API evolution GPUExtent3D.depth -> depthOrArrayLayers (Step 2) Still leave deprecated `depth` functional as there are some references in other clients. Using `depth` and `depthOrArrayLayers` at the same time is invalid. Add DeprecatedAPITests. Bug: chromium:1176969 Change-Id: Ia06645e4f3c17588323dd36b11f9f3988b2e3aba Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/44640 Commit-Queue: Shrek Shao <shrekshao@google.com> Reviewed-by: Austin Eng <enga@chromium.org>

diff --git a/src/dawn_native/CommandBuffer.cpp b/src/dawn_native/CommandBuffer.cpp
index bde0c63..9fd1a99 100644
--- a/src/dawn_native/CommandBuffer.cpp
+++ b/src/dawn_native/CommandBuffer.cpp

@@ -77,7 +77,8 @@
                                                    const Extent3D& copySize) {
         switch (copy.texture->GetDimension()) {
             case wgpu::TextureDimension::e2D:
-                return {copy.aspect, {copy.origin.z, copySize.depth}, {copy.mipLevel, 1}};
+                return {
+                    copy.aspect, {copy.origin.z, copySize.depthOrArrayLayers}, {copy.mipLevel, 1}};
             default:
                 UNREACHABLE();
         }
@@ -181,7 +182,7 @@
         }
 
         const uint64_t overwrittenRangeSize =
-            copyTextureDataSizePerRow * heightInBlocks * copy->copySize.depth;
+            copyTextureDataSizePerRow * heightInBlocks * copy->copySize.depthOrArrayLayers;
         if (copy->destination.buffer->GetSize() > overwrittenRangeSize) {
             return false;
         }

diff --git a/src/dawn_native/CommandEncoder.cpp b/src/dawn_native/CommandEncoder.cpp
index 9c85531f..00e6255 100644
--- a/src/dawn_native/CommandEncoder.cpp
+++ b/src/dawn_native/CommandEncoder.cpp

@@ -626,11 +626,14 @@
                                              const ImageCopyTexture* destination,
                                              const Extent3D* copySize) {
         mEncodingContext.TryEncode(this, [&](CommandAllocator* allocator) -> MaybeError {
+            Extent3D fixedCopySize = *copySize;
+            DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(GetDevice(), &fixedCopySize));
+
             if (GetDevice()->IsValidationEnabled()) {
                 DAWN_TRY(ValidateImageCopyBuffer(GetDevice(), *source));
                 DAWN_TRY(ValidateCanUseAs(source->buffer, wgpu::BufferUsage::CopySrc));
 
-                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *destination, *copySize));
+                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *destination, fixedCopySize));
                 DAWN_TRY(ValidateCanUseAs(destination->texture, wgpu::TextureUsage::CopyDst));
                 DAWN_TRY(ValidateTextureSampleCountInBufferCopyCommands(destination->texture));
 
@@ -639,25 +642,26 @@
                 // because in the latter we divide copyExtent.width by blockWidth and
                 // copyExtent.height by blockHeight while the divisibility conditions are
                 // checked in validating texture copy range.
-                DAWN_TRY(ValidateTextureCopyRange(*destination, *copySize));
+                DAWN_TRY(ValidateTextureCopyRange(*destination, fixedCopySize));
             }
             const TexelBlockInfo& blockInfo =
                 destination->texture->GetFormat().GetAspectInfo(destination->aspect).block;
             TextureDataLayout srcLayout = FixUpDeprecatedTextureDataLayoutOptions(
-                GetDevice(), source->layout, blockInfo, *copySize);
+                GetDevice(), source->layout, blockInfo, fixedCopySize);
             if (GetDevice()->IsValidationEnabled()) {
                 DAWN_TRY(ValidateLinearTextureCopyOffset(srcLayout, blockInfo));
                 DAWN_TRY(ValidateLinearTextureData(srcLayout, source->buffer->GetSize(), blockInfo,
-                                                   *copySize));
+                                                   fixedCopySize));
 
                 mTopLevelBuffers.insert(source->buffer);
                 mTopLevelTextures.insert(destination->texture);
             }
 
-            ApplyDefaultTextureDataLayoutOptions(&srcLayout, blockInfo, *copySize);
+            ApplyDefaultTextureDataLayoutOptions(&srcLayout, blockInfo, fixedCopySize);
 
             // Skip noop copies.
-            if (copySize->width != 0 && copySize->height != 0 && copySize->depth != 0) {
+            if (fixedCopySize.width != 0 && fixedCopySize.height != 0 &&
+                fixedCopySize.depthOrArrayLayers != 0) {
                 // Record the copy command.
                 CopyBufferToTextureCmd* copy =
                     allocator->Allocate<CopyBufferToTextureCmd>(Command::CopyBufferToTexture);
@@ -670,7 +674,7 @@
                 copy->destination.mipLevel = destination->mipLevel;
                 copy->destination.aspect =
                     ConvertAspect(destination->texture->GetFormat(), destination->aspect);
-                copy->copySize = *copySize;
+                copy->copySize = fixedCopySize;
             }
 
             return {};
@@ -681,8 +685,11 @@
                                              const ImageCopyBuffer* destination,
                                              const Extent3D* copySize) {
         mEncodingContext.TryEncode(this, [&](CommandAllocator* allocator) -> MaybeError {
+            Extent3D fixedCopySize = *copySize;
+            DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(GetDevice(), &fixedCopySize));
+
             if (GetDevice()->IsValidationEnabled()) {
-                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *source, *copySize));
+                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *source, fixedCopySize));
                 DAWN_TRY(ValidateCanUseAs(source->texture, wgpu::TextureUsage::CopySrc));
                 DAWN_TRY(ValidateTextureSampleCountInBufferCopyCommands(source->texture));
                 DAWN_TRY(ValidateTextureDepthStencilToBufferCopyRestrictions(*source));
@@ -694,25 +701,26 @@
                 // because in the latter we divide copyExtent.width by blockWidth and
                 // copyExtent.height by blockHeight while the divisibility conditions are
                 // checked in validating texture copy range.
-                DAWN_TRY(ValidateTextureCopyRange(*source, *copySize));
+                DAWN_TRY(ValidateTextureCopyRange(*source, fixedCopySize));
             }
             const TexelBlockInfo& blockInfo =
                 source->texture->GetFormat().GetAspectInfo(source->aspect).block;
             TextureDataLayout dstLayout = FixUpDeprecatedTextureDataLayoutOptions(
-                GetDevice(), destination->layout, blockInfo, *copySize);
+                GetDevice(), destination->layout, blockInfo, fixedCopySize);
             if (GetDevice()->IsValidationEnabled()) {
                 DAWN_TRY(ValidateLinearTextureCopyOffset(dstLayout, blockInfo));
                 DAWN_TRY(ValidateLinearTextureData(dstLayout, destination->buffer->GetSize(),
-                                                   blockInfo, *copySize));
+                                                   blockInfo, fixedCopySize));
 
                 mTopLevelTextures.insert(source->texture);
                 mTopLevelBuffers.insert(destination->buffer);
             }
 
-            ApplyDefaultTextureDataLayoutOptions(&dstLayout, blockInfo, *copySize);
+            ApplyDefaultTextureDataLayoutOptions(&dstLayout, blockInfo, fixedCopySize);
 
             // Skip noop copies.
-            if (copySize->width != 0 && copySize->height != 0 && copySize->depth != 0) {
+            if (fixedCopySize.width != 0 && fixedCopySize.height != 0 &&
+                fixedCopySize.depthOrArrayLayers != 0) {
                 // Record the copy command.
                 CopyTextureToBufferCmd* copy =
                     allocator->Allocate<CopyTextureToBufferCmd>(Command::CopyTextureToBuffer);
@@ -724,7 +732,7 @@
                 copy->destination.offset = dstLayout.offset;
                 copy->destination.bytesPerRow = dstLayout.bytesPerRow;
                 copy->destination.rowsPerImage = dstLayout.rowsPerImage;
-                copy->copySize = *copySize;
+                copy->copySize = fixedCopySize;
             }
 
             return {};
@@ -735,18 +743,20 @@
                                               const ImageCopyTexture* destination,
                                               const Extent3D* copySize) {
         mEncodingContext.TryEncode(this, [&](CommandAllocator* allocator) -> MaybeError {
+            Extent3D fixedCopySize = *copySize;
+            DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(GetDevice(), &fixedCopySize));
             if (GetDevice()->IsValidationEnabled()) {
                 DAWN_TRY(GetDevice()->ValidateObject(source->texture));
                 DAWN_TRY(GetDevice()->ValidateObject(destination->texture));
 
-                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *source, *copySize));
-                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *destination, *copySize));
+                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *source, fixedCopySize));
+                DAWN_TRY(ValidateImageCopyTexture(GetDevice(), *destination, fixedCopySize));
 
                 DAWN_TRY(
-                    ValidateTextureToTextureCopyRestrictions(*source, *destination, *copySize));
+                    ValidateTextureToTextureCopyRestrictions(*source, *destination, fixedCopySize));
 
-                DAWN_TRY(ValidateTextureCopyRange(*source, *copySize));
-                DAWN_TRY(ValidateTextureCopyRange(*destination, *copySize));
+                DAWN_TRY(ValidateTextureCopyRange(*source, fixedCopySize));
+                DAWN_TRY(ValidateTextureCopyRange(*destination, fixedCopySize));
 
                 DAWN_TRY(ValidateCanUseAs(source->texture, wgpu::TextureUsage::CopySrc));
                 DAWN_TRY(ValidateCanUseAs(destination->texture, wgpu::TextureUsage::CopyDst));
@@ -756,7 +766,8 @@
             }
 
             // Skip noop copies.
-            if (copySize->width != 0 && copySize->height != 0 && copySize->depth != 0) {
+            if (fixedCopySize.width != 0 && fixedCopySize.height != 0 &&
+                fixedCopySize.depthOrArrayLayers != 0) {
                 CopyTextureToTextureCmd* copy =
                     allocator->Allocate<CopyTextureToTextureCmd>(Command::CopyTextureToTexture);
                 copy->source.texture = source->texture;
@@ -768,7 +779,7 @@
                 copy->destination.mipLevel = destination->mipLevel;
                 copy->destination.aspect =
                     ConvertAspect(destination->texture->GetFormat(), destination->aspect);
-                copy->copySize = *copySize;
+                copy->copySize = fixedCopySize;
             }
 
             return {};

diff --git a/src/dawn_native/CommandValidation.cpp b/src/dawn_native/CommandValidation.cpp
index e8e2b2e..e4ad9e2 100644
--- a/src/dawn_native/CommandValidation.cpp
+++ b/src/dawn_native/CommandValidation.cpp

@@ -103,7 +103,7 @@
         uint32_t heightInBlocks = copySize.height / blockInfo.height;
         uint64_t bytesInLastRow = Safe32x32(widthInBlocks, blockInfo.byteSize);
 
-        if (copySize.depth == 0) {
+        if (copySize.depthOrArrayLayers == 0) {
             return 0;
         }
 
@@ -122,14 +122,14 @@
         //
         // This means that if the computation of depth * bytesPerImage doesn't overflow, none of the
         // computations for requiredBytesInCopy will. (and it's not a very pessimizing check)
-        ASSERT(copySize.depth <= 1 || (bytesPerRow != wgpu::kCopyStrideUndefined &&
-                                       rowsPerImage != wgpu::kCopyStrideUndefined));
+        ASSERT(copySize.depthOrArrayLayers <= 1 || (bytesPerRow != wgpu::kCopyStrideUndefined &&
+                                                    rowsPerImage != wgpu::kCopyStrideUndefined));
         uint64_t bytesPerImage = Safe32x32(bytesPerRow, rowsPerImage);
-        if (bytesPerImage > std::numeric_limits<uint64_t>::max() / copySize.depth) {
+        if (bytesPerImage > std::numeric_limits<uint64_t>::max() / copySize.depthOrArrayLayers) {
             return DAWN_VALIDATION_ERROR("requiredBytesInCopy is too large.");
         }
 
-        uint64_t requiredBytesInCopy = bytesPerImage * (copySize.depth - 1);
+        uint64_t requiredBytesInCopy = bytesPerImage * (copySize.depthOrArrayLayers - 1);
         if (heightInBlocks > 0) {
             ASSERT(heightInBlocks <= 1 || bytesPerRow != wgpu::kCopyStrideUndefined);
             uint64_t bytesInLastImage = Safe32x32(bytesPerRow, heightInBlocks - 1) + bytesInLastRow;
@@ -159,14 +159,14 @@
         TextureDataLayout layout = originalLayout;
 
         if (copyExtent.height != 0 && layout.rowsPerImage == 0) {
-            if (copyExtent.depth > 1) {
+            if (copyExtent.depthOrArrayLayers > 1) {
                 device->EmitDeprecationWarning(
                     "rowsPerImage soon must be non-zero if copy depth > 1 (it will no longer "
                     "default to the copy height).");
                 ASSERT(copyExtent.height % blockInfo.height == 0);
                 uint32_t heightInBlocks = copyExtent.height / blockInfo.height;
                 layout.rowsPerImage = heightInBlocks;
-            } else if (copyExtent.depth == 1) {
+            } else if (copyExtent.depthOrArrayLayers == 1) {
                 device->EmitDeprecationWarning(
                     "rowsPerImage soon must be non-zero or unspecified if copy depth == 1 (it will "
                     "no longer default to the copy height).");
@@ -179,7 +179,7 @@
         ASSERT(copyExtent.width % blockInfo.width == 0);
         uint32_t widthInBlocks = copyExtent.width / blockInfo.width;
         uint32_t bytesInLastRow = widthInBlocks * blockInfo.byteSize;
-        if (copyExtent.height == 1 && copyExtent.depth == 1 &&
+        if (copyExtent.height == 1 && copyExtent.depthOrArrayLayers == 1 &&
             bytesInLastRow > layout.bytesPerRow) {
             device->EmitDeprecationWarning(
                 "Soon, even if copy height == 1, bytesPerRow must be >= the byte size of each row "
@@ -203,11 +203,11 @@
             uint32_t widthInBlocks = copyExtent.width / blockInfo.width;
             uint32_t bytesInLastRow = widthInBlocks * blockInfo.byteSize;
 
-            ASSERT(heightInBlocks <= 1 && copyExtent.depth <= 1);
+            ASSERT(heightInBlocks <= 1 && copyExtent.depthOrArrayLayers <= 1);
             layout->bytesPerRow = Align(bytesInLastRow, kTextureBytesPerRowAlignment);
         }
         if (layout->rowsPerImage == wgpu::kCopyStrideUndefined) {
-            ASSERT(copyExtent.depth <= 1);
+            ASSERT(copyExtent.depthOrArrayLayers <= 1);
             layout->rowsPerImage = heightInBlocks;
         }
     }
@@ -219,8 +219,9 @@
         ASSERT(copyExtent.height % blockInfo.height == 0);
         uint32_t heightInBlocks = copyExtent.height / blockInfo.height;
 
-        if (copyExtent.depth > 1 && (layout.bytesPerRow == wgpu::kCopyStrideUndefined ||
-                                     layout.rowsPerImage == wgpu::kCopyStrideUndefined)) {
+        if (copyExtent.depthOrArrayLayers > 1 &&
+            (layout.bytesPerRow == wgpu::kCopyStrideUndefined ||
+             layout.rowsPerImage == wgpu::kCopyStrideUndefined)) {
             return DAWN_VALIDATION_ERROR(
                 "If copy depth > 1, bytesPerRow and rowsPerImage must be specified.");
         }
@@ -316,7 +317,7 @@
         // For 2D textures, include the array layer as depth so it can be checked with other
         // dimensions.
         ASSERT(texture->GetDimension() == wgpu::TextureDimension::e2D);
-        mipSize.depth = texture->GetArrayLayers();
+        mipSize.depthOrArrayLayers = texture->GetArrayLayers();
 
