Support cooperative vector (#6223)

* Support cooperative vector without Vulkan-header update

Adding a Slang support for cooperative vector.
But this commit doesn't have Vulkan-header update.

Author: jkwak-work

docs/user-guide/a3-02-reference-capability-atoms.md

@@ -269,6 +269,14 @@ Versions
 > HLSL shader model 6.7 and related capabilities of other targets.
 > Includes related GLSL/SPIRV extensions.
 
+`sm_6_8_version`
+> HLSL shader model 6.8 and related capabilities of other targets.
+> Does not include related GLSL/SPIRV extensions.
+
+`sm_6_8`
+> HLSL shader model 6.8 and related capabilities of other targets.
+> Includes related GLSL/SPIRV extensions.
+
 `GLSL_130`
 > GLSL 130 and related capabilities of other targets.
 
@@ -407,6 +415,12 @@ Extensions
 `SPV_GOOGLE_user_type`
 > Represents the SPIR-V extension for SPV_GOOGLE_user_type.
 
+`SPV_EXT_replicated_composites`
+> Represents the SPIR-V extension for SPV_EXT_replicated_composites.
+
+`SPV_NV_cooperative_vector`
+> Represents the SPIR-V extension for SPV_NV_cooperative_vector.
+
 `spvAtomicFloat32AddEXT`
 > Represents the SPIR-V capability for atomic float 32 add operations.
 
@@ -509,6 +523,15 @@ Extensions
 `spvDemoteToHelperInvocation`
 > Represents the SPIR-V capability for demoting to helper invocation.
 
+`spvReplicatedCompositesEXT`
+> Represents the SPIR-V capability for replicated composites
+
+`spvCooperativeVectorNV`
+> Represents the SPIR-V capability for cooperative vectors
+
+`spvCooperativeVectorTrainingNV`
+> Represents the SPIR-V capability for cooperative vector training
+
 `spvMaximalReconvergenceKHR`
 > Represents the SPIR-V capability for maximal reconvergence.
 
@@ -861,6 +884,14 @@ Compound Capabilities
 `bufferreference_int64`
 > Capabilities needed to use GLSL buffer-reference's with int64
 
+`cooperative_vector`
+> Capabilities needed to use cooperative vectors
+> Note that cpp and cuda are supported via a fallback non-cooperative implementation
+> No HLSL shader model bound yet
+
+`cooperative_vector_training`
+> Capabilities needed to train cooperative vectors
+
 `any_stage`
 > Collection of all shader stages
 
@@ -1220,6 +1251,9 @@ Other
 `DX_6_7`
 > Use `sm_6_7` instead
 
+`DX_6_8`
+> Use `sm_6_8` instead
+
 `GLSL_410_SPIRV_1_0`
 > User should not use this capability
 

source/slang/core.meta.slang

@@ -3234,6 +3234,18 @@ int __alignOf_intrinsic