[memory-leak] Fix CoTaskMem pixel-buffer leak in SKImage.Create raster failure path by github-actions[bot] · Pull Request #4455 · mono/SkiaSharp

github-actions · 2026-07-16T12:58:30Z

🤖 Produced automatically by the Fixer - Memory Leak agentic workflow using the memory-leak-fixer skill (focus area 10 — allocation-failure path).

The fix

SKImage.Create(SKImageInfo) allocates an unmanaged pixel buffer with Marshal.AllocCoTaskMem and relies on the native raster image's release proc (SKImageRasterReleaseProxyForCoTaskMem, which calls Marshal.FreeCoTaskMem) to free it. That proc only runs when the native image is destroyed — so when sk_image_new_raster returns null (a valid-BytesSize info that Skia rejects, e.g. Rgba8888 + AlphaType.Unknown), no image is created, the proc never fires, and the buffer leaks.

The fix frees the buffer explicitly on the null-handle path — the idiomatic focus-area-10 pattern (return null on failure, but release any allocation made along the way):

var image = GetObject(SkiaApi.sk_image_new_raster(pixmap.Handle, DelegateProxies.SKImageRasterReleaseProxyForCoTaskMem, null));
if (image == null)
    Marshal.FreeCoTaskMem(pixels);
return image;

Managed-C# only (binding/SkiaSharp/SKImage.cs), no public signature change — ABI stable. The factory still returns null on failure (no exception-surface change).

Proof (red → green)

New regression test tests/Tests/SkiaSharp/SKImageTest.cs::FailedRasterCreateDoesNotLeakPixelBuffer drives 200 failing 4 MB Create calls and asserts private memory does not grow by a large fraction of what was allocated.

Without fix: FAIL — private memory grew 805 MB over 200 failed calls (~800 MB allocated → every buffer leaked).
With fix: PASS — growth ≈ 0. Full SKImageTest class green (45 passed / 7 hardware-skipped).

Commands (single-TFM, prebuilt natives via externals-download):

dotnet build tests/SkiaSharp.Tests.Console/SkiaSharp.Tests.Console.csproj \
  -f net10.0 -p:AllTargetFrameworks=net10.0 -p:BasicTargetFrameworks=net10.0 -p:PlatformTargetFrameworks=
dotnet SkiaSharp.Tests.dll --filter-method "*FailedRasterCreateDoesNotLeakPixelBuffer*"

Scope note

Framework bug, empirically proven. Managed-C# only, ABI-safe, internal-only change. Distinct from open PR #4453 (GCHandle leaks in FromPixels/FromTexture/SKSurface.Create/SKData.Create) — that PR does not touch SKImage.Create(SKImageInfo) or the CoTaskMem pixel buffer.

Labels: tenet/performance + perf/memory-leak.

Fixes #4454

Generated by Fixer - Memory Leak · ● 16.7M · ◷

SKImage.Create(SKImageInfo) allocates an unmanaged pixel buffer with Marshal.AllocCoTaskMem and relies on the native raster image's release proc to free it. When sk_image_new_raster fails (e.g. a valid-BytesSize info Skia rejects), no image is created, the release proc never runs, and the buffer leaks. Free it explicitly on the failure path. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