         // All texture dimensions are in uint32_t so by doing checks in uint64_t we avoid
         // overflows.
@@ -324,8 +325,9 @@
                 static_cast<uint64_t>(mipSize.width) ||
             static_cast<uint64_t>(textureCopy.origin.y) + static_cast<uint64_t>(copySize.height) >
                 static_cast<uint64_t>(mipSize.height) ||
-            static_cast<uint64_t>(textureCopy.origin.z) + static_cast<uint64_t>(copySize.depth) >
-                static_cast<uint64_t>(mipSize.depth)) {
+            static_cast<uint64_t>(textureCopy.origin.z) +
+                    static_cast<uint64_t>(copySize.depthOrArrayLayers) >
+                static_cast<uint64_t>(mipSize.depthOrArrayLayers)) {
             return DAWN_VALIDATION_ERROR("Touching outside of the texture");
         }
 
@@ -422,7 +424,7 @@
         if (src.texture == dst.texture && src.mipLevel == dst.mipLevel) {
             ASSERT(src.texture->GetDimension() == wgpu::TextureDimension::e2D &&
                    dst.texture->GetDimension() == wgpu::TextureDimension::e2D);
-            if (IsRangeOverlapped(src.origin.z, dst.origin.z, copySize.depth)) {
+            if (IsRangeOverlapped(src.origin.z, dst.origin.z, copySize.depthOrArrayLayers)) {
                 return DAWN_VALIDATION_ERROR(
                     "Copy subresources cannot be overlapped when copying within the same "
                     "texture.");

diff --git a/src/dawn_native/Device.cpp b/src/dawn_native/Device.cpp
index 6e082dd..cf65c42 100644
--- a/src/dawn_native/Device.cpp
+++ b/src/dawn_native/Device.cpp

@@ -1247,10 +1247,12 @@
     ResultOrError<Ref<TextureBase>> DeviceBase::CreateTextureInternal(
         const TextureDescriptor* descriptor) {
         DAWN_TRY(ValidateIsAlive());
+        TextureDescriptor fixedDescriptor = *descriptor;
+        DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(this, &(fixedDescriptor.size)));
         if (IsValidationEnabled()) {
-            DAWN_TRY(ValidateTextureDescriptor(this, descriptor));
+            DAWN_TRY(ValidateTextureDescriptor(this, &fixedDescriptor));
         }
-        return CreateTextureImpl(descriptor);
+        return CreateTextureImpl(&fixedDescriptor);
     }
 
     MaybeError DeviceBase::CreateTextureViewInternal(TextureViewBase** result,

diff --git a/src/dawn_native/Queue.cpp b/src/dawn_native/Queue.cpp
index e3e9141..d5fa05a 100644
--- a/src/dawn_native/Queue.cpp
+++ b/src/dawn_native/Queue.cpp

@@ -115,9 +115,9 @@
             uint64_t imageAdditionalStride =
                 dataLayout.bytesPerRow * (dataRowsPerImage - alignedRowsPerImage);
 
-            CopyTextureData(dstPointer, srcPointer, writeSizePixel.depth, alignedRowsPerImage,
-                            imageAdditionalStride, alignedBytesPerRow, optimallyAlignedBytesPerRow,
-                            dataLayout.bytesPerRow);
+            CopyTextureData(dstPointer, srcPointer, writeSizePixel.depthOrArrayLayers,
+                            alignedRowsPerImage, imageAdditionalStride, alignedBytesPerRow,
+                            optimallyAlignedBytesPerRow, dataLayout.bytesPerRow);
 
             return uploadHandle;
         }
@@ -305,17 +305,21 @@
                                                size_t dataSize,
                                                const TextureDataLayout* dataLayout,
                                                const Extent3D* writeSize) {
-        DAWN_TRY(ValidateWriteTexture(destination, dataSize, dataLayout, writeSize));
+        Extent3D fixedWriteSize = *writeSize;
+        DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(GetDevice(), &fixedWriteSize));
 
-        if (writeSize->width == 0 || writeSize->height == 0 || writeSize->depth == 0) {
+        DAWN_TRY(ValidateWriteTexture(destination, dataSize, dataLayout, &fixedWriteSize));
+
+        if (fixedWriteSize.width == 0 || fixedWriteSize.height == 0 ||
+            fixedWriteSize.depthOrArrayLayers == 0) {
             return {};
         }
 
         const TexelBlockInfo& blockInfo =
             destination->texture->GetFormat().GetAspectInfo(destination->aspect).block;
         TextureDataLayout layout = *dataLayout;
-        ApplyDefaultTextureDataLayoutOptions(&layout, blockInfo, *writeSize);
-        return WriteTextureImpl(*destination, data, layout, *writeSize);
+        ApplyDefaultTextureDataLayoutOptions(&layout, blockInfo, fixedWriteSize);
+        return WriteTextureImpl(*destination, data, layout, fixedWriteSize);
     }
 
     MaybeError QueueBase::WriteTextureImpl(const ImageCopyTexture& destination,
@@ -375,12 +379,14 @@
         const ImageCopyTexture* destination,
         const Extent3D* copySize,
         const CopyTextureForBrowserOptions* options) {
+        Extent3D fixedCopySize = *copySize;
+        DAWN_TRY(FixUpDeprecatedGPUExtent3DDepth(GetDevice(), &fixedCopySize));
         if (GetDevice()->IsValidationEnabled()) {
-            DAWN_TRY(
-                ValidateCopyTextureForBrowser(GetDevice(), source, destination, copySize, options));
+            DAWN_TRY(ValidateCopyTextureForBrowser(GetDevice(), source, destination, &fixedCopySize,
+                                                   options));
         }
 
-        return DoCopyTextureForBrowser(GetDevice(), source, destination, copySize, options);
+        return DoCopyTextureForBrowser(GetDevice(), source, destination, &fixedCopySize, options);
     }
 
     MaybeError QueueBase::ValidateSubmit(uint32_t commandCount,

diff --git a/src/dawn_native/SwapChain.cpp b/src/dawn_native/SwapChain.cpp
index 84f5a56..96ef193 100644
--- a/src/dawn_native/SwapChain.cpp
+++ b/src/dawn_native/SwapChain.cpp

@@ -180,7 +180,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = mWidth;
         descriptor.size.height = mHeight;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = mFormat;
         descriptor.mipLevelCount = 1;

diff --git a/src/dawn_native/Texture.cpp b/src/dawn_native/Texture.cpp
index 21ae34b..8679e95 100644
--- a/src/dawn_native/Texture.cpp
+++ b/src/dawn_native/Texture.cpp

@@ -118,7 +118,7 @@
                 // Multisampled 2D array texture is not supported because on Metal it requires the
                 // version of macOS be greater than 10.14.
                 if (descriptor->dimension != wgpu::TextureDimension::e2D ||
-                    descriptor->size.depth > 1) {
+                    descriptor->size.depthOrArrayLayers > 1) {
                     return DAWN_VALIDATION_ERROR("Multisampled texture must be 2D with depth=1");
                 }
 
@@ -164,7 +164,7 @@
 
         MaybeError ValidateTextureSize(const TextureDescriptor* descriptor, const Format* format) {
             ASSERT(descriptor->size.width != 0 && descriptor->size.height != 0 &&
-                   descriptor->size.depth != 0);
+                   descriptor->size.depthOrArrayLayers != 0);
 
             Extent3D maxExtent;
             switch (descriptor->dimension) {
@@ -182,7 +182,7 @@
             }
             if (descriptor->size.width > maxExtent.width ||
                 descriptor->size.height > maxExtent.height ||
-                descriptor->size.depth > maxExtent.depth) {
+                descriptor->size.depthOrArrayLayers > maxExtent.depthOrArrayLayers) {
                 return DAWN_VALIDATION_ERROR("Texture dimension (width, height or depth) exceeded");
             }
 
@@ -191,7 +191,8 @@
                 maxMippedDimension = std::max(maxMippedDimension, descriptor->size.height);
             }
             if (descriptor->dimension == wgpu::TextureDimension::e3D) {
-                maxMippedDimension = std::max(maxMippedDimension, descriptor->size.depth);
+                maxMippedDimension =
+                    std::max(maxMippedDimension, descriptor->size.depthOrArrayLayers);
             }
             if (Log2(maxMippedDimension) + 1 < descriptor->mipLevelCount) {
                 return DAWN_VALIDATION_ERROR("Texture has too many mip levels");
@@ -243,6 +244,26 @@
 
     }  // anonymous namespace
 
+    MaybeError FixUpDeprecatedGPUExtent3DDepth(DeviceBase* device, Extent3D* extent) {
+        if (extent->depth != 1) {
+            // deprecated depth is assigned
+            if (extent->depthOrArrayLayers != 1) {
+                // both deprecated and updated API is used
+                return DAWN_VALIDATION_ERROR(
+                    "Deprecated GPUExtent3D.depth and updated GPUExtent3D.depthOrArrayLengths are "
+                    "both assigned.");
+            }
+
+            extent->depthOrArrayLayers = extent->depth;
+
+            device->EmitDeprecationWarning(
+                "GPUExtent3D.depth is deprecated. Please use GPUExtent3D.depthOrArrayLayers "
+                "instead.");
+        }
+
+        return {};
+    }
+
     MaybeError ValidateTextureDescriptor(const DeviceBase* device,
                                          const TextureDescriptor* descriptor) {
         if (descriptor == nullptr) {
@@ -264,7 +285,7 @@
 
         // TODO(jiawei.shao@intel.com): check stuff based on the dimension
         if (descriptor->size.width == 0 || descriptor->size.height == 0 ||
-            descriptor->size.depth == 0 || descriptor->mipLevelCount == 0) {
+            descriptor->size.depthOrArrayLayers == 0 || descriptor->mipLevelCount == 0) {
             return DAWN_VALIDATION_ERROR("Cannot create an empty texture");
         }
 
@@ -399,7 +420,8 @@
           mSampleCount(descriptor->sampleCount),
           mUsage(descriptor->usage),
           mState(state) {
-        uint32_t subresourceCount = mMipLevelCount * mSize.depth * GetAspectCount(mFormat.aspects);
+        uint32_t subresourceCount =
+            mMipLevelCount * mSize.depthOrArrayLayers * GetAspectCount(mFormat.aspects);
         mIsSubresourceContentInitializedAtIndex = std::vector<bool>(subresourceCount, false);
 
         // Add readonly storage usage if the texture has a storage usage. The validation rules in
@@ -446,7 +468,7 @@
     uint32_t TextureBase::GetDepth() const {
         ASSERT(!IsError());
         ASSERT(mDimension == wgpu::TextureDimension::e3D);
-        return mSize.depth;
+        return mSize.depthOrArrayLayers;
     }
     uint32_t TextureBase::GetArrayLayers() const {
         ASSERT(!IsError());
@@ -455,7 +477,7 @@
         if (mDimension == wgpu::TextureDimension::e3D) {
             return 1;
         }
-        return mSize.depth;
+        return mSize.depthOrArrayLayers;
     }
     uint32_t TextureBase::GetNumMipLevels() const {
         ASSERT(!IsError());
@@ -554,7 +576,7 @@
             return extent;
         }
 
-        extent.depth = std::max(mSize.depth >> level, 1u);
+        extent.depthOrArrayLayers = std::max(mSize.depthOrArrayLayers >> level, 1u);
         return extent;
     }
 
@@ -584,7 +606,7 @@
         uint32_t clampedCopyExtentHeight = (origin.y + extent.height > virtualSizeAtLevel.height)
                                                ? (virtualSizeAtLevel.height - origin.y)
                                                : extent.height;
-        return {clampedCopyExtentWidth, clampedCopyExtentHeight, extent.depth};
+        return {clampedCopyExtentWidth, clampedCopyExtentHeight, extent.depthOrArrayLayers};
     }
 
     TextureViewBase* TextureBase::CreateView(const TextureViewDescriptor* descriptor) {

diff --git a/src/dawn_native/Texture.h b/src/dawn_native/Texture.h
index 52cfa06..108e11b 100644
--- a/src/dawn_native/Texture.h
+++ b/src/dawn_native/Texture.h

@@ -38,6 +38,8 @@
 
     bool IsValidSampleCount(uint32_t sampleCount);
 
+    MaybeError FixUpDeprecatedGPUExtent3DDepth(DeviceBase* device, Extent3D* extent);
+
     static constexpr wgpu::TextureUsage kReadOnlyTextureUsages =
         wgpu::TextureUsage::CopySrc | wgpu::TextureUsage::Sampled | kReadOnlyStorageTexture;
 

diff --git a/src/dawn_native/d3d12/CommandBufferD3D12.cpp b/src/dawn_native/d3d12/CommandBufferD3D12.cpp
index 0d8d83a..020d0c6 100644
--- a/src/dawn_native/d3d12/CommandBufferD3D12.cpp
+++ b/src/dawn_native/d3d12/CommandBufferD3D12.cpp

@@ -81,8 +81,8 @@
                    copySize.width == srcSize.width &&      //
                    copySize.height == dstSize.height &&    //
                    copySize.height == srcSize.height &&    //
-                   copySize.depth == dstSize.depth &&      //
-                   copySize.depth == srcSize.depth;
+                   copySize.depthOrArrayLayers == dstSize.depthOrArrayLayers &&  //
+                   copySize.depthOrArrayLayers == srcSize.depthOrArrayLayers;
         }
 
         void RecordCopyTextureToBufferFromTextureCopySplit(ID3D12GraphicsCommandList* commandList,
@@ -143,7 +143,7 @@
             // that uses copySplits.copies2D[1].
             std::array<uint64_t, TextureCopySplits::kMaxTextureCopySplits>
                 bufferOffsetsForNextSlice = {{0u, 0u}};
-            for (uint32_t copySlice = 0; copySlice < copySize.depth; ++copySlice) {
+            for (uint32_t copySlice = 0; copySlice < copySize.depthOrArrayLayers; ++copySlice) {
                 const uint32_t splitIndex = copySlice % copySplits.copies2D.size();
 
                 const Texture2DCopySplit& copySplitPerLayerBase = copySplits.copies2D[splitIndex];
@@ -882,7 +882,7 @@
                         // it is not allowed to copy with overlapped subresources, but we still
                         // add the ASSERT here as a reminder for this possible misuse.
                         ASSERT(!IsRangeOverlapped(copy->source.origin.z, copy->destination.origin.z,
-                                                  copy->copySize.depth));
+                                                  copy->copySize.depthOrArrayLayers));
                     }
                     source->TrackUsageAndTransitionNow(commandContext, wgpu::TextureUsage::CopySrc,
                                                        srcRange);
@@ -908,7 +908,8 @@
                             copy->copySize.width, copy->copySize.height, 1u};
 
                         for (Aspect aspect : IterateEnumMask(srcRange.aspects)) {
-                            for (uint32_t slice = 0; slice < copy->copySize.depth; ++slice) {
+                            for (uint32_t slice = 0; slice < copy->copySize.depthOrArrayLayers;
+                                 ++slice) {
                                 D3D12_TEXTURE_COPY_LOCATION srcLocation =
                                     ComputeTextureCopyLocationForTexture(
                                         source, copy->source.mipLevel,

diff --git a/src/dawn_native/d3d12/D3D12Backend.cpp b/src/dawn_native/d3d12/D3D12Backend.cpp
index 22d13dd..3975698 100644
--- a/src/dawn_native/d3d12/D3D12Backend.cpp
+++ b/src/dawn_native/d3d12/D3D12Backend.cpp

