From 09608f209641d8b8662516b300571359f437c168 Mon Sep 17 00:00:00 2001 From: Ashley Coleman Date: Tue, 21 Apr 2026 13:56:17 -0600 Subject: [PATCH 1/4] Update release notes --- docs/ReleaseNotes.md | 32 ++++++++++++++++++++++++++------ 1 file changed, 26 insertions(+), 6 deletions(-) diff --git a/docs/ReleaseNotes.md b/docs/ReleaseNotes.md index 7ff424587f..8486e591a2 100644 --- a/docs/ReleaseNotes.md +++ b/docs/ReleaseNotes.md @@ -17,15 +17,16 @@ The included licenses apply to the following files: ## Changelog -### Upcoming Release - -Place release notes for the upcoming release below this line and remove this -line upon naming the release. Refer to previous for appropriate section names. +### Version 1.10.2605 #### Experimental Shader Model 6.10 -- Removed experimental Cooperative Vector, this has been replaced by LinAlg matrix. -- Implement GetGroupWaveIndex and GetGroupWaveCount in experimental Shader Model 6.10. +- Removed experimental Cooperative Vector, this has been replaced by LinAlg Matrix. +- Added DirectX LinAlg Matrix API in experimental Shader Model 6.10. + - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0035-linalg-matrix.md) + - New `Matrix` type that operates under three "scopes". ThreadScope, WaveScope, ThreadGroupScope + - Supports a comprehensive set of Matrix operations including matrix-vector and matrix-matrix multiply +- Added GetGroupWaveIndex and GetGroupWaveCount in experimental Shader Model 6.10. - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0048-group-wave-index.md) - GetGroupWaveIndex: New intrinsic for Compute, Mesh, Amplification and Node shaders which returns the index of the wave within the thread group that the the thread is executing. - GetGroupWaveCount: New intrinsic for Compute, Mesh, Amplification and Node @@ -35,6 +36,13 @@ line upon naming the release. Refer to previous for appropriate section names. - `DebugBreak()` triggers a breakpoint if a debugger is attached. - `dx::IsDebuggerPresent()` returns true if a debugger is attached. - SPIR-V: `DebugBreak()` emits `NonSemantic.DebugBreak` extended instruction; `IsDebuggerPresent()` is not supported. +- Added TriangleObjectPositions in experimental Shader Model 6.10. + - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0041-triangle-object-positions.md) +- Added CluterID in experimental Shader Model 6.10. + - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0045-clustered-geometry.md) +- Added Variable GroupShared Limits in experimental Shader Model 6.10. + - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0049-variable-groupshared-memory.md) + #### Bug Fixes @@ -45,6 +53,16 @@ line upon naming the release. Refer to previous for appropriate section names. - User-defined conversion operators (e.g., `operator float4()`) now produce an error instead of being silently ignored. [#5103](https://github.com/microsoft/DirectXShaderCompiler/pull/8206) +- Fixed SPIRV crash with out-of-line template decl + [#8079](https://github.com/microsoft/DirectXShaderCompiler/pull/8079) +- Fixed crash in CanConvert from incomplete type + [#8110](https://github.com/microsoft/DirectXShaderCompiler/pull/8110) +- Fixed crash in diagnostics when source contains embedded nulls + [#8164](https://github.com/microsoft/DirectXShaderCompiler/pull/8164) +- Fixed OOB in subscript indexing of col-major matrix in cbuffer + [#7866](https://github.com/microsoft/DirectXShaderCompiler/pull/7866) +- Allow groupshared args in templates + [#8217](https://github.com/microsoft/DirectXShaderCompiler/pull/8217) - DXIL validation: added validation for `CreateHandleFromBinding`. - DXIL validation now rejects non-standard integer bit widths (e.g. `i25`) in instructions. @@ -55,6 +73,8 @@ line upon naming the release. Refer to previous for appropriate section names. default, with `/Fi` to override the output filename. The old FXC-style `/P ` positional syntax has been renamed to `/Po`. [#4611](https://github.com/microsoft/DirectXShaderCompiler/issues/4611). +- Added support for -Fre for Metal + [#8159](https://github.com/microsoft/DirectXShaderCompiler/pull/8159) ### Version 1.9.2602 From c7152e3e0ec18cd88c03e707740a937159468b81 Mon Sep 17 00:00:00 2001 From: Ashley Coleman Date: Tue, 21 Apr 2026 15:12:25 -0600 Subject: [PATCH 2/4] add missing entry --- docs/ReleaseNotes.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/ReleaseNotes.md b/docs/ReleaseNotes.md index 8486e591a2..8a460a27d1 100644 --- a/docs/ReleaseNotes.md +++ b/docs/ReleaseNotes.md @@ -63,6 +63,8 @@ The included licenses apply to the following files: [#7866](https://github.com/microsoft/DirectXShaderCompiler/pull/7866) - Allow groupshared args in templates [#8217](https://github.com/microsoft/DirectXShaderCompiler/pull/8217) +- Fixed RawBufferVectorLoad/Store to use 32-bit element types for min precision types instead of 16-bit + [#8274](https://github.com/microsoft/DirectXShaderCompiler/pull/8274) - DXIL validation: added validation for `CreateHandleFromBinding`. - DXIL validation now rejects non-standard integer bit widths (e.g. `i25`) in instructions. From 2515a3154f8b48b9ab0b1f24af4ea895bfa80af5 Mon Sep 17 00:00:00 2001 From: Ashley Coleman Date: Tue, 21 Apr 2026 17:08:14 -0600 Subject: [PATCH 3/4] Address comments --- docs/ReleaseNotes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/ReleaseNotes.md b/docs/ReleaseNotes.md index 8a460a27d1..adb2b045ab 100644 --- a/docs/ReleaseNotes.md +++ b/docs/ReleaseNotes.md @@ -21,11 +21,11 @@ The included licenses apply to the following files: #### Experimental Shader Model 6.10 -- Removed experimental Cooperative Vector, this has been replaced by LinAlg Matrix. - Added DirectX LinAlg Matrix API in experimental Shader Model 6.10. - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0035-linalg-matrix.md) - New `Matrix` type that operates under three "scopes". ThreadScope, WaveScope, ThreadGroupScope - Supports a comprehensive set of Matrix operations including matrix-vector and matrix-matrix multiply +- Removed experimental Cooperative Vector, this has been replaced by LinAlg Matrix. - Added GetGroupWaveIndex and GetGroupWaveCount in experimental Shader Model 6.10. - [proposal](https://github.com/microsoft/hlsl-specs/blob/main/proposals/0048-group-wave-index.md) - GetGroupWaveIndex: New intrinsic for Compute, Mesh, Amplification and Node shaders which returns the index of the wave within the thread group that the the thread is executing. From 35a07ba3190372f97042ce537087d287ff14b828 Mon Sep 17 00:00:00 2001 From: Ashley Coleman Date: Tue, 21 Apr 2026 17:36:14 -0600 Subject: [PATCH 4/4] Update docs/ReleaseNotes.md Co-authored-by: Damyan Pepper --- docs/ReleaseNotes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/ReleaseNotes.md b/docs/ReleaseNotes.md index adb2b045ab..d5789a630e 100644 --- a/docs/ReleaseNotes.md +++ b/docs/ReleaseNotes.md @@ -17,7 +17,7 @@ The included licenses apply to the following files: ## Changelog -### Version 1.10.2605 +### Version 1.10.2605 (preview) #### Experimental Shader Model 6.10