…4476) Add regression-tracking benchmarks for common colour/interop paths (#4476) Derived from: #4370, #4385, #4442, #4453, #4455 CI: Track - Benchmarks run 29608575980 (green, all OS/roles) The permanent regression-tracking suite (SkiaSharp.Benchmarks.Tracking) landed recently alongside a wave of perf/agentic PRs, but it had no coverage for the managed colour/interop paths those PRs actually touch. This adds a small, curated set of trackers so a future regression (or the improvement itself) shows up as a step in the persisted per-OS time/allocation history. Following the suite's scaled-pyramid philosophy (a few small common-path trackers plus one broad composite, not a micro-benchmark per PR), four benchmarks are added under Tracking/Benchmarks/ (auto-linked into the source-mode project, so they also drive the PR column): * ColorMathBenchmark - managed SKColor->SKColorF (#4370) and SKPMColor PreMultiply/UnPreMultiply (#4385). Zero-alloc; the ToColorF/ToColor ratio (the reverse operator stays native) is a managed-regression signal even on noisy shared runners. These two ports are the real, dashboard-visible speed wins - their time lines should drop when the ports merge. * RasterImageLifecycleBenchmark - create+dispose churn of SKImage.Create raster (#4455) and SKData.Create with a managed release proc (#4453). Allocation is the tracked signal for that object-lifecycle area. * RuntimeEffectShaderBenchmark - a per-frame animated SkSL shader: build uniforms (the SKColor uniform pays the #4370 convert), ToShader, draw. Covers the uniforms lifecycle whose SKData leak #4442 touched. * SceneRenderBenchmark - one broad frame (transforms, linear+radial gradients, filled/stroked primitives, a built path, a scaled image draw, a colour-filter layer, clipping, managed colour maths) so many unrelated future optimizations each nudge it - a merge indicator rather than a micro-benchmark. Scope notes: the failure-path leak fixes (#4453, #4455), the native uniforms leak (#4442), and the UAF/double-free fixes (#4372, #4468) are correctness/native-memory changes that a throughput + managed-allocation dashboard largely will not register; they are represented here by area, not as leak assertions (those belong in soak/ leak tests). #4372 and #4468 are intentionally not benchmarked (niche debug canvases; HarfBuzz lives in a separate assembly the tracking project does not reference). All benchmarks use public API only (the project references the nightly NuGet, not internal SkiaApi), are deterministic and machine-independent (fixed seeds, no fonts or external content), and compile against every benchmarked role - verified against the 3.119.4 baseline (prev-major) and 4.151 nightly, with only SKPath/SKPathBuilder behind the existing #if. A short local run and the full CI matrix (Linux/Windows/ macOS, all version roles plus the source-built PR column) emit time and allocation data for every case. Co-authored-by: Matthew Leibowitz <mattleibow@live.com> Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

…f47978d8b90

github-actions · 2026-07-17T21:05:15Z

📦 Try the packages from this PR

Warning

Do not run these scripts without first reviewing the code in this PR.

Step 1 — Download the packages

bash / macOS / Linux:

curl -fsSL https://raw.githubusercontent.com/mono/SkiaSharp/main/scripts/get-skiasharp-pr.sh | bash -s -- 4455

PowerShell / Windows:

iex "& { $(irm https://raw.githubusercontent.com/mono/SkiaSharp/main/scripts/get-skiasharp-pr.ps1) } 4455"

Step 2 — Add the local NuGet source

dotnet nuget add source ~/.skiasharp/hives/pr-4455/packages --name skiasharp-pr-4455

More options

Option	Description
`--successful-only` / `-SuccessfulOnly`	Only use successful builds
`--force` / `-Force`	Overwrite previously downloaded packages
`--list` / `-List`	List available artifacts without downloading
`--build-id ID` / `-BuildId ID`	Download from a specific build

Or download manually from Azure Pipelines — look for the nuget artifact on the build for this PR.

Remove the source when you're done:

dotnet nuget remove source skiasharp-pr-4455

github-actions · 2026-07-17T21:35:12Z

📖 Documentation Preview

The documentation for this PR has been deployed and is available at:

🔗 View Staging Site
🔗 View Staging Docs
🔗 View Staging Gallery (Blazor)
🔗 View Staging Gallery (Uno Platform)
🔗 View Staging SkiaFiddle

This preview will be updated automatically when you push new commits to this PR.

This comment is automatically updated by the documentation staging workflow.

Copilot

Pull request overview

This PR fixes an unmanaged memory leak in SKImage.Create(SKImageInfo) when raster image creation fails: the method allocates a CoTaskMem pixel buffer, but if the native factory returns null the release-proc never runs, so the buffer must be freed explicitly. It also adds a regression test to catch the leak.

Changes:

Free the CoTaskMem pixel buffer when sk_image_new_raster returns a null image.
Add a regression test that repeatedly hits the failure path and asserts process private memory doesn’t grow significantly.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
binding/SkiaSharp/SKImage.cs	Frees the unmanaged pixel buffer on native raster-create failure to prevent leaks.
tests/Tests/SkiaSharp/SKImageTest.cs	Adds a regression test that exercises the failure path repeatedly and checks for memory growth.

The FailedRasterCreateDoesNotLeakPixelBuffer regression test measures the leak via System.Diagnostics.Process.PrivateMemorySize64. That API is not supported on WASM (no System.Diagnostics.Process at all), iOS, or Mac Catalyst, so the test threw PlatformNotSupportedException and failed the Tests WASM (Linux) and Tests iOS (macOS) legs (and Mac Catalyst). Guard the test with SkipOnPlatform(IsBrowser || IsIOS || IsMacCatalyst), matching the existing platform-skip pattern in this suite. The SKImage.Create fix itself is platform-agnostic and remains exercised by the desktop and Android test legs, where PrivateMemorySize64 is available. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 4649389c-a7bd-443d-8c26-7c6b16cda819

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

…f47978d8b90

github-actions · 2026-07-22T21:53:30Z

📊 SkiaSharp benchmarks — PR #4455

⭐ this PR (full source build) vs 🌙 nightly · Linux · Windows · macOS

Informational only — this never blocks the PR. 🟢 faster / less allocation · 🔴 slower / more allocation; moves under 5% are hidden as noise.