@@ -79,7 +79,7 @@
         TextureDescriptor textureDescriptor = {};
         textureDescriptor.usage = static_cast<wgpu::TextureUsage>(descriptor->usage);
         textureDescriptor.dimension = static_cast<wgpu::TextureDimension>(mDimension);
-        textureDescriptor.size = {mSize.width, mSize.height, mSize.depth};
+        textureDescriptor.size = {mSize.width, mSize.height, mSize.depthOrArrayLayers};
         textureDescriptor.format = static_cast<wgpu::TextureFormat>(mFormat);
         textureDescriptor.mipLevelCount = mMipLevelCount;
         textureDescriptor.sampleCount = mSampleCount;

diff --git a/src/dawn_native/d3d12/TextureCopySplitter.cpp b/src/dawn_native/d3d12/TextureCopySplitter.cpp
index e8cfae2..d75257e 100644
--- a/src/dawn_native/d3d12/TextureCopySplitter.cpp
+++ b/src/dawn_native/d3d12/TextureCopySplitter.cpp

@@ -131,7 +131,7 @@
             copy.copies[0].bufferOffset = texelOffset;
             copy.copies[0].bufferSize.width = copySize.width + texelOffset.x;
             copy.copies[0].bufferSize.height = rowsPerImageInTexels + texelOffset.y;
-            copy.copies[0].bufferSize.depth = copySize.depth;
+            copy.copies[0].bufferSize.depthOrArrayLayers = copySize.depthOrArrayLayers;
 
             return copy;
         }
@@ -178,12 +178,12 @@
         uint32_t texelsPerRow = bytesPerRow / blockInfo.byteSize * blockInfo.width;
         copy.copies[0].copySize.width = texelsPerRow - texelOffset.x;
         copy.copies[0].copySize.height = copySize.height;
-        copy.copies[0].copySize.depth = copySize.depth;
+        copy.copies[0].copySize.depthOrArrayLayers = copySize.depthOrArrayLayers;
 
         copy.copies[0].bufferOffset = texelOffset;
         copy.copies[0].bufferSize.width = texelsPerRow;
         copy.copies[0].bufferSize.height = rowsPerImageInTexels + texelOffset.y;
-        copy.copies[0].bufferSize.depth = copySize.depth;
+        copy.copies[0].bufferSize.depthOrArrayLayers = copySize.depthOrArrayLayers;
 
         copy.copies[1].textureOffset.x = origin.x + copy.copies[0].copySize.width;
         copy.copies[1].textureOffset.y = origin.y;
@@ -192,14 +192,14 @@
         ASSERT(copySize.width > copy.copies[0].copySize.width);
         copy.copies[1].copySize.width = copySize.width - copy.copies[0].copySize.width;
         copy.copies[1].copySize.height = copySize.height;
-        copy.copies[1].copySize.depth = copySize.depth;
+        copy.copies[1].copySize.depthOrArrayLayers = copySize.depthOrArrayLayers;
 
         copy.copies[1].bufferOffset.x = 0;
         copy.copies[1].bufferOffset.y = texelOffset.y + blockInfo.height;
         copy.copies[1].bufferOffset.z = 0;
         copy.copies[1].bufferSize.width = copy.copies[1].copySize.width;
         copy.copies[1].bufferSize.height = rowsPerImageInTexels + texelOffset.y + blockInfo.height;
-        copy.copies[1].bufferSize.depth = copySize.depth;
+        copy.copies[1].bufferSize.depthOrArrayLayers = copySize.depthOrArrayLayers;
 
         return copy;
     }
@@ -233,7 +233,7 @@
 
         // When the copy only refers one texture 2D array layer copies.copies2D[1] will never be
         // used so we can safely early return here.
-        if (copySize.depth == 1) {
+        if (copySize.depthOrArrayLayers == 1) {
             return copies;
         }
 

diff --git a/src/dawn_native/d3d12/TextureD3D12.cpp b/src/dawn_native/d3d12/TextureD3D12.cpp
index 6b90c0a..ba70e43 100644
--- a/src/dawn_native/d3d12/TextureD3D12.cpp
+++ b/src/dawn_native/d3d12/TextureD3D12.cpp

@@ -344,7 +344,7 @@
             return DAWN_VALIDATION_ERROR("Mip level count must be 1");
         }
 
-        if (descriptor->size.depth != 1) {
+        if (descriptor->size.depthOrArrayLayers != 1) {
             return DAWN_VALIDATION_ERROR("Depth must be 1");
         }
 
@@ -360,7 +360,7 @@
         const D3D12_RESOURCE_DESC d3dDescriptor = d3d12Resource->GetDesc();
         if ((dawnDescriptor->size.width != d3dDescriptor.Width) ||
             (dawnDescriptor->size.height != d3dDescriptor.Height) ||
-            (dawnDescriptor->size.depth != 1)) {
+            (dawnDescriptor->size.depthOrArrayLayers != 1)) {
             return DAWN_VALIDATION_ERROR("D3D12 texture size doesn't match descriptor");
         }
 
@@ -485,7 +485,7 @@
         const Extent3D& size = GetSize();
         resourceDescriptor.Width = size.width;
         resourceDescriptor.Height = size.height;
-        resourceDescriptor.DepthOrArraySize = size.depth;
+        resourceDescriptor.DepthOrArraySize = size.depthOrArrayLayers;
 
         // This will need to be much more nuanced when WebGPU has
         // texture view compatibility rules.

diff --git a/src/dawn_native/d3d12/UtilsD3D12.cpp b/src/dawn_native/d3d12/UtilsD3D12.cpp
index 7f6d08f..7f3df52 100644
--- a/src/dawn_native/d3d12/UtilsD3D12.cpp
+++ b/src/dawn_native/d3d12/UtilsD3D12.cpp

@@ -93,7 +93,7 @@
             texture->GetD3D12CopyableSubresourceFormat(aspect);
         bufferLocation.PlacedFootprint.Footprint.Width = bufferSize.width;
         bufferLocation.PlacedFootprint.Footprint.Height = bufferSize.height;
-        bufferLocation.PlacedFootprint.Footprint.Depth = bufferSize.depth;
+        bufferLocation.PlacedFootprint.Footprint.Depth = bufferSize.depthOrArrayLayers;
         bufferLocation.PlacedFootprint.Footprint.RowPitch = rowPitch;
         return bufferLocation;
     }
@@ -105,7 +105,7 @@
         sourceRegion.front = offset.z;
         sourceRegion.right = offset.x + copySize.width;
         sourceRegion.bottom = offset.y + copySize.height;
-        sourceRegion.back = offset.z + copySize.depth;
+        sourceRegion.back = offset.z + copySize.depthOrArrayLayers;
         return sourceRegion;
     }
 
@@ -201,7 +201,7 @@
         std::array<uint64_t, TextureCopySplits::kMaxTextureCopySplits> bufferOffsetsForNextSlice = {
             {0u, 0u}};
 
-        for (uint32_t copySlice = 0; copySlice < copySize.depth; ++copySlice) {
+        for (uint32_t copySlice = 0; copySlice < copySize.depthOrArrayLayers; ++copySlice) {
             const uint32_t splitIndex = copySlice % copySplits.copies2D.size();
 
             const Texture2DCopySplit& copySplitPerLayerBase = copySplits.copies2D[splitIndex];

diff --git a/src/dawn_native/metal/CommandBufferMTL.mm b/src/dawn_native/metal/CommandBufferMTL.mm
index 58b85fb..a8794d7 100644
--- a/src/dawn_native/metal/CommandBufferMTL.mm
+++ b/src/dawn_native/metal/CommandBufferMTL.mm

@@ -638,7 +638,7 @@
                         const TextureBufferCopySplit::CopyInfo& copyInfo = splitCopies.copies[i];
 
                         const uint32_t copyBaseLayer = copyInfo.textureOrigin.z;
-                        const uint32_t copyLayerCount = copyInfo.copyExtent.depth;
+                        const uint32_t copyLayerCount = copyInfo.copyExtent.depthOrArrayLayers;
                         const MTLOrigin textureOrigin =
                             MTLOriginMake(copyInfo.textureOrigin.x, copyInfo.textureOrigin.y, 0);
                         const MTLSize copyExtent =
@@ -688,7 +688,7 @@
                         const TextureBufferCopySplit::CopyInfo& copyInfo = splitCopies.copies[i];
 
                         const uint32_t copyBaseLayer = copyInfo.textureOrigin.z;
-                        const uint32_t copyLayerCount = copyInfo.copyExtent.depth;
+                        const uint32_t copyLayerCount = copyInfo.copyExtent.depthOrArrayLayers;
                         const MTLOrigin textureOrigin =
                             MTLOriginMake(copyInfo.textureOrigin.x, copyInfo.textureOrigin.y, 0);
                         const MTLSize copyExtent =
@@ -738,7 +738,7 @@
                     const MTLOrigin destinationOriginNoLayer =
                         MTLOriginMake(copy->destination.origin.x, copy->destination.origin.y, 0);
 
-                    for (uint32_t slice = 0; slice < copy->copySize.depth; ++slice) {
+                    for (uint32_t slice = 0; slice < copy->copySize.depthOrArrayLayers; ++slice) {
                         [commandContext->EnsureBlit()
                               copyFromTexture:srcTexture->GetMTLTexture()
                                   sourceSlice:copy->source.origin.z + slice

diff --git a/src/dawn_native/metal/DeviceMTL.mm b/src/dawn_native/metal/DeviceMTL.mm
index e570639..6dcf97d 100644
--- a/src/dawn_native/metal/DeviceMTL.mm
+++ b/src/dawn_native/metal/DeviceMTL.mm

@@ -313,7 +313,7 @@
         const Extent3D clampedSize =
             texture->ClampToMipLevelVirtualSize(dst->mipLevel, dst->origin, copySizePixels);
         const uint32_t copyBaseLayer = dst->origin.z;
-        const uint32_t copyLayerCount = copySizePixels.depth;
+        const uint32_t copyLayerCount = copySizePixels.depthOrArrayLayers;
         const uint64_t bytesPerImage = dataLayout.rowsPerImage * dataLayout.bytesPerRow;
 
         MTLBlitOption blitOption = ComputeMTLBlitOption(texture->GetFormat(), dst->aspect);

diff --git a/src/dawn_native/metal/TextureMTL.mm b/src/dawn_native/metal/TextureMTL.mm
index d10c675..4918174 100644
--- a/src/dawn_native/metal/TextureMTL.mm
+++ b/src/dawn_native/metal/TextureMTL.mm

@@ -275,7 +275,7 @@
             return DAWN_VALIDATION_ERROR("IOSurface mip level count must be 1");
         }
 
-        if (descriptor->size.depth != 1) {
+        if (descriptor->size.depthOrArrayLayers != 1) {
             return DAWN_VALIDATION_ERROR("IOSurface array layer count must be 1");
         }
 
@@ -285,7 +285,7 @@
 
         if (descriptor->size.width != IOSurfaceGetWidthOfPlane(ioSurface, plane) ||
             descriptor->size.height != IOSurfaceGetHeightOfPlane(ioSurface, plane) ||
-            descriptor->size.depth != 1) {
+            descriptor->size.depthOrArrayLayers != 1) {
             return DAWN_VALIDATION_ERROR("IOSurface size doesn't match descriptor");
         }
 
@@ -317,7 +317,7 @@
 
         // Choose the correct MTLTextureType and paper over differences in how the array layer count
         // is specified.
-        mtlDesc.depth = descriptor->size.depth;
+        mtlDesc.depth = descriptor->size.depthOrArrayLayers;
         mtlDesc.arrayLength = 1;
         switch (descriptor->dimension) {
             case wgpu::TextureDimension::e2D:
@@ -532,8 +532,8 @@
                              (largestMipSize.height / blockInfo.height),
                          512llu);
 
-            // TODO(enga): Multiply by largestMipSize.depth and do a larger 3D copy to clear a whole
-            // range of subresources when tracking that is improved.
+            // TODO(enga): Multiply by largestMipSize.depthOrArrayLayers and do a larger 3D copy to
+            // clear a whole range of subresources when tracking that is improved.
             uint64_t bufferSize = largestMipBytesPerImage * 1;
 
             if (bufferSize > std::numeric_limits<NSUInteger>::max()) {

diff --git a/src/dawn_native/metal/UtilsMetal.mm b/src/dawn_native/metal/UtilsMetal.mm
index 953deb9..bc41ec5 100644
--- a/src/dawn_native/metal/UtilsMetal.mm
+++ b/src/dawn_native/metal/UtilsMetal.mm

@@ -60,9 +60,9 @@
         // compute the correct range when checking if the buffer is big enough to contain the
         // data for the whole copy. Instead of looking at the position of the last texel in the
         // buffer, it computes the volume of the 3D box with bytesPerRow * (rowsPerImage /
-        // format.blockHeight) * copySize.depth. For example considering the pixel buffer below
-        // where in memory, each row data (D) of the texture is followed by some padding data
-        // (P):
+        // format.blockHeight) * copySize.depthOrArrayLayers. For example considering the pixel
+        // buffer below where in memory, each row data (D) of the texture is followed by some
+        // padding data (P):
         //     |DDDDDDD|PP|
         //     |DDDDDDD|PP|
         //     |DDDDDDD|PP|
@@ -85,7 +85,8 @@
         ASSERT(texture->GetDimension() == wgpu::TextureDimension::e2D);
 
         // Check whether buffer size is big enough.
-        bool needWorkaround = bufferSize - bufferOffset < bytesPerImage * copyExtent.depth;
+        bool needWorkaround =
+            bufferSize - bufferOffset < bytesPerImage * copyExtent.depthOrArrayLayers;
         if (!needWorkaround) {
             copy.count = 1;
             copy.copies[0].bufferOffset = bufferOffset;
@@ -93,25 +94,25 @@
             copy.copies[0].bytesPerImage = bytesPerImage;
             copy.copies[0].textureOrigin = origin;
             copy.copies[0].copyExtent = {clampedCopyExtent.width, clampedCopyExtent.height,
-                                         copyExtent.depth};
+                                         copyExtent.depthOrArrayLayers};
             return copy;
         }
 
         uint64_t currentOffset = bufferOffset;
 
         // Doing all the copy except the last image.
-        if (copyExtent.depth > 1) {
+        if (copyExtent.depthOrArrayLayers > 1) {
             copy.copies[copy.count].bufferOffset = currentOffset;
             copy.copies[copy.count].bytesPerRow = bytesPerRow;
             copy.copies[copy.count].bytesPerImage = bytesPerImage;
             copy.copies[copy.count].textureOrigin = origin;
             copy.copies[copy.count].copyExtent = {clampedCopyExtent.width, clampedCopyExtent.height,
-                                                  copyExtent.depth - 1};
+                                                  copyExtent.depthOrArrayLayers - 1};
 
             ++copy.count;
 
             // Update offset to copy to the last image.
-            currentOffset += (copyExtent.depth - 1) * bytesPerImage;
+            currentOffset += (copyExtent.depthOrArrayLayers - 1) * bytesPerImage;
         }
 
         // Doing all the copy in last image except the last row.
@@ -121,7 +122,7 @@
             copy.copies[copy.count].bytesPerRow = bytesPerRow;
             copy.copies[copy.count].bytesPerImage = bytesPerRow * (copyBlockRowCount - 1);
             copy.copies[copy.count].textureOrigin = {origin.x, origin.y,
-                                                     origin.z + copyExtent.depth - 1};
+                                                     origin.z + copyExtent.depthOrArrayLayers - 1};
 
             ASSERT(copyExtent.height - blockInfo.height <
                    texture->GetMipLevelVirtualSize(mipLevel).height);
@@ -146,7 +147,7 @@
         copy.copies[copy.count].bytesPerImage = lastRowDataSize;
         copy.copies[copy.count].textureOrigin = {origin.x,
                                                  origin.y + copyExtent.height - blockInfo.height,
-                                                 origin.z + copyExtent.depth - 1};
+                                                 origin.z + copyExtent.depthOrArrayLayers - 1};
         copy.copies[copy.count].copyExtent = {clampedCopyExtent.width, lastRowCopyExtentHeight, 1};
         ++copy.count;
 

diff --git a/src/dawn_native/opengl/CommandBufferGL.cpp b/src/dawn_native/opengl/CommandBufferGL.cpp
index e78c68b..6573d12 100644
--- a/src/dawn_native/opengl/CommandBufferGL.cpp
+++ b/src/dawn_native/opengl/CommandBufferGL.cpp

@@ -472,7 +472,7 @@
                 blitMask |= GL_STENCIL_BUFFER_BIT;
             }
             // Iterate over all layers, doing a single blit for each.
-            for (uint32_t layer = 0; layer < copySize.depth; ++layer) {
+            for (uint32_t layer = 0; layer < copySize.depthOrArrayLayers; ++layer) {
                 // Bind all required aspects for this layer.
                 for (Aspect aspect : IterateEnumMask(src.aspect)) {
                     GLenum glAttachment;
@@ -652,7 +652,8 @@
                             if (texture->GetArrayLayers() > 1) {
                                 // TODO(jiawei.shao@intel.com): do a single copy when the data is
                                 // correctly packed.
-                                for (size_t copyZ = 0; copyZ < copyExtent.depth; ++copyZ) {
+                                for (size_t copyZ = 0; copyZ < copyExtent.depthOrArrayLayers;
+                                     ++copyZ) {
                                     uintptr_t offsetPerImage = static_cast<uintptr_t>(
                                         src.offset + copyZ * src.bytesPerRow * src.rowsPerImage);
                                     uint32_t dstOriginY = dst.origin.y;
@@ -702,13 +703,15 @@
 
                             uint64_t copyDataSize = (copySize.width / blockInfo.width) *
                                                     (copySize.height / blockInfo.height) *
-                                                    blockInfo.byteSize * copySize.depth;
+                                                    blockInfo.byteSize *
+                                                    copySize.depthOrArrayLayers;
 
                             if (texture->GetArrayLayers() > 1) {
                                 gl.CompressedTexSubImage3D(
                                     target, dst.mipLevel, dst.origin.x, dst.origin.y, dst.origin.z,
-                                    copyExtent.width, copyExtent.height, copyExtent.depth,
-                                    format.internalFormat, copyDataSize,
+                                    copyExtent.width, copyExtent.height,
+                                    copyExtent.depthOrArrayLayers, format.internalFormat,
+                                    copyDataSize,
                                     reinterpret_cast<void*>(static_cast<uintptr_t>(src.offset)));
                             } else {
                                 gl.CompressedTexSubImage2D(
@@ -734,8 +737,8 @@
                                 if (texture->GetArrayLayers() > 1) {
                                     gl.TexSubImage3D(target, dst.mipLevel, dst.origin.x,
                                                      dst.origin.y, dst.origin.z, copySize.width,
-                                                     copySize.height, copySize.depth, format.format,
-                                                     format.type,
+                                                     copySize.height, copySize.depthOrArrayLayers,
+                                                     format.format, format.type,
                                                      reinterpret_cast<void*>(
                                                          static_cast<uintptr_t>(src.offset)));
                                 } else {
@@ -839,7 +842,7 @@
                             }
 
                             const uint64_t bytesPerImage = dst.bytesPerRow * dst.rowsPerImage;
-                            for (uint32_t layer = 0; layer < copySize.depth; ++layer) {
+                            for (uint32_t layer = 0; layer < copySize.depthOrArrayLayers; ++layer) {
                                 gl.FramebufferTextureLayer(GL_READ_FRAMEBUFFER, glAttachment,
                                                            texture->GetHandle(), src.mipLevel,
                                                            src.origin.z + layer);
@@ -892,7 +895,8 @@
                                             src.mipLevel, src.origin.x, src.origin.y, src.origin.z,
                                             dstTexture->GetHandle(), dstTexture->GetGLTarget(),
                                             dst.mipLevel, dst.origin.x, dst.origin.y, dst.origin.z,
-                                            copySize.width, copySize.height, copy->copySize.depth);
+                                            copySize.width, copySize.height,
+                                            copy->copySize.depthOrArrayLayers);
                     } else {
                         CopyTextureToTextureWithBlit(gl, src, dst, copySize);
                     }

diff --git a/src/dawn_native/opengl/QueueGL.cpp b/src/dawn_native/opengl/QueueGL.cpp
index 23eb143..17260b9 100644
--- a/src/dawn_native/opengl/QueueGL.cpp
+++ b/src/dawn_native/opengl/QueueGL.cpp

@@ -59,7 +59,8 @@
         const OpenGLFunctions& gl = ToBackend(GetDevice())->gl;
 
         Texture* texture = ToBackend(destination.texture);
-        SubresourceRange range(Aspect::Color, {destination.origin.z, writeSizePixel.depth},
+        SubresourceRange range(Aspect::Color,
+                               {destination.origin.z, writeSizePixel.depthOrArrayLayers},
                                {destination.mipLevel, 1});
         if (IsCompleteSubresourceCopiedTo(texture, writeSizePixel, destination.mipLevel)) {
             texture->SetIsSubresourceContentInitialized(true, range);
@@ -97,7 +98,7 @@
                 const uint8_t* slice = static_cast<const uint8_t*>(data);
 
                 for (uint32_t z = destination.origin.z;
-                     z < destination.origin.z + writeSizePixel.depth; ++z) {
+                     z < destination.origin.z + writeSizePixel.depthOrArrayLayers; ++z) {
                     const uint8_t* d = slice;
 
                     for (uint32_t y = destination.origin.y;
@@ -122,8 +123,8 @@
                 gl.PixelStorei(GL_UNPACK_IMAGE_HEIGHT, dataLayout.rowsPerImage * blockInfo.height);
                 gl.TexSubImage3D(target, destination.mipLevel, destination.origin.x,
                                  destination.origin.y, destination.origin.z, writeSizePixel.width,
-                                 writeSizePixel.height, writeSizePixel.depth, format.format,
-                                 format.type, data);
+                                 writeSizePixel.height, writeSizePixel.depthOrArrayLayers,
+                                 format.format, format.type, data);
                 gl.PixelStorei(GL_UNPACK_IMAGE_HEIGHT, 0);
             }
             gl.PixelStorei(GL_UNPACK_ROW_LENGTH, 0);
@@ -138,7 +139,7 @@
                 }
             } else {
                 const uint8_t* slice = static_cast<const uint8_t*>(data);
-                for (uint32_t z = 0; z < writeSizePixel.depth; ++z) {
+                for (uint32_t z = 0; z < writeSizePixel.depthOrArrayLayers; ++z) {
                     const uint8_t* d = slice;
                     for (uint32_t y = 0; y < writeSizePixel.height; ++y) {
                         gl.TexSubImage3D(target, destination.mipLevel, destination.origin.x,

diff --git a/src/dawn_native/opengl/TextureGL.cpp b/src/dawn_native/opengl/TextureGL.cpp
index 5582d97..c7fc9ef 100644
--- a/src/dawn_native/opengl/TextureGL.cpp
+++ b/src/dawn_native/opengl/TextureGL.cpp

@@ -29,7 +29,7 @@
         GLenum TargetForTexture(const TextureDescriptor* descriptor) {
             switch (descriptor->dimension) {
                 case wgpu::TextureDimension::e2D:
-                    if (descriptor->size.depth > 1) {
+                    if (descriptor->size.depthOrArrayLayers > 1) {
                         ASSERT(descriptor->sampleCount == 1);
                         return GL_TEXTURE_2D_ARRAY;
                     } else {

diff --git a/src/dawn_native/vulkan/CommandBufferVk.cpp b/src/dawn_native/vulkan/CommandBufferVk.cpp
index 98bfb58..62a0caf 100644
--- a/src/dawn_native/vulkan/CommandBufferVk.cpp
+++ b/src/dawn_native/vulkan/CommandBufferVk.cpp

@@ -58,7 +58,7 @@
             Extent3D imageExtentDst = ComputeTextureCopyExtent(dstCopy, copySize);
             return imageExtentSrc.width == imageExtentDst.width &&
                    imageExtentSrc.height == imageExtentDst.height &&
-                   imageExtentSrc.depth == imageExtentDst.depth;
+                   imageExtentSrc.depthOrArrayLayers == imageExtentDst.depthOrArrayLayers;
         }
 
         VkImageCopy ComputeImageCopyRegion(const TextureCopy& srcCopy,
@@ -76,7 +76,7 @@
             region.srcSubresource.aspectMask = VulkanAspectMask(aspect);
             region.srcSubresource.mipLevel = srcCopy.mipLevel;
             region.srcSubresource.baseArrayLayer = srcCopy.origin.z;
-            region.srcSubresource.layerCount = copySize.depth;
+            region.srcSubresource.layerCount = copySize.depthOrArrayLayers;
 
             region.srcOffset.x = srcCopy.origin.x;
             region.srcOffset.y = srcCopy.origin.y;
@@ -85,7 +85,7 @@
             region.dstSubresource.aspectMask = VulkanAspectMask(aspect);
             region.dstSubresource.mipLevel = dstCopy.mipLevel;
             region.dstSubresource.baseArrayLayer = dstCopy.origin.z;
-            region.dstSubresource.layerCount = copySize.depth;
+            region.dstSubresource.layerCount = copySize.depthOrArrayLayers;
 
             region.dstOffset.x = dstCopy.origin.x;
             region.dstOffset.y = dstCopy.origin.y;
@@ -459,7 +459,7 @@
         // Create the temporary buffer. Note that We don't need to respect WebGPU's 256 alignment
         // because it isn't a hard constraint in Vulkan.
         uint64_t tempBufferSize =
-            widthInBlocks * heightInBlocks * copySize.depth * blockInfo.byteSize;
+            widthInBlocks * heightInBlocks * copySize.depthOrArrayLayers * blockInfo.byteSize;
         BufferDescriptor tempBufferDescriptor;
         tempBufferDescriptor.size = tempBufferSize;
         tempBufferDescriptor.usage = wgpu::BufferUsage::CopySrc | wgpu::BufferUsage::CopyDst;
@@ -689,8 +689,8 @@
                         // subresources should all be GENERAL instead of what we set now. Currently
                         // it is not allowed to copy with overlapped subresources, but we still
                         // add the ASSERT here as a reminder for this possible misuse.
-                        ASSERT(
-                            !IsRangeOverlapped(src.origin.z, dst.origin.z, copy->copySize.depth));
+                        ASSERT(!IsRangeOverlapped(src.origin.z, dst.origin.z,
+                                                  copy->copySize.depthOrArrayLayers));
                     }
 
                     // TODO after Yunchao's CL

diff --git a/src/dawn_native/vulkan/TextureVk.cpp b/src/dawn_native/vulkan/TextureVk.cpp
index 870b59f..28f9499 100644
--- a/src/dawn_native/vulkan/TextureVk.cpp
+++ b/src/dawn_native/vulkan/TextureVk.cpp

@@ -199,12 +199,12 @@
                 case wgpu::TextureDimension::e2D:
                     info->imageType = VK_IMAGE_TYPE_2D;
                     info->extent = {size.width, size.height, 1};
-                    info->arrayLayers = size.depth;
+                    info->arrayLayers = size.depthOrArrayLayers;
                     break;
 
                 case wgpu::TextureDimension::e3D:
                     info->imageType = VK_IMAGE_TYPE_3D;
-                    info->extent = {size.width, size.height, size.depth};
+                    info->extent = {size.width, size.height, size.depthOrArrayLayers};
                     info->arrayLayers = 1;
                     break;
 
@@ -456,7 +456,7 @@
             return DAWN_VALIDATION_ERROR("Mip level count must be 1");
         }
 
-        if (descriptor->size.depth != 1) {
+        if (descriptor->size.depthOrArrayLayers != 1) {
             return DAWN_VALIDATION_ERROR("Array layer count must be 1");
         }
 

diff --git a/src/dawn_native/vulkan/UtilsVulkan.cpp b/src/dawn_native/vulkan/UtilsVulkan.cpp
index 7e58319..6167201 100644
--- a/src/dawn_native/vulkan/UtilsVulkan.cpp
+++ b/src/dawn_native/vulkan/UtilsVulkan.cpp

@@ -130,7 +130,7 @@
                 region.imageOffset.z = 0;
 
                 region.imageSubresource.baseArrayLayer = textureCopy.origin.z;
-                region.imageSubresource.layerCount = copySize.depth;
+                region.imageSubresource.layerCount = copySize.depthOrArrayLayers;
 
                 Extent3D imageExtent = ComputeTextureCopyExtent(textureCopy, copySize);
                 region.imageExtent.width = imageExtent.width;

diff --git a/src/tests/end2end/BindGroupTests.cpp b/src/tests/end2end/BindGroupTests.cpp
index a518e89..e44aeb2 100644
--- a/src/tests/end2end/BindGroupTests.cpp
+++ b/src/tests/end2end/BindGroupTests.cpp

@@ -298,7 +298,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kRTSize;
     descriptor.size.height = kRTSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/BufferZeroInitTests.cpp b/src/tests/end2end/BufferZeroInitTests.cpp
index 327723e..67226ef 100644
--- a/src/tests/end2end/BufferZeroInitTests.cpp
+++ b/src/tests/end2end/BufferZeroInitTests.cpp

@@ -96,7 +96,7 @@
 
         wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
 
-        for (uint32_t arrayLayer = 0; arrayLayer < size.depth; ++arrayLayer) {
+        for (uint32_t arrayLayer = 0; arrayLayer < size.depthOrArrayLayers; ++arrayLayer) {
             wgpu::TextureViewDescriptor viewDescriptor;
             viewDescriptor.format = format;
             viewDescriptor.dimension = wgpu::TextureViewDimension::e2D;
@@ -145,7 +145,7 @@
         const uint64_t expectedValueCount = bufferSize / sizeof(float);
         std::vector<float> expectedValues(expectedValueCount, 0.f);
 
-        for (uint32_t slice = 0; slice < spec.textureSize.depth; ++slice) {
+        for (uint32_t slice = 0; slice < spec.textureSize.depthOrArrayLayers; ++slice) {
             const uint64_t baseOffsetBytesPerSlice =
                 spec.bufferOffset + spec.bytesPerRow * spec.rowsPerImage * slice;
             for (uint32_t y = 0; y < spec.textureSize.height; ++y) {
@@ -963,7 +963,8 @@
     constexpr wgpu::Extent3D kTextureSize = {64u, 4u, 3u};
 
     // bytesPerRow == texelBlockSizeInBytes * copySize.width && rowsPerImage == copySize.height &&
-    // bytesPerRow * (rowsPerImage * (copySize.depth - 1) + copySize.height) == buffer.size
+    // bytesPerRow * (rowsPerImage * (copySize.depthOrArrayLayers - 1) + copySize.height) ==
+    // buffer.size
     {
         TestBufferZeroInitInCopyTextureToBuffer(
             {kTextureSize, 0u, 0u, kTextureBytesPerRowAlignment, kTextureSize.height, 0u});
@@ -976,7 +977,7 @@
             {kTextureSize, 0u, 0u, kTextureBytesPerRowAlignment, kRowsPerImage, 1u});
     }
 