⏱️ Times are raw BenchmarkDotNet means, and the ⭐ PR and baseline legs run on separate CI runners, so microbenchmarks can swing run-to-run — treat small time deltas as noise. Allocations are deterministic and the reliable signal. The interactive perf-dashboard (linked below) applies smoothing for the trend view.

Highlights

⏱️ Time — 🔴 28 slower · 🟢 35 faster

🟢 RuntimeEffectShaderBenchmark.DrawFrame · Windows · 1.12 ms → 701.67 µs (-37%)
🟢 ColorMathBenchmark.ToColor(Colors: 4096) · Windows · 15.52 µs → 10.66 µs (-31%)
🟢 ColorMathBenchmark.UnPreMultiply(Colors: 4096) · Windows · 12.22 µs → 8.73 µs (-28%)
🟢 ColorMathBenchmark.ToColor(Colors: 4096) · Linux · 15.39 µs → 11.76 µs (-24%)
🟢 ColorMathBenchmark.PreMultiply(Colors: 4096) · Windows · 13.19 µs → 10.12 µs (-23%)
…and 58 more (see details below)

Full per-OS benchmark deltas

Linux

⏱️ Time (vs 🌙 nightly 4.151.0-nightly.83)

Benchmark	baseline	this PR	Δ
`ColorMathBenchmark.ToColor(Colors: 4096)`	15.39 µs	11.76 µs	🟢 -24%
`MatrixOpsBenchmark.Concat(Count: 4096)`	217.38 µs	253.76 µs	🔴 +17%
`MatrixOpsBenchmark.Invert(Count: 4096)`	131.81 µs	150.32 µs	🔴 +14%
`MatrixMapPointsBenchmark.MapPoint(Points: 4096)`	71.94 µs	81.73 µs	🔴 +14%
`MatrixMapPointsBenchmark.MapPoints(Points: 256)`	204.4 ns	232.0 ns	🔴 +13%
`ColorMathBenchmark.PreMultiply(Colors: 4096)`	12.85 µs	14.49 µs	🔴 +13%
`MatrixMapPointsBenchmark.MapPoint(Points: 256)`	4.50 µs	5.08 µs	🔴 +13%
`RasterImageLifecycleBenchmark.CreateDataWithReleaseProc(Count: 256)`	98.98 µs	111.47 µs	🔴 +13%
`MatrixMapPointsBenchmark.MapVector(Points: 4096)`	171.72 µs	193.27 µs	🔴 +13%
`ColorMathBenchmark.UnPreMultiply(Colors: 4096)`	11.45 µs	12.65 µs	🔴 +10%
`MatrixMapPointsBenchmark.MapRadius(Points: 256)`	13.96 µs	15.33 µs	🔴 +10%
`MatrixMapPointsBenchmark.MapVector(Points: 256)`	10.79 µs	11.84 µs	🔴 +10%
`MatrixMapPointsBenchmark.MapRect(Points: 256)`	11.93 µs	12.81 µs	🔴 +7%
`MatrixMapPointsBenchmark.MapRect(Points: 4096)`	190.96 µs	204.88 µs	🔴 +7%
`ColorMathBenchmark.ToColorF(Colors: 4096)`	13.13 µs	14.07 µs	🔴 +7%
`RuntimeEffectShaderBenchmark.DrawFrame`	2.83 ms	3.03 ms	🔴 +7%
`PathBoundsBenchmark.TightBounds(Points: 1024)`	10.65 µs	11.39 µs	🔴 +7%
`BitmapDrawBenchmark.DrawUnscaledTiles(Tiles: 64)`	251.27 µs	267.95 µs	🔴 +7%
`CanvasDrawBenchmark.Draw(Shapes: 64)`	4.16 ms	4.41 ms	🔴 +6%
`CanvasDrawBenchmark.Draw(Shapes: 512)`	33.24 ms	35.17 ms	🔴 +6%
`BitmapDrawBenchmark.DrawUnscaledTiles(Tiles: 256)`	940.85 µs	994.12 µs	🔴 +6%
`BitmapDrawBenchmark.DrawScaledTiles(Tiles: 64)`	355.99 µs	376.05 µs	🔴 +6%
`BitmapDrawBenchmark.DrawScaledTiles(Tiles: 256)`	1.34 ms	1.40 ms	🔴 +5%