-    // bytesPerRow * rowsPerImage * copySize.depth < buffer.size
+    // bytesPerRow * rowsPerImage * copySize.depthOrArrayLayers < buffer.size
     {
         constexpr uint64_t kExtraBufferSize = 16u;
         TestBufferZeroInitInCopyTextureToBuffer({kTextureSize, 0u, kExtraBufferSize,

diff --git a/src/tests/end2end/ColorStateTests.cpp b/src/tests/end2end/ColorStateTests.cpp
index b9b6d6c..91db43a 100644
--- a/src/tests/end2end/ColorStateTests.cpp
+++ b/src/tests/end2end/ColorStateTests.cpp

@@ -771,7 +771,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kRTSize;
     descriptor.size.height = kRTSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/CompressedTextureFormatTests.cpp b/src/tests/end2end/CompressedTextureFormatTests.cpp
index a95f834a..06b8afa 100644
--- a/src/tests/end2end/CompressedTextureFormatTests.cpp
+++ b/src/tests/end2end/CompressedTextureFormatTests.cpp

@@ -64,14 +64,14 @@
             copyRowsPerImage = copyHeightInBlock;
         }
         uint32_t copyBytesPerImage = copyBytesPerRow * copyRowsPerImage;
-        uint32_t uploadBufferSize =
-            copyConfig.bufferOffset + copyBytesPerImage * copyConfig.copyExtent3D.depth;
+        uint32_t uploadBufferSize = copyConfig.bufferOffset +
+                                    copyBytesPerImage * copyConfig.copyExtent3D.depthOrArrayLayers;
 
         // Fill data with the pre-prepared one-block compressed texture data.
         std::vector<uint8_t> data(uploadBufferSize, 0);
         std::vector<uint8_t> oneBlockCompressedTextureData =
             GetOneBlockBCFormatTextureData(copyConfig.textureDescriptor.format);
-        for (uint32_t layer = 0; layer < copyConfig.copyExtent3D.depth; ++layer) {
+        for (uint32_t layer = 0; layer < copyConfig.copyExtent3D.depthOrArrayLayers; ++layer) {
             for (uint32_t h = 0; h < copyHeightInBlock; ++h) {
                 for (uint32_t w = 0; w < copyWidthInBlock; ++w) {
                     uint32_t uploadBufferOffset = copyConfig.bufferOffset +
@@ -227,14 +227,14 @@
         if (config.copyOrigin3D.y + config.copyExtent3D.height > virtualSizeAtLevel.height) {
             noPaddingExtent3D.height = virtualSizeAtLevel.height - config.copyOrigin3D.y;
         }
-        noPaddingExtent3D.depth = 1u;
+        noPaddingExtent3D.depthOrArrayLayers = 1u;
 
         std::vector<RGBA8> expectedData =
             GetExpectedData(config.textureDescriptor.format, noPaddingExtent3D);
 
         wgpu::Origin3D firstLayerCopyOrigin = {config.copyOrigin3D.x, config.copyOrigin3D.y, 0};
         for (uint32_t layer = config.copyOrigin3D.z;
-             layer < config.copyOrigin3D.z + config.copyExtent3D.depth; ++layer) {
+             layer < config.copyOrigin3D.z + config.copyExtent3D.depthOrArrayLayers; ++layer) {
             wgpu::BindGroup bindGroup = CreateBindGroupForTest(
                 renderPipeline.GetBindGroupLayout(0), bcTexture, config.textureDescriptor.format,
                 layer, config.viewMipmapLevel);
@@ -391,7 +391,7 @@
     static std::vector<RGBA8> FillExpectedData(const wgpu::Extent3D& testRegion,
                                                RGBA8 leftColorInBlock,
                                                RGBA8 rightColorInBlock) {
-        ASSERT(testRegion.depth == 1);
+        ASSERT(testRegion.depthOrArrayLayers == 1);
 
         std::vector<RGBA8> expectedData(testRegion.width * testRegion.height, leftColorInBlock);
         for (uint32_t y = 0; y < testRegion.height; ++y) {
@@ -409,7 +409,7 @@
     static wgpu::Extent3D GetVirtualSizeAtLevel(const CopyConfig& config) {
         return {config.textureDescriptor.size.width >> config.viewMipmapLevel,
                 config.textureDescriptor.size.height >> config.viewMipmapLevel,
-                config.textureDescriptor.size.depth};
+                config.textureDescriptor.size.depthOrArrayLayers};
     }
 
     static wgpu::Extent3D GetPhysicalSizeAtLevel(const CopyConfig& config) {
@@ -484,7 +484,7 @@
     config.copyExtent3D = config.textureDescriptor.size;
 
     constexpr uint32_t kArrayLayerCount = 3;
-    config.textureDescriptor.size.depth = kArrayLayerCount;
+    config.textureDescriptor.size.depthOrArrayLayers = kArrayLayerCount;
     config.copyOrigin3D.z = kArrayLayerCount - 1;
 
     for (wgpu::TextureFormat format : utils::kBCFormats) {
@@ -973,8 +973,8 @@
 }
 
 // Test the workaround in the B2T copies when (bufferSize - bufferOffset < bytesPerImage *
-// copyExtent.depth) on Metal backends. As copyExtent.depth can only be 1 for BC formats, on Metal
-// backend we will use two copies to implement such copy.
+// copyExtent.depthOrArrayLayers) on Metal backends. As copyExtent.depthOrArrayLayers can only be 1
+// for BC formats, on Metal backend we will use two copies to implement such copy.
 TEST_P(CompressedTextureBCFormatTest, LargeImageHeight) {
     // TODO(jiawei.shao@intel.com): find out why this test fails on Windows Intel OpenGL drivers.
     DAWN_SKIP_TEST_IF(IsIntel() && IsOpenGL() && IsWindows());
@@ -995,7 +995,7 @@
 }
 
 // Test the workaround in the B2T copies when (bufferSize - bufferOffset < bytesPerImage *
-// copyExtent.depth) and copyExtent needs to be clamped.
+// copyExtent.depthOrArrayLayers) and copyExtent needs to be clamped.
 TEST_P(CompressedTextureBCFormatTest, LargeImageHeightAndClampedCopyExtent) {
     // TODO(jiawei.shao@intel.com): find out why this test fails on Windows Intel OpenGL drivers.
     DAWN_SKIP_TEST_IF(IsIntel() && IsOpenGL() && IsWindows());
@@ -1055,7 +1055,7 @@
     config.rowsPerImage = 8;
 
     config.copyExtent3D = config.textureDescriptor.size;
-    config.copyExtent3D.depth = kArrayLayerCount;
+    config.copyExtent3D.depthOrArrayLayers = kArrayLayerCount;
 
     for (wgpu::TextureFormat format : utils::kBCFormats) {
         config.textureDescriptor.format = format;
@@ -1084,7 +1084,7 @@
     constexpr uint32_t kCopyLayerCount = 2;
     config.copyOrigin3D = {0, 0, kCopyBaseArrayLayer};
     config.copyExtent3D = config.textureDescriptor.size;
-    config.copyExtent3D.depth = kCopyLayerCount;
+    config.copyExtent3D.depthOrArrayLayers = kCopyLayerCount;
 
     for (wgpu::TextureFormat format : utils::kBCFormats) {
         config.textureDescriptor.format = format;

diff --git a/src/tests/end2end/CopyTests.cpp b/src/tests/end2end/CopyTests.cpp
index f6d6a3d..0c457cb 100644
--- a/src/tests/end2end/CopyTests.cpp
+++ b/src/tests/end2end/CopyTests.cpp

@@ -46,7 +46,7 @@
     static std::vector<uint8_t> GetExpectedTextureData(const utils::TextureDataCopyLayout& layout) {
         uint32_t bytesPerTexelBlock = layout.bytesPerRow / layout.texelBlocksPerRow;
         std::vector<uint8_t> textureData(layout.byteLength);
-        for (uint32_t layer = 0; layer < layout.mipSize.depth; ++layer) {
+        for (uint32_t layer = 0; layer < layout.mipSize.depthOrArrayLayers; ++layer) {
             const uint32_t byteOffsetPerSlice = layout.bytesPerImage * layer;
             for (uint32_t y = 0; y < layout.mipSize.height; ++y) {
                 for (uint32_t x = 0; x < layout.mipSize.width * bytesPerTexelBlock; ++x) {
@@ -64,7 +64,7 @@
     static std::vector<RGBA8> GetExpectedTextureDataRGBA8(
         const utils::TextureDataCopyLayout& layout) {
         std::vector<RGBA8> textureData(layout.texelBlockCount);
-        for (uint32_t layer = 0; layer < layout.mipSize.depth; ++layer) {
+        for (uint32_t layer = 0; layer < layout.mipSize.depthOrArrayLayers; ++layer) {
             const uint32_t texelIndexOffsetPerSlice = layout.texelBlocksPerImage * layer;
             for (uint32_t y = 0; y < layout.mipSize.height; ++y) {
                 for (uint32_t x = 0; x < layout.mipSize.width; ++x) {
@@ -185,7 +185,7 @@
         // Texels in single slice.
         const uint32_t texelCountInCopyRegion = utils::GetTexelCountInCopyRegion(
             bufferSpec.bytesPerRow, bufferSpec.rowsPerImage, copySizePerSlice, textureSpec.format);
-        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depth;
+        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depthOrArrayLayers;
         std::vector<RGBA8> expected(texelCountInCopyRegion);
         for (uint32_t slice = textureSpec.copyOrigin.z; slice < maxArrayLayer; ++slice) {
             // Pack the data used to create the upload buffer in the specified copy region to have
@@ -207,7 +207,7 @@
                 << ", " << textureSpec.copyOrigin.y << ", " << textureSpec.copyOrigin.z << "), ("
                 << textureSpec.copyOrigin.x + copySize.width << ", "
                 << textureSpec.copyOrigin.y + copySize.height << ", "
-                << textureSpec.copyOrigin.z + copySize.depth << ")) from "
+                << textureSpec.copyOrigin.z + copySize.depthOrArrayLayers << ")) from "
                 << textureSpec.textureSize.width << " x " << textureSpec.textureSize.height
                 << " texture at mip level " << textureSpec.copyLevel << " layer " << slice << " to "
                 << bufferSpec.size << "-byte buffer with offset " << bufferOffset
@@ -258,7 +258,7 @@
                 textureSpec.format, textureSpec.textureSize, textureSpec.copyLevel,
                 bufferSpec.rowsPerImage);
 
-        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depth;
+        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depthOrArrayLayers;
 
         wgpu::ImageCopyBuffer imageCopyBuffer = utils::CreateImageCopyBuffer(
             buffer, bufferSpec.offset, bufferSpec.bytesPerRow, bufferSpec.rowsPerImage);
@@ -333,7 +333,9 @@
         // `level` mip level
         const utils::TextureDataCopyLayout srcDataCopyLayout =
             utils::GetTextureDataCopyLayoutForTexture2DAtLevel(
-                format, {srcSpec.textureSize.width, srcSpec.textureSize.height, copySize.depth},
+                format,
+                {srcSpec.textureSize.width, srcSpec.textureSize.height,
+                 copySize.depthOrArrayLayers},
                 srcSpec.copyLevel);
 
         // Initialize the source texture
@@ -358,10 +360,12 @@
         encoder.CopyTextureToTexture(&srcImageCopyTexture, &dstImageCopyTexture, &copySize);
 
         // Copy the data from the srcSpec.copyOrigin.z-th layer to (srcSpec.copyOrigin.z +
-        // copySize.depth)-th layer of dstTexture to outputBuffer
+        // copySize.depthOrArrayLayers)-th layer of dstTexture to outputBuffer
         const utils::TextureDataCopyLayout dstDataCopyLayout =
             utils::GetTextureDataCopyLayoutForTexture2DAtLevel(
-                format, {dstSpec.textureSize.width, dstSpec.textureSize.height, copySize.depth},
+                format,
+                {dstSpec.textureSize.width, dstSpec.textureSize.height,
+                 copySize.depthOrArrayLayers},
                 dstSpec.copyLevel);
         wgpu::BufferDescriptor outputBufferDescriptor;
         outputBufferDescriptor.size = dstDataCopyLayout.byteLength;
@@ -384,7 +388,7 @@
                 bytesPerTexel);
 
             std::vector<uint8_t> expectedDstDataPerSlice(validDataSizePerDstTextureLayer);
-            for (uint32_t slice = 0; slice < copySize.depth; ++slice) {
+            for (uint32_t slice = 0; slice < copySize.depthOrArrayLayers; ++slice) {
                 // For each source texture array slice involved in the copy, emulate the T2T copy
                 // on the CPU side by "copying" the copy data from the "source texture"
                 // (srcTextureCopyData) to the "destination texture" (expectedDstDataPerSlice).

diff --git a/src/tests/end2end/CopyTextureForBrowserTests.cpp b/src/tests/end2end/CopyTextureForBrowserTests.cpp
index deb3a70..17c3fd6 100644
--- a/src/tests/end2end/CopyTextureForBrowserTests.cpp
+++ b/src/tests/end2end/CopyTextureForBrowserTests.cpp

@@ -36,7 +36,7 @@
 
     static std::vector<RGBA8> GetSourceTextureData(const utils::TextureDataCopyLayout& layout) {
         std::vector<RGBA8> textureData(layout.texelBlockCount);
-        for (uint32_t layer = 0; layer < layout.mipSize.depth; ++layer) {
+        for (uint32_t layer = 0; layer < layout.mipSize.depthOrArrayLayers; ++layer) {
             const uint32_t sliceOffset = layout.texelBlocksPerImage * layer;
             for (uint32_t y = 0; y < layout.mipSize.height; ++y) {
                 const uint32_t rowOffset = layout.texelBlocksPerRow * y;
@@ -141,7 +141,8 @@
         const utils::TextureDataCopyLayout copyLayout =
             utils::GetTextureDataCopyLayoutForTexture2DAtLevel(
                 kTextureFormat,
-                {srcSpec.textureSize.width, srcSpec.textureSize.height, copySize.depth},
+                {srcSpec.textureSize.width, srcSpec.textureSize.height,
+                 copySize.depthOrArrayLayers},
                 srcSpec.level);
 
         const std::vector<RGBA8> textureArrayCopyData = GetSourceTextureData(copyLayout);

diff --git a/src/tests/end2end/D3D12ResourceWrappingTests.cpp b/src/tests/end2end/D3D12ResourceWrappingTests.cpp
index 47f311b..0a703c7 100644
--- a/src/tests/end2end/D3D12ResourceWrappingTests.cpp
+++ b/src/tests/end2end/D3D12ResourceWrappingTests.cpp

@@ -191,7 +191,7 @@
 // Test an error occurs if the descriptor depth isn't 1
 TEST_P(D3D12SharedHandleValidation, InvalidDepth) {
     DAWN_SKIP_TEST_IF(UsesWire());
-    baseDawnDescriptor.size.depth = 2;
+    baseDawnDescriptor.size.depthOrArrayLayers = 2;
 
     wgpu::Texture texture;
     ComPtr<ID3D11Texture2D> d3d11Texture;

diff --git a/src/tests/end2end/DeprecatedAPITests.cpp b/src/tests/end2end/DeprecatedAPITests.cpp
index 9d69f89..e1b8491 100644
--- a/src/tests/end2end/DeprecatedAPITests.cpp
+++ b/src/tests/end2end/DeprecatedAPITests.cpp

@@ -172,6 +172,291 @@
     EXPECT_DEPRECATION_WARNING(queue.CreateFence());
 }
 
+// Test GPUExtent3D.depth deprecation in TextureDescriptor.size
+TEST_P(DeprecationTests, GPUExtent3DDepthDeprecationTextureDescriptor) {
+    wgpu::TextureDescriptor kBaseDesc;
+    kBaseDesc.usage = wgpu::TextureUsage::Sampled;
+    kBaseDesc.size.width = 1;
+    kBaseDesc.size.height = 1;
+    kBaseDesc.format = wgpu::TextureFormat::RGBA8Unorm;
+    kBaseDesc.dimension = wgpu::TextureDimension::e2D;
+
+    {
+        // Valid: default
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        wgpu::Texture texture;
+        texture = device.CreateTexture(&desc);
+    }
+    {
+        // Warning: use deprecated depth but still valid
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        desc.mipLevelCount = 2;
+        desc.size.width = 2;
+        desc.size.height = 2;
+        desc.size.depth = 2;
+        wgpu::Texture texture;
+        EXPECT_DEPRECATION_WARNING(texture = device.CreateTexture(&desc));
+    }
+    {
+        // Warning: use deprecated depth
+        // Error: use deprecated depth and the descriptor is invalid
+        // because 2D texture with depth == 0 is not allowed
+        // This is to verify the deprecated depth is picked up by the implementation.
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        desc.size.depth = 0;
+        wgpu::Texture texture;
+        ASSERT_DEVICE_ERROR(EXPECT_DEPRECATION_WARNING(texture = device.CreateTexture(&desc)));
+    }
+    {
+        // Error: use both deprecated depth and depthOrArrayLayers
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        desc.size.depth = 2;
+        desc.size.depthOrArrayLayers = 2;
+        wgpu::Texture texture;
+        ASSERT_DEVICE_ERROR(texture = device.CreateTexture(&desc));
+    }
+    {
+        // Valid: use updated depthOrArrayLayers
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        desc.mipLevelCount = 2;
+        desc.size.width = 2;
+        desc.size.height = 2;
+        desc.size.depthOrArrayLayers = 2;
+        wgpu::Texture texture;
+        texture = device.CreateTexture(&desc);
+    }
+    {
+        // Error: use updated depthOrArrayLayers and the descriptor is invalid
+        // because 2D texture with depthOrArrayLayers == 0 is not allowed
+        wgpu::TextureDescriptor desc = kBaseDesc;
+        desc.size.depthOrArrayLayers = 0;
+        wgpu::Texture texture;
+        ASSERT_DEVICE_ERROR(texture = device.CreateTexture(&desc));
+    }
+}
+
+// Test GPUExtent3D.depth deprecation in CopyBufferToTexture, CopyTextureToBuffer, and
+// CopyTextureToTexture
+TEST_P(DeprecationTests, GPUExtent3DDepthDeprecationCopy) {
+    wgpu::BufferDescriptor bufferDesc;
+    bufferDesc.size = 4 * 256;
+    bufferDesc.usage = wgpu::BufferUsage::CopySrc | wgpu::BufferUsage::CopyDst;
+    wgpu::Buffer srcBuffer = device.CreateBuffer(&bufferDesc);
+
+    wgpu::TextureDescriptor dstTextureDesc;
+    dstTextureDesc.usage = wgpu::TextureUsage::CopyDst;
+    dstTextureDesc.size.width = 4;
+    dstTextureDesc.size.height = 4;
+    dstTextureDesc.size.depthOrArrayLayers = 1;
+    dstTextureDesc.format = wgpu::TextureFormat::RGBA8Unorm;
+    dstTextureDesc.dimension = wgpu::TextureDimension::e2D;
+    wgpu::Texture dstTexture = device.CreateTexture(&dstTextureDesc);
+
+    wgpu::TextureDescriptor srcTextureDesc = dstTextureDesc;
+    srcTextureDesc.usage = wgpu::TextureUsage::CopySrc;
+    wgpu::Texture srcTexture = device.CreateTexture(&srcTextureDesc);
+
+    wgpu::ImageCopyBuffer imageCopyBuffer = utils::CreateImageCopyBuffer(srcBuffer, 0, 256, 4);
+    wgpu::ImageCopyTexture imageCopyDstTexture =
+        utils::CreateImageCopyTexture(dstTexture, 0, {0, 0, 0}, wgpu::TextureAspect::All);
+    wgpu::ImageCopyTexture imageCopySrcTexture =
+        utils::CreateImageCopyTexture(srcTexture, 0, {0, 0, 0}, wgpu::TextureAspect::All);
+
+    wgpu::Extent3D kBaseExtent3D;
+    kBaseExtent3D.width = 4;
+    kBaseExtent3D.height = 4;
+
+    {
+        // Valid: default
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+        encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D);
+        encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D);
+        encoder.CopyTextureToTexture(&imageCopySrcTexture, &imageCopyDstTexture, &extent3D);
+        encoder.Finish();
+    }
+    {
+        // Warning: empty copy use deprecated depth == 0 but still valid
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depth = 0;
+        wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+        EXPECT_DEPRECATION_WARNING(
+            encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D));
+        EXPECT_DEPRECATION_WARNING(
+            encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D));
+        EXPECT_DEPRECATION_WARNING(
+            encoder.CopyTextureToTexture(&imageCopySrcTexture, &imageCopyDstTexture, &extent3D));
+        encoder.Finish();
+    }
+    {
+        // Warning: use deprecated depth
+        // Error: depth > 1
+        // This is to verify the deprecated depth is picked up by the implementation.
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.depth = 2;
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            EXPECT_DEPRECATION_WARNING(
+                encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D));
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            EXPECT_DEPRECATION_WARNING(
+                encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D));
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            EXPECT_DEPRECATION_WARNING(encoder.CopyTextureToTexture(
+                &imageCopySrcTexture, &imageCopyDstTexture, &extent3D));
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+    }
+    {
+        // Error: use both deprecated depth and depthOrArrayLayers
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depth = 0;
+        extent3D.depthOrArrayLayers = 0;
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyTextureToTexture(&imageCopySrcTexture, &imageCopyDstTexture, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+    }
+    {
+        // Valid: use updated depthOrArrayLayers
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depthOrArrayLayers = 0;
+        wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+        encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D);
+        encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D);
+        encoder.CopyTextureToTexture(&imageCopySrcTexture, &imageCopyDstTexture, &extent3D);
+        encoder.Finish();
+    }
+    {
+        // Error: use updated depthOrArrayLayers and is invalid
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.depthOrArrayLayers = 2;
+        wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyBufferToTexture(&imageCopyBuffer, &imageCopyDstTexture, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyTextureToBuffer(&imageCopySrcTexture, &imageCopyBuffer, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+        {
+            wgpu::CommandEncoder encoder = device.CreateCommandEncoder();
+            encoder.CopyTextureToTexture(&imageCopySrcTexture, &imageCopyDstTexture, &extent3D);
+            ASSERT_DEVICE_ERROR(encoder.Finish());
+        }
+    }
+}
+
+// Test GPUExtent3D.depth deprecation in WriteTexture
+TEST_P(DeprecationTests, GPUExtent3DDepthDeprecationWriteTexture) {
+    wgpu::TextureDescriptor dstTextureDesc;
+    dstTextureDesc.usage = wgpu::TextureUsage::CopyDst;
+    dstTextureDesc.size.width = 4;
+    dstTextureDesc.size.height = 4;
+    dstTextureDesc.size.depthOrArrayLayers = 1;
+    dstTextureDesc.format = wgpu::TextureFormat::RGBA8Unorm;
+    dstTextureDesc.dimension = wgpu::TextureDimension::e2D;
+    wgpu::Texture dstTexture = device.CreateTexture(&dstTextureDesc);
+
+    size_t dataSize = 4 * 256;
+    std::vector<uint8_t> data(dataSize);
+
+    wgpu::TextureDataLayout textureDataLayout;
+    textureDataLayout.offset = 0;
+    textureDataLayout.bytesPerRow = 256;
+    textureDataLayout.rowsPerImage = 4;
+
+    wgpu::ImageCopyTexture imageCopyDstTexture =
+        utils::CreateImageCopyTexture(dstTexture, 0, {0, 0, 0}, wgpu::TextureAspect::All);
+
+    wgpu::Extent3D kBaseExtent3D;
+    kBaseExtent3D.width = 4;
+    kBaseExtent3D.height = 4;
+
+    {
+        // Valid: default
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        wgpu::Queue queue = device.GetQueue();
+        queue.WriteTexture(&imageCopyDstTexture, data.data(), dataSize, &textureDataLayout,
+                           &extent3D);
+    }
+    {
+        // Warning: use deprecated depth == 0 but still valid
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depth = 0;
+        wgpu::Queue queue = device.GetQueue();
+        EXPECT_DEPRECATION_WARNING(queue.WriteTexture(&imageCopyDstTexture, data.data(), dataSize,
+                                                      &textureDataLayout, &extent3D));
+    }
+    {
+        // Warning: use deprecated depth
+        // Error: depth > 1 for 2D textures
+        // This is to verify the deprecated depth is picked up by the implementation.
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.depth = 2;
+        wgpu::Queue queue = device.GetQueue();
+        ASSERT_DEVICE_ERROR(EXPECT_DEPRECATION_WARNING(queue.WriteTexture(
+            &imageCopyDstTexture, data.data(), dataSize, &textureDataLayout, &extent3D)));
+    }
+    {
+        // Error: use both deprecated depth and depthOrArrayLayers
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depth = 0;
+        extent3D.depthOrArrayLayers = 0;
+        wgpu::Queue queue = device.GetQueue();
+        ASSERT_DEVICE_ERROR(queue.WriteTexture(&imageCopyDstTexture, data.data(), dataSize,
+                                               &textureDataLayout, &extent3D));
+    }
+    {
+        // Valid: use updated depthOrArrayLayers
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.width = 0;
+        extent3D.height = 0;
+        extent3D.depthOrArrayLayers = 0;
+        wgpu::Queue queue = device.GetQueue();
+        queue.WriteTexture(&imageCopyDstTexture, data.data(), dataSize, &textureDataLayout,
+                           &extent3D);
+    }
+    {
+        // Error: use updated depthOrArrayLayers and depthOrArrayLayers > 1 for 2D textures
+        wgpu::Extent3D extent3D = kBaseExtent3D;
+        extent3D.depthOrArrayLayers = 2;
+        wgpu::Queue queue = device.GetQueue();
+        ASSERT_DEVICE_ERROR(queue.WriteTexture(&imageCopyDstTexture, data.data(), dataSize,
+                                               &textureDataLayout, &extent3D));
+    }
+}
+
 DAWN_INSTANTIATE_TEST(DeprecationTests,
                       D3D12Backend(),
                       MetalBackend(),

diff --git a/src/tests/end2end/DepthStencilStateTests.cpp b/src/tests/end2end/DepthStencilStateTests.cpp
index 3696c19..460f77b 100644
--- a/src/tests/end2end/DepthStencilStateTests.cpp
+++ b/src/tests/end2end/DepthStencilStateTests.cpp

@@ -29,7 +29,7 @@
         renderTargetDescriptor.dimension = wgpu::TextureDimension::e2D;
         renderTargetDescriptor.size.width = kRTSize;
         renderTargetDescriptor.size.height = kRTSize;
-        renderTargetDescriptor.size.depth = 1;
+        renderTargetDescriptor.size.depthOrArrayLayers = 1;
         renderTargetDescriptor.sampleCount = 1;
         renderTargetDescriptor.format = wgpu::TextureFormat::RGBA8Unorm;
         renderTargetDescriptor.mipLevelCount = 1;
@@ -43,7 +43,7 @@
         depthDescriptor.dimension = wgpu::TextureDimension::e2D;
         depthDescriptor.size.width = kRTSize;
         depthDescriptor.size.height = kRTSize;
-        depthDescriptor.size.depth = 1;
+        depthDescriptor.size.depthOrArrayLayers = 1;
         depthDescriptor.sampleCount = 1;
         depthDescriptor.format = wgpu::TextureFormat::Depth24PlusStencil8;
         depthDescriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/DeviceLostTests.cpp b/src/tests/end2end/DeviceLostTests.cpp
index 3e2752f..0ddac92 100644
--- a/src/tests/end2end/DeviceLostTests.cpp
+++ b/src/tests/end2end/DeviceLostTests.cpp

@@ -234,7 +234,7 @@
     wgpu::TextureDescriptor descriptor;
     descriptor.size.width = 4;
     descriptor.size.height = 4;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.mipLevelCount = 1;
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.usage = wgpu::TextureUsage::RenderAttachment;

diff --git a/src/tests/end2end/IOSurfaceWrappingTests.cpp b/src/tests/end2end/IOSurfaceWrappingTests.cpp
index d26fb8e..10ebad2 100644
--- a/src/tests/end2end/IOSurfaceWrappingTests.cpp
+++ b/src/tests/end2end/IOSurfaceWrappingTests.cpp

@@ -182,7 +182,7 @@
 // Test an error occurs if the descriptor depth isn't 1
 TEST_P(IOSurfaceValidationTests, InvalidDepth) {
     DAWN_SKIP_TEST_IF(UsesWire());
-    descriptor.size.depth = 2;
+    descriptor.size.depthOrArrayLayers = 2;
 
     ASSERT_DEVICE_ERROR(wgpu::Texture texture =
                             WrapIOSurface(&descriptor, defaultIOSurface.get(), 0));

diff --git a/src/tests/end2end/MultisampledRenderingTests.cpp b/src/tests/end2end/MultisampledRenderingTests.cpp
index 62e0ae8..2eeadbb 100644
--- a/src/tests/end2end/MultisampledRenderingTests.cpp
+++ b/src/tests/end2end/MultisampledRenderingTests.cpp

@@ -101,7 +101,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = kWidth << (mipLevelCount - 1);
         descriptor.size.height = kHeight << (mipLevelCount - 1);
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = sampleCount;
         descriptor.format = format;
         descriptor.mipLevelCount = mipLevelCount;

diff --git a/src/tests/end2end/NonzeroTextureCreationTests.cpp b/src/tests/end2end/NonzeroTextureCreationTests.cpp
index 91fab10..e479fee 100644
--- a/src/tests/end2end/NonzeroTextureCreationTests.cpp
+++ b/src/tests/end2end/NonzeroTextureCreationTests.cpp

@@ -33,7 +33,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = 1;
@@ -53,7 +53,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.mipLevelCount = 1;
     descriptor.usage = wgpu::TextureUsage::RenderAttachment | wgpu::TextureUsage::CopySrc;
@@ -75,7 +75,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = mipLevels;
@@ -99,7 +99,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = arrayLayers;
+    descriptor.size.depthOrArrayLayers = arrayLayers;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = 1;
@@ -125,7 +125,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Snorm;
     descriptor.mipLevelCount = 1;
@@ -163,7 +163,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kSize;
     descriptor.size.height = kSize;
-    descriptor.size.depth = 2;
+    descriptor.size.depthOrArrayLayers = 2;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Snorm;
     descriptor.mipLevelCount = 1;
@@ -196,7 +196,7 @@
     baseDescriptor.dimension = wgpu::TextureDimension::e2D;
     baseDescriptor.size.width = kSize;
     baseDescriptor.size.height = kSize;
-    baseDescriptor.size.depth = 1;
+    baseDescriptor.size.depthOrArrayLayers = 1;
     baseDescriptor.sampleCount = 1;
     baseDescriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     baseDescriptor.mipLevelCount = 1;
@@ -208,10 +208,10 @@
         wgpu::TextureDescriptor descriptor = baseDescriptor;
         // Some textures may be cleared with render pass load/store ops.
         // Test above the max attachment count.
-        descriptor.size.depth = kMaxColorAttachments + 1;
+        descriptor.size.depthOrArrayLayers = kMaxColorAttachments + 1;
         wgpu::Texture texture = device.CreateTexture(&descriptor);
 