Windows

⏱️ Time (vs 🌙 nightly 4.151.0-nightly.83)

Benchmark	baseline	this PR	Δ
`RuntimeEffectShaderBenchmark.DrawFrame`	1.12 ms	701.67 µs	🟢 -37%
`ColorMathBenchmark.ToColor(Colors: 4096)`	15.52 µs	10.66 µs	🟢 -31%
`ColorMathBenchmark.UnPreMultiply(Colors: 4096)`	12.22 µs	8.73 µs	🟢 -28%
`ColorMathBenchmark.PreMultiply(Colors: 4096)`	13.19 µs	10.12 µs	🟢 -23%
`BitmapDrawBenchmark.DrawUnscaledTiles(Tiles: 256)`	942.89 µs	734.99 µs	🟢 -22%
`MatrixMapPointsBenchmark.MapPoints(Points: 256)`	173.5 ns	208.5 ns	🔴 +20%
`MatrixMapPointsBenchmark.MapPoint(Points: 4096)`	79.32 µs	64.49 µs	🟢 -19%
`RasterImageLifecycleBenchmark.CreateDataWithReleaseProc(Count: 256)`	100.03 µs	118.46 µs	🔴 +18%
`CanvasDrawBenchmark.Draw(Shapes: 64)`	2.60 ms	2.14 ms	🟢 -18%
`MatrixOpsBenchmark.Concat(Count: 4096)`	230.64 µs	190.41 µs	🟢 -17%
`CanvasDrawBenchmark.Draw(Shapes: 512)`	20.95 ms	17.32 ms	🟢 -17%
`MatrixMapPointsBenchmark.MapPoint(Points: 256)`	4.85 µs	4.04 µs	🟢 -17%
`LargeImageScaleBenchmark.UpscaleCrossfade(Size: 2048)`	10.32 ms	8.61 ms	🟢 -17%
`BitmapDrawBenchmark.DrawUnscaledTiles(Tiles: 64)`	251.45 µs	210.19 µs	🟢 -16%
`MatrixOpsBenchmark.Invert(Count: 4096)`	132.16 µs	111.21 µs	🟢 -16%
`BitmapDrawBenchmark.DrawScaledTiles(Tiles: 256)`	1.26 ms	1.07 ms	🟢 -15%
`LargeImageScaleBenchmark.UpscaleCrossfade(Size: 1024)`	2.56 ms	2.21 ms	🟢 -14%
`PathBoundsBenchmark.TightBounds(Points: 1024)`	15.93 µs	13.81 µs	🟢 -13%
`SceneRenderBenchmark.RenderFrame(Complexity: 4)`	10.07 ms	8.73 ms	🟢 -13%
`ColorParseBenchmark.Parse(Iterations: 1000)`	77.86 µs	67.60 µs	🟢 -13%
`SceneRenderBenchmark.RenderFrame(Complexity: 1)`	2.11 ms	1.84 ms	🟢 -13%
`PathBoundsBenchmark.TightBounds(Points: 64)`	1.54 µs	1.35 µs	🟢 -12%
`BitmapDrawBenchmark.DrawScaledTiles(Tiles: 64)`	334.04 µs	304.84 µs	🟢 -9%
`MatrixMapPointsBenchmark.MapRect(Points: 4096)`	121.43 µs	111.22 µs	🟢 -8%
`ColorMathBenchmark.ToColorF(Colors: 4096)`	13.29 µs	14.05 µs	🔴 +6%

macOS

⏱️ Time (vs 🌙 nightly 4.151.0-nightly.83)

Benchmark	baseline	this PR	Δ
`PathBoundsBenchmark.TightBounds(Points: 64)`	933.0 ns	838.2 ns	🟢 -10%
`SceneRenderBenchmark.RenderFrame(Complexity: 1)`	1.27 ms	1.18 ms	🟢 -7%
`SceneRenderBenchmark.RenderFrame(Complexity: 4)`	5.79 ms	5.39 ms	🟢 -7%
`ColorMathBenchmark.ToColor(Colors: 4096)`	20.01 µs	21.36 µs	🔴 +7%
`MatrixMapPointsBenchmark.MapPoint(Points: 4096)`	19.40 µs	18.12 µs	🟢 -7%
`RasterImageLifecycleBenchmark.CreateRasterImage(Count: 256)`	159.99 µs	149.52 µs	🟢 -7%
`MatrixOpsBenchmark.Invert(Count: 4096)`	69.83 µs	65.36 µs	🟢 -6%
`PathBoundsBenchmark.TightBounds(Points: 1024)`	8.71 µs	8.17 µs	🟢 -6%
`MatrixOpsBenchmark.Concat(Count: 4096)`	118.81 µs	111.64 µs	🟢 -6%
`RuntimeEffectShaderBenchmark.DrawFrame`	551.19 µs	517.92 µs	🟢 -6%
`RasterImageLifecycleBenchmark.CreateDataWithReleaseProc(Count: 256)`	68.74 µs	72.82 µs	🔴 +6%
`MatrixMapPointsBenchmark.MapRadius(Points: 256)`	6.51 µs	6.14 µs	🟢 -6%
`MatrixMapPointsBenchmark.MapRadius(Points: 4096)`	104.04 µs	98.53 µs	🟢 -5%
`BitmapDrawBenchmark.DrawUnscaledTiles(Tiles: 64)`	172.30 µs	181.15 µs	🔴 +5%
`MatrixMapPointsBenchmark.MapPoint(Points: 256)`	1.21 µs	1.15 µs	🟢 -5%

📈 Full interactive perf-dashboard & run details →

github-actions · 2026-07-23T03:03:12Z

📦 Artifact size report

Packages from this PR (build 1522410) vs the latest nightly baseline 4.151.0-nightly.83 (observed 2026-07-22).

Total .nupkg size: 536.7 MB → 535.4 MB (−1.3 MB, -0.2%)

Packages

⚠️ marks growth over 500.0 KB or 2%. Changes under 50.0 KB are treated as noise.

Package	baseline	this PR	Δ	Δ%
`SkiaSharp.NativeAssets.WinUI`	109.0 MB	107.1 MB	🟢 −1.9 MB	-1.7%
`SkiaSharp.NativeAssets.NanoServer` ⚠️	21.7 MB	22.5 MB	🔴 +841.6 KB	+3.8%
`HarfBuzzSharp.NativeAssets.Win32`	20.8 MB	20.5 MB	🟢 −258.2 KB	-1.2%

+37 package(s) unchanged (< 50.0 KB).

Per-file changes

`HarfBuzzSharp.NativeAssets.Win32`

File	Size
`runtimes/win-x86/native/libHarfBuzzSharp.pdb`	22.2 MB → 21.9 MB (🟢 −272.0 KB)
`runtimes/win-x64/native/libHarfBuzzSharp.pdb`	22.0 MB → 21.8 MB (🟢 −240.0 KB)
`runtimes/win-arm64/native/libHarfBuzzSharp.pdb`	22.3 MB → 22.3 MB (🟢 −8.0 KB)

`SkiaSharp.NativeAssets.NanoServer`

File	Size
`runtimes/win-x64/native/libSkiaSharp.pdb`	68.8 MB → 70.2 MB (🔴 +1.5 MB)
🧬 `win-x64` (`runtimes/win-x64/native/libSkiaSharp.dll`)	9.6 MB → 10.2 MB (🔴 +596.5 KB)

`SkiaSharp.NativeAssets.WinUI`

File	Size
`runtimes/win-x86/native/libGLESv2.pdb`	67.3 MB → 67.0 MB (🟢 −240.0 KB)
`runtimes/win-x64/native/libGLESv2.pdb`	66.6 MB → 66.4 MB (🟢 −208.0 KB)
`runtimes/win-arm64/native/libGLESv2.pdb`	64.5 MB → 64.3 MB (🟢 −192.0 KB)
`runtimes/win-x64/native/libEGL.pdb`	6.4 MB → 6.3 MB (🟢 −32.0 KB)
`runtimes/win-x86/native/libEGL.pdb`	6.5 MB → 6.5 MB (🟢 −16.0 KB)
`runtimes/win-arm64/native/SkiaSharp.Views.WinUI.Native.pdb`	67.7 MB → 67.7 MB (🟢 −8.0 KB)
`runtimes/win-x64/native/SkiaSharp.Views.WinUI.Native.pdb`	68.0 MB → 68.0 MB (🟢 −8.0 KB)
🧬 `win-x64` (`runtimes/win-x64/native/libGLESv2.dll`)	4.9 MB → 4.9 MB (🔴 +2.0 KB)
🧬 `win-arm64` (`runtimes/win-arm64/native/libGLESv2.dll`)	4.4 MB → 4.4 MB (🔴 +1.0 KB)

_{Informational only — this never blocks the PR. Native binaries are labelled by os/arch.}

github-actions Bot added partner/agentic-workflows Issues and PRs created by SkiaSharp agentic workflows. perf/memory-leak Unbounded memory growth: leaked native handles or undisposed objects. Implies tenet/performance. tenet/performance Performance related issues labels Jul 16, 2026

github-project-automation Bot added this to SkiaSharp Backlog Jul 16, 2026

mattleibow mentioned this pull request Jul 17, 2026

Add regression-tracking benchmarks for common colour/interop paths #4476

Merged

Merge branch 'main' into dev/memory-leak-image-create-cotaskmem-7cb41…

d5e45bc

…f47978d8b90

mattleibow requested a review from Copilot July 20, 2026 15:02

Copilot started reviewing on behalf of mattleibow July 20, 2026 15:03 View session

Copilot AI reviewed Jul 20, 2026

View reviewed changes

Comment thread binding/SkiaSharp/SKImage.cs

Comment thread tests/Tests/SkiaSharp/SKImageTest.cs Outdated

Comment thread tests/Tests/SkiaSharp/SKImageTest.cs Outdated

Comment thread tests/Tests/SkiaSharp/SKImageTest.cs Outdated

mattleibow and others added 3 commits July 20, 2026 17:19

Potential fix for pull request finding

54e5457

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

Apply suggestions from code review

3888e34

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

mattleibow marked this pull request as ready for review July 22, 2026 21:53

Merge branch 'main' into dev/memory-leak-image-create-cotaskmem-7cb41…

5aca152

…f47978d8b90

mattleibow merged commit 2bd34ad into main Jul 22, 2026
27 of 87 checks passed

mattleibow deleted the dev/memory-leak-image-create-cotaskmem-7cb41f47978d8b90 branch July 22, 2026 23:43

github-project-automation Bot moved this to Done in SkiaSharp Backlog Jul 22, 2026

github-actions Bot mentioned this pull request Jul 23, 2026

[docs] Regenerate API diffs and release notes #4526

Closed

This was referenced Jul 23, 2026

[docs] docs: regenerate API diffs and release notes #4529

Closed

[docs] Regenerate API diffs and release notes #4530

Closed

[docs] regenerate API diffs and release notes #4533

Merged

[docs] Regenerate API diffs and release notes #4551

Closed

github-actions Bot mentioned this pull request Jul 28, 2026

[docs] regenerate API diffs and release notes #4563

Merged

1 task

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[memory-leak] Fix CoTaskMem pixel-buffer leak in SKImage.Create raster failure path - #4455

[memory-leak] Fix CoTaskMem pixel-buffer leak in SKImage.Create raster failure path#4455
mattleibow merged 6 commits into
mainfrom
dev/memory-leak-image-create-cotaskmem-7cb41f47978d8b90

github-actions Bot commented Jul 16, 2026

Uh oh!

github-actions Bot commented Jul 17, 2026

Uh oh!

github-actions Bot commented Jul 17, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions Bot commented Jul 22, 2026 •

edited

Loading

Linux

Windows

macOS

Uh oh!

Uh oh!

github-actions Bot commented Jul 23, 2026

`HarfBuzzSharp.NativeAssets.Win32`

`SkiaSharp.NativeAssets.NanoServer`

`SkiaSharp.NativeAssets.WinUI`

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

github-actions Bot commented Jul 16, 2026

The fix

Proof (red → green)

Scope note

Uh oh!

github-actions Bot commented Jul 17, 2026

📦 Try the packages from this PR

Uh oh!

github-actions Bot commented Jul 17, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions Bot commented Jul 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📊 SkiaSharp benchmarks — PR #4455

Highlights

Linux

Windows

macOS

Uh oh!

Uh oh!

github-actions Bot commented Jul 23, 2026

📦 Artifact size report

Packages

HarfBuzzSharp.NativeAssets.Win32

SkiaSharp.NativeAssets.NanoServer

SkiaSharp.NativeAssets.WinUI

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

github-actions Bot commented Jul 22, 2026 •

edited

Loading

`HarfBuzzSharp.NativeAssets.Win32`

`SkiaSharp.NativeAssets.NanoServer`

`SkiaSharp.NativeAssets.WinUI`