-        for (uint32_t i = 0; i < descriptor.size.depth; ++i) {
+        for (uint32_t i = 0; i < descriptor.size.depthOrArrayLayers; ++i) {
             EXPECT_TEXTURE_RGBA8_EQ(&filled, texture, 0, 0, 1, 1, 0, i);
         }
     }
@@ -230,11 +230,11 @@
         wgpu::TextureDescriptor descriptor = baseDescriptor;
         // Some textures may be cleared with render pass load/store ops.
         // Test above the max attachment count.
-        descriptor.size.depth = kMaxColorAttachments + 1;
+        descriptor.size.depthOrArrayLayers = kMaxColorAttachments + 1;
         descriptor.mipLevelCount = 3;
         wgpu::Texture texture = device.CreateTexture(&descriptor);
 
-        for (uint32_t i = 0; i < descriptor.size.depth; ++i) {
+        for (uint32_t i = 0; i < descriptor.size.depthOrArrayLayers; ++i) {
             for (uint32_t j = 0; j < descriptor.mipLevelCount; ++j) {
                 EXPECT_TEXTURE_RGBA8_EQ(&filled, texture, 0, 0, 1, 1, j, i);
             }
@@ -248,7 +248,7 @@
     baseDescriptor.dimension = wgpu::TextureDimension::e2D;
     baseDescriptor.size.width = kSize;
     baseDescriptor.size.height = kSize;
-    baseDescriptor.size.depth = 1;
+    baseDescriptor.size.depthOrArrayLayers = 1;
     baseDescriptor.sampleCount = 1;
     baseDescriptor.format = wgpu::TextureFormat::RGBA8Snorm;
     baseDescriptor.mipLevelCount = 1;
@@ -260,10 +260,10 @@
         wgpu::TextureDescriptor descriptor = baseDescriptor;
         // Some textures may be cleared with render pass load/store ops.
         // Test above the max attachment count.
-        descriptor.size.depth = kMaxColorAttachments + 1;
+        descriptor.size.depthOrArrayLayers = kMaxColorAttachments + 1;
         wgpu::Texture texture = device.CreateTexture(&descriptor);
 
-        for (uint32_t i = 0; i < descriptor.size.depth; ++i) {
+        for (uint32_t i = 0; i < descriptor.size.depthOrArrayLayers; ++i) {
             EXPECT_TEXTURE_RGBA8_EQ(&filled, texture, 0, 0, 1, 1, 0, i);
         }
     }
@@ -282,11 +282,11 @@
         wgpu::TextureDescriptor descriptor = baseDescriptor;
         // Some textures may be cleared with render pass load/store ops.
         // Test above the max attachment count.
-        descriptor.size.depth = kMaxColorAttachments + 1;
+        descriptor.size.depthOrArrayLayers = kMaxColorAttachments + 1;
         descriptor.mipLevelCount = 3;
         wgpu::Texture texture = device.CreateTexture(&descriptor);
 
-        for (uint32_t i = 0; i < descriptor.size.depth; ++i) {
+        for (uint32_t i = 0; i < descriptor.size.depthOrArrayLayers; ++i) {
             for (uint32_t j = 0; j < descriptor.mipLevelCount; ++j) {
                 EXPECT_TEXTURE_RGBA8_EQ(&filled, texture, 0, 0, 1, 1, j, i);
             }

diff --git a/src/tests/end2end/QueueTests.cpp b/src/tests/end2end/QueueTests.cpp
index bd8fd9d..4f635e9 100644
--- a/src/tests/end2end/QueueTests.cpp
+++ b/src/tests/end2end/QueueTests.cpp

@@ -286,7 +286,7 @@
         const uint32_t bytesPerTexel = utils::GetTexelBlockSizeInBytes(kTextureFormat);
         wgpu::Extent3D mipSize = {textureSpec.textureSize.width >> textureSpec.level,
                                   textureSpec.textureSize.height >> textureSpec.level,
-                                  textureSpec.textureSize.depth};
+                                  textureSpec.textureSize.depthOrArrayLayers};
         uint32_t bytesPerRow = dataSpec.bytesPerRow;
         if (bytesPerRow == wgpu::kCopyStrideUndefined) {
             bytesPerRow = mipSize.width * bytesPerTexel;
@@ -296,7 +296,7 @@
             dataSpec.rowsPerImage > 0 ? dataSpec.rowsPerImage : mipSize.height;
         uint32_t bytesPerImage = bytesPerRow * appliedRowsPerImage;
 
-        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depth;
+        const uint32_t maxArrayLayer = textureSpec.copyOrigin.z + copySize.depthOrArrayLayers;
 
         uint64_t dataOffset = dataSpec.offset;
         const uint32_t texelCountLastLayer =

diff --git a/src/tests/end2end/RenderPassLoadOpTests.cpp b/src/tests/end2end/RenderPassLoadOpTests.cpp
index c8bc7ce..25d9427 100644
--- a/src/tests/end2end/RenderPassLoadOpTests.cpp
+++ b/src/tests/end2end/RenderPassLoadOpTests.cpp

@@ -60,7 +60,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = kRTSize;
         descriptor.size.height = kRTSize;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
         descriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/RenderPassTests.cpp b/src/tests/end2end/RenderPassTests.cpp
index 55eb5cf..5d65ae8 100644
--- a/src/tests/end2end/RenderPassTests.cpp
+++ b/src/tests/end2end/RenderPassTests.cpp

@@ -59,7 +59,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = kRTSize;
         descriptor.size.height = kRTSize;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = kFormat;
         descriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/SamplerFilterAnisotropicTests.cpp b/src/tests/end2end/SamplerFilterAnisotropicTests.cpp
index 7dd08de..0f5b080 100644
--- a/src/tests/end2end/SamplerFilterAnisotropicTests.cpp
+++ b/src/tests/end2end/SamplerFilterAnisotropicTests.cpp

@@ -97,7 +97,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = textureWidthLevel0;
         descriptor.size.height = textureHeightLevel0;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
         descriptor.mipLevelCount = mipLevelCount;

diff --git a/src/tests/end2end/SamplerTests.cpp b/src/tests/end2end/SamplerTests.cpp
index 9855154..c3ce811 100644
--- a/src/tests/end2end/SamplerTests.cpp
+++ b/src/tests/end2end/SamplerTests.cpp

@@ -93,7 +93,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = 2;
         descriptor.size.height = 2;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
         descriptor.mipLevelCount = 1;

diff --git a/src/tests/end2end/SubresourceRenderAttachmentTests.cpp b/src/tests/end2end/SubresourceRenderAttachmentTests.cpp
index 010376c..fdcd891 100644
--- a/src/tests/end2end/SubresourceRenderAttachmentTests.cpp
+++ b/src/tests/end2end/SubresourceRenderAttachmentTests.cpp

@@ -121,7 +121,7 @@
         renderTargetDesc.dimension = wgpu::TextureDimension::e2D;
         renderTargetDesc.size.width = kTextureSize;
         renderTargetDesc.size.height = kTextureSize;
-        renderTargetDesc.size.depth = kArrayLayerCount;
+        renderTargetDesc.size.depthOrArrayLayers = kArrayLayerCount;
         renderTargetDesc.sampleCount = 1;
         renderTargetDesc.format = format;
         renderTargetDesc.mipLevelCount = kMipLevelCount;

diff --git a/src/tests/end2end/TextureViewTests.cpp b/src/tests/end2end/TextureViewTests.cpp
index 1396d6e..4a34bd6 100644
--- a/src/tests/end2end/TextureViewTests.cpp
+++ b/src/tests/end2end/TextureViewTests.cpp

@@ -37,7 +37,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = 1;
         descriptor.format = kDefaultFormat;
         descriptor.mipLevelCount = mipLevelCount;

diff --git a/src/tests/end2end/TextureZeroInitTests.cpp b/src/tests/end2end/TextureZeroInitTests.cpp
index 8076c13..11ab992 100644
--- a/src/tests/end2end/TextureZeroInitTests.cpp
+++ b/src/tests/end2end/TextureZeroInitTests.cpp

@@ -45,7 +45,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = kSize;
         descriptor.size.height = kSize;
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = 1;
         descriptor.format = format;
         descriptor.mipLevelCount = mipLevelCount;
@@ -1425,7 +1425,7 @@
     wgpu::TextureDescriptor descriptor;
     descriptor.size.width = kUnalignedSize;
     descriptor.size.height = kUnalignedSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.format = wgpu::TextureFormat::R8Snorm;
     descriptor.usage = wgpu::TextureUsage::CopySrc;
     wgpu::Texture texture = device.CreateTexture(&descriptor);

diff --git a/src/tests/perf_tests/DrawCallPerf.cpp b/src/tests/perf_tests/DrawCallPerf.cpp
index a2cd14e..8000549 100644
--- a/src/tests/perf_tests/DrawCallPerf.cpp
+++ b/src/tests/perf_tests/DrawCallPerf.cpp

@@ -287,7 +287,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = kTextureSize;
         descriptor.size.height = kTextureSize;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.usage = wgpu::TextureUsage::RenderAttachment;
 
         descriptor.format = wgpu::TextureFormat::RGBA8Unorm;

diff --git a/src/tests/perf_tests/SubresourceTrackingPerf.cpp b/src/tests/perf_tests/SubresourceTrackingPerf.cpp
index 7f32843..0ddf864 100644
--- a/src/tests/perf_tests/SubresourceTrackingPerf.cpp
+++ b/src/tests/perf_tests/SubresourceTrackingPerf.cpp

@@ -64,7 +64,7 @@
         mMaterials = device.CreateTexture(&materialDesc);
 
         wgpu::TextureDescriptor uploadTexDesc = materialDesc;
-        uploadTexDesc.size.depth = 1;
+        uploadTexDesc.size.depthOrArrayLayers = 1;
         uploadTexDesc.mipLevelCount = 1;
         uploadTexDesc.usage = wgpu::TextureUsage::CopySrc;
         mUploadTexture = device.CreateTexture(&uploadTexDesc);

diff --git a/src/tests/unittests/d3d12/CopySplitTests.cpp b/src/tests/unittests/d3d12/CopySplitTests.cpp
index adc546d..e18b5be 100644
--- a/src/tests/unittests/d3d12/CopySplitTests.cpp
+++ b/src/tests/unittests/d3d12/CopySplitTests.cpp

@@ -49,7 +49,8 @@
             const auto& copy = copySplit.copies[i];
             ASSERT_LE(copy.bufferOffset.x + copy.copySize.width, copy.bufferSize.width);
             ASSERT_LE(copy.bufferOffset.y + copy.copySize.height, copy.bufferSize.height);
-            ASSERT_LE(copy.bufferOffset.z + copy.copySize.depth, copy.bufferSize.depth);
+            ASSERT_LE(copy.bufferOffset.z + copy.copySize.depthOrArrayLayers,
+                      copy.bufferSize.depthOrArrayLayers);
         }
     }
 
@@ -75,9 +76,9 @@
                 bool overlapY =
                     RangesOverlap(a.textureOffset.y, a.textureOffset.y + a.copySize.height,
                                   b.textureOffset.y, b.textureOffset.y + b.copySize.height);
-                bool overlapZ =
-                    RangesOverlap(a.textureOffset.z, a.textureOffset.z + a.copySize.depth,
-                                  b.textureOffset.z, b.textureOffset.z + b.copySize.depth);
+                bool overlapZ = RangesOverlap(
+                    a.textureOffset.z, a.textureOffset.z + a.copySize.depthOrArrayLayers,
+                    b.textureOffset.z, b.textureOffset.z + b.copySize.depthOrArrayLayers);
                 ASSERT_TRUE(!overlapX || !overlapY || !overlapZ);
             }
         }
@@ -93,7 +94,8 @@
         uint32_t minZ = copySplit.copies[0].textureOffset.z;
         uint32_t maxX = copySplit.copies[0].textureOffset.x + copySplit.copies[0].copySize.width;
         uint32_t maxY = copySplit.copies[0].textureOffset.y + copySplit.copies[0].copySize.height;
-        uint32_t maxZ = copySplit.copies[0].textureOffset.z + copySplit.copies[0].copySize.depth;
+        uint32_t maxZ =
+            copySplit.copies[0].textureOffset.z + copySplit.copies[0].copySize.depthOrArrayLayers;
 
         for (uint32_t i = 1; i < copySplit.count; ++i) {
             const auto& copy = copySplit.copies[i];
@@ -102,7 +104,7 @@
             minZ = std::min(minZ, copy.textureOffset.z);
             maxX = std::max(maxX, copy.textureOffset.x + copy.copySize.width);
             maxY = std::max(maxY, copy.textureOffset.y + copy.copySize.height);
-            maxZ = std::max(maxZ, copy.textureOffset.z + copy.copySize.depth);
+            maxZ = std::max(maxZ, copy.textureOffset.z + copy.copySize.depthOrArrayLayers);
         }
 
         ASSERT_EQ(minX, textureSpec.x);
@@ -119,7 +121,7 @@
         uint32_t count = 0;
         for (uint32_t i = 0; i < copySplit.count; ++i) {
             const auto& copy = copySplit.copies[i];
-            count += copy.copySize.width * copy.copySize.height * copy.copySize.depth;
+            count += copy.copySize.width * copy.copySize.height * copy.copySize.depthOrArrayLayers;
         }
         ASSERT_EQ(count, textureSpec.width * textureSpec.height * textureSpec.depth);
     }
@@ -192,11 +194,12 @@
             const auto& copy = copySplit.copies[i];
             os << "  " << i << ": Texture at (" << copy.textureOffset.x << ", "
                << copy.textureOffset.y << ", " << copy.textureOffset.z << "), size ("
-               << copy.copySize.width << ", " << copy.copySize.height << ", " << copy.copySize.depth
-               << ")" << std::endl;
+               << copy.copySize.width << ", " << copy.copySize.height << ", "
+               << copy.copySize.depthOrArrayLayers << ")" << std::endl;
             os << "  " << i << ": Buffer at (" << copy.bufferOffset.x << ", " << copy.bufferOffset.y
                << ", " << copy.bufferOffset.z << "), footprint (" << copy.bufferSize.width << ", "
-               << copy.bufferSize.height << ", " << copy.bufferSize.depth << ")" << std::endl;
+               << copy.bufferSize.height << ", " << copy.bufferSize.depthOrArrayLayers << ")"
+               << std::endl;
         }
         return os;
     }

diff --git a/src/tests/unittests/validation/CopyCommandsValidationTests.cpp b/src/tests/unittests/validation/CopyCommandsValidationTests.cpp
index 62e4c9c..29b664a 100644
--- a/src/tests/unittests/validation/CopyCommandsValidationTests.cpp
+++ b/src/tests/unittests/validation/CopyCommandsValidationTests.cpp

@@ -41,7 +41,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = sampleCount;
         descriptor.format = format;
         descriptor.mipLevelCount = mipLevelCount;
@@ -663,7 +663,7 @@
     ASSERT_DEVICE_ERROR(wgpu::Buffer errorBuffer = device.CreateBuffer(&errorBufferDescriptor));
 
     wgpu::TextureDescriptor errorTextureDescriptor;
-    errorTextureDescriptor.size.depth = 0;
+    errorTextureDescriptor.size.depthOrArrayLayers = 0;
     ASSERT_DEVICE_ERROR(wgpu::Texture errorTexture = device.CreateTexture(&errorTextureDescriptor));
 
     wgpu::ImageCopyBuffer errorImageCopyBuffer = utils::CreateImageCopyBuffer(errorBuffer, 0, 0, 0);
@@ -1252,7 +1252,7 @@
     ASSERT_DEVICE_ERROR(wgpu::Buffer errorBuffer = device.CreateBuffer(&errorBufferDescriptor));
 
     wgpu::TextureDescriptor errorTextureDescriptor;
-    errorTextureDescriptor.size.depth = 0;
+    errorTextureDescriptor.size.depthOrArrayLayers = 0;
     ASSERT_DEVICE_ERROR(wgpu::Texture errorTexture = device.CreateTexture(&errorTextureDescriptor));
 
     wgpu::ImageCopyBuffer errorImageCopyBuffer = utils::CreateImageCopyBuffer(errorBuffer, 0, 0, 0);
@@ -1533,12 +1533,12 @@
         TestT2TCopy(utils::Expectation::Success, source, 0, {0, 0, 1}, destination, 0, {0, 0, 1},
                     {16, 16, 1});
 
-        // Copy multiple slices (srcImageCopyTexture.arrayLayer + copySize.depth ==
+        // Copy multiple slices (srcImageCopyTexture.arrayLayer + copySize.depthOrArrayLayers ==
         // srcImageCopyTexture.texture.arrayLayerCount)
         TestT2TCopy(utils::Expectation::Success, source, 0, {0, 0, 2}, destination, 0, {0, 0, 0},
                     {16, 16, 2});
 
-        // Copy multiple slices (dstImageCopyTexture.arrayLayer + copySize.depth ==
+        // Copy multiple slices (dstImageCopyTexture.arrayLayer + copySize.depthOrArrayLayers ==
         // dstImageCopyTexture.texture.arrayLayerCount)
         TestT2TCopy(utils::Expectation::Success, source, 0, {0, 0, 0}, destination, 0, {0, 0, 2},
                     {16, 16, 2});

diff --git a/src/tests/unittests/validation/QueueWriteTextureValidationTests.cpp b/src/tests/unittests/validation/QueueWriteTextureValidationTests.cpp
index f115c30c..6d8a2e7 100644
--- a/src/tests/unittests/validation/QueueWriteTextureValidationTests.cpp
+++ b/src/tests/unittests/validation/QueueWriteTextureValidationTests.cpp

@@ -38,7 +38,7 @@
             descriptor.dimension = wgpu::TextureDimension::e2D;
             descriptor.size.width = size.width;
             descriptor.size.height = size.height;
-            descriptor.size.depth = size.depth;
+            descriptor.size.depthOrArrayLayers = size.depthOrArrayLayers;
             descriptor.sampleCount = sampleCount;
             descriptor.format = format;
             descriptor.mipLevelCount = mipLevelCount;
@@ -358,7 +358,7 @@
     // Test WriteTexture with texture in error state causes errors.
     TEST_F(QueueWriteTextureValidationTest, TextureInErrorState) {
         wgpu::TextureDescriptor errorTextureDescriptor;
-        errorTextureDescriptor.size.depth = 0;
+        errorTextureDescriptor.size.depthOrArrayLayers = 0;
         ASSERT_DEVICE_ERROR(wgpu::Texture errorTexture =
                                 device.CreateTexture(&errorTextureDescriptor));
         wgpu::ImageCopyTexture errorImageCopyTexture =

diff --git a/src/tests/unittests/validation/RenderPassDescriptorValidationTests.cpp b/src/tests/unittests/validation/RenderPassDescriptorValidationTests.cpp
index 9de9bdc..b02ec28 100644
--- a/src/tests/unittests/validation/RenderPassDescriptorValidationTests.cpp
+++ b/src/tests/unittests/validation/RenderPassDescriptorValidationTests.cpp

@@ -55,7 +55,7 @@
         descriptor.dimension = dimension;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = sampleCount;
         descriptor.format = format;
         descriptor.mipLevelCount = mipLevelCount;

diff --git a/src/tests/unittests/validation/RenderPipelineValidationTests.cpp b/src/tests/unittests/validation/RenderPipelineValidationTests.cpp
index 93175ea..1a9aaac 100644
--- a/src/tests/unittests/validation/RenderPipelineValidationTests.cpp
+++ b/src/tests/unittests/validation/RenderPipelineValidationTests.cpp

@@ -238,7 +238,7 @@
     wgpu::TextureDescriptor baseTextureDescriptor;
     baseTextureDescriptor.size.width = 4;
     baseTextureDescriptor.size.height = 4;
-    baseTextureDescriptor.size.depth = 1;
+    baseTextureDescriptor.size.depthOrArrayLayers = 1;
     baseTextureDescriptor.mipLevelCount = 1;
     baseTextureDescriptor.dimension = wgpu::TextureDimension::e2D;
     baseTextureDescriptor.usage = wgpu::TextureUsage::RenderAttachment;

diff --git a/src/tests/unittests/validation/TextureValidationTests.cpp b/src/tests/unittests/validation/TextureValidationTests.cpp
index 7ff68f0..48da264 100644
--- a/src/tests/unittests/validation/TextureValidationTests.cpp
+++ b/src/tests/unittests/validation/TextureValidationTests.cpp

@@ -33,7 +33,7 @@
             wgpu::TextureDescriptor descriptor;
             descriptor.size.width = kWidth;
             descriptor.size.height = kHeight;
-            descriptor.size.depth = kDefaultDepth;
+            descriptor.size.depthOrArrayLayers = kDefaultDepth;
             descriptor.mipLevelCount = kDefaultMipLevels;
             descriptor.sampleCount = kDefaultSampleCount;
             descriptor.dimension = wgpu::TextureDimension::e2D;
@@ -109,7 +109,7 @@
         {
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
             descriptor.sampleCount = 4;
-            descriptor.size.depth = 2;
+            descriptor.size.depthOrArrayLayers = 2;
 
             ASSERT_DEVICE_ERROR(device.CreateTexture(&descriptor));
         }
@@ -217,7 +217,7 @@
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
             descriptor.size.width = 32;
             descriptor.size.height = 8;
-            descriptor.size.depth = 64;
+            descriptor.size.depthOrArrayLayers = 64;
             descriptor.dimension = wgpu::TextureDimension::e3D;
             // Non square mip map halves width, height and depth until a 1x1x1 dimension for a 3D
             // texture. So there are 7 mipmaps at most: 32 * 8 * 64, 16 * 4 * 32, 8 * 2 * 16,
@@ -231,7 +231,7 @@
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
             descriptor.size.width = 32;
             descriptor.size.height = 8;
-            descriptor.size.depth = 64;
+            descriptor.size.depthOrArrayLayers = 64;
             // Non square mip map halves width and height until a 1x1 dimension for a 2D texture,
             // even its depth > 1. So there are 6 mipmaps at most: 32 * 8, 16 * 4, 8 * 2, 4 * 1, 2 *
             // 1, 1 * 1.
@@ -261,21 +261,21 @@
         {
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
 
-            descriptor.size.depth = kMaxTextureArrayLayers + 1u;
+            descriptor.size.depthOrArrayLayers = kMaxTextureArrayLayers + 1u;
             ASSERT_DEVICE_ERROR(device.CreateTexture(&descriptor));
         }
 
         // Array layer count less than kMaxTextureArrayLayers is allowed
         {
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
-            descriptor.size.depth = kMaxTextureArrayLayers >> 1;
+            descriptor.size.depthOrArrayLayers = kMaxTextureArrayLayers >> 1;
             device.CreateTexture(&descriptor);
         }
 
         // Array layer count equal to kMaxTextureArrayLayers is allowed
         {
             wgpu::TextureDescriptor descriptor = defaultDescriptor;
-            descriptor.size.depth = kMaxTextureArrayLayers;
+            descriptor.size.depthOrArrayLayers = kMaxTextureArrayLayers;
             device.CreateTexture(&descriptor);
         }
     }
@@ -602,7 +602,7 @@
         for (wgpu::TextureFormat format : utils::kBCFormats) {
             wgpu::TextureDescriptor descriptor = CreateDefaultTextureDescriptor();
             descriptor.format = format;
-            descriptor.size.depth = 6;
+            descriptor.size.depthOrArrayLayers = 6;
             device.CreateTexture(&descriptor);
         }
     }
@@ -612,7 +612,7 @@
         for (wgpu::TextureFormat format : utils::kBCFormats) {
             wgpu::TextureDescriptor descriptor = CreateDefaultTextureDescriptor();
             descriptor.format = format;
-            descriptor.size.depth = 4;
+            descriptor.size.depthOrArrayLayers = 4;
             descriptor.dimension = wgpu::TextureDimension::e3D;
             ASSERT_DEVICE_ERROR(device.CreateTexture(&descriptor));
         }

diff --git a/src/tests/unittests/validation/TextureViewValidationTests.cpp b/src/tests/unittests/validation/TextureViewValidationTests.cpp
index e2f5476..31ad3c1 100644
--- a/src/tests/unittests/validation/TextureViewValidationTests.cpp
+++ b/src/tests/unittests/validation/TextureViewValidationTests.cpp

@@ -35,7 +35,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = arrayLayerCount;
+        descriptor.size.depthOrArrayLayers = arrayLayerCount;
         descriptor.sampleCount = sampleCount;
         descriptor.format = kDefaultTextureFormat;
         descriptor.mipLevelCount = mipLevelCount;

diff --git a/src/tests/unittests/validation/ValidationTest.cpp b/src/tests/unittests/validation/ValidationTest.cpp
index 27b3c9c..d4a5ba9 100644
--- a/src/tests/unittests/validation/ValidationTest.cpp
+++ b/src/tests/unittests/validation/ValidationTest.cpp

@@ -187,7 +187,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = width;
     descriptor.size.height = height;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = attachmentFormat;
     descriptor.mipLevelCount = 1;

diff --git a/src/tests/white_box/D3D12DescriptorHeapTests.cpp b/src/tests/white_box/D3D12DescriptorHeapTests.cpp
index 41985b1..cbbe21d 100644
--- a/src/tests/white_box/D3D12DescriptorHeapTests.cpp
+++ b/src/tests/white_box/D3D12DescriptorHeapTests.cpp

@@ -73,7 +73,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = format;
         descriptor.mipLevelCount = 1;
@@ -747,7 +747,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = kRTSize;
     descriptor.size.height = kRTSize;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::RGBA8Unorm;
     descriptor.mipLevelCount = 1;

diff --git a/src/tests/white_box/D3D12ResourceHeapTests.cpp b/src/tests/white_box/D3D12ResourceHeapTests.cpp
index 450cb9e..b9606ce 100644
--- a/src/tests/white_box/D3D12ResourceHeapTests.cpp
+++ b/src/tests/white_box/D3D12ResourceHeapTests.cpp

@@ -54,7 +54,7 @@
     descriptor.dimension = wgpu::TextureDimension::e2D;
     descriptor.size.width = 8;
     descriptor.size.height = 8;
-    descriptor.size.depth = 1;
+    descriptor.size.depthOrArrayLayers = 1;
     descriptor.sampleCount = 1;
     descriptor.format = wgpu::TextureFormat::BC1RGBAUnorm;
     descriptor.mipLevelCount = 1;

diff --git a/src/tests/white_box/VulkanImageWrappingTestsDmaBuf.cpp b/src/tests/white_box/VulkanImageWrappingTestsDmaBuf.cpp
index 1ab886b..c5b36e6 100644
--- a/src/tests/white_box/VulkanImageWrappingTestsDmaBuf.cpp
+++ b/src/tests/white_box/VulkanImageWrappingTestsDmaBuf.cpp

@@ -225,7 +225,7 @@
 
     // Test an error occurs if the descriptor depth isn't 1
     TEST_P(VulkanImageWrappingValidationTests, InvalidDepth) {
-        defaultDescriptor.size.depth = 2;
+        defaultDescriptor.size.depthOrArrayLayers = 2;
 
         ASSERT_DEVICE_ERROR(wgpu::Texture texture =
                                 WrapVulkanImage(device, &defaultDescriptor, defaultFd,
@@ -770,7 +770,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = 640;
         descriptor.size.height = 480;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = wgpu::TextureFormat::BGRA8Unorm;
         descriptor.mipLevelCount = 1;

diff --git a/src/tests/white_box/VulkanImageWrappingTestsOpaqueFD.cpp b/src/tests/white_box/VulkanImageWrappingTestsOpaqueFD.cpp
index 7df34ab..9e16567 100644
--- a/src/tests/white_box/VulkanImageWrappingTestsOpaqueFD.cpp
+++ b/src/tests/white_box/VulkanImageWrappingTestsOpaqueFD.cpp

@@ -314,7 +314,7 @@
     // Test an error occurs if the descriptor depth isn't 1
     TEST_P(VulkanImageWrappingValidationTests, InvalidDepth) {
         DAWN_SKIP_TEST_IF(UsesWire());
-        defaultDescriptor.size.depth = 2;
+        defaultDescriptor.size.depthOrArrayLayers = 2;
 
         ASSERT_DEVICE_ERROR(wgpu::Texture texture = WrapVulkanImage(
                                 device, &defaultDescriptor, defaultFd, defaultAllocationSize,
@@ -929,7 +929,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = 640;
         descriptor.size.height = 480;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = wgpu::TextureFormat::BGRA8Unorm;
         descriptor.mipLevelCount = 1;

diff --git a/src/utils/TestUtils.cpp b/src/utils/TestUtils.cpp
index 323c3c2..673ad78 100644
--- a/src/utils/TestUtils.cpp
+++ b/src/utils/TestUtils.cpp

@@ -40,7 +40,8 @@
         TextureDataCopyLayout layout;
 
         layout.mipSize = {textureSizeAtLevel0.width >> mipmapLevel,
-                          textureSizeAtLevel0.height >> mipmapLevel, textureSizeAtLevel0.depth};
+                          textureSizeAtLevel0.height >> mipmapLevel,
+                          textureSizeAtLevel0.depthOrArrayLayers};
 
         layout.bytesPerRow = GetMinimumBytesPerRow(format, layout.mipSize.width);
 
@@ -83,7 +84,7 @@
         ASSERT(copyExtent.height % blockHeight == 0);
         uint32_t heightInBlocks = copyExtent.height / blockHeight;
         return RequiredBytesInCopy(bytesPerRow, rowsPerImage, widthInBlocks, heightInBlocks,
-                                   copyExtent.depth, blockSize);
+                                   copyExtent.depthOrArrayLayers, blockSize);
     }
 
     uint64_t RequiredBytesInCopy(uint64_t bytesPerRow,

diff --git a/src/utils/WGPUHelpers.cpp b/src/utils/WGPUHelpers.cpp
index 8370c91..751c4e5 100644
--- a/src/utils/WGPUHelpers.cpp
+++ b/src/utils/WGPUHelpers.cpp

@@ -258,7 +258,7 @@
         descriptor.dimension = wgpu::TextureDimension::e2D;
         descriptor.size.width = width;
         descriptor.size.height = height;
-        descriptor.size.depth = 1;
+        descriptor.size.depthOrArrayLayers = 1;
         descriptor.sampleCount = 1;
         descriptor.format = BasicRenderPass::kDefaultColorFormat;
         descriptor.mipLevelCount = 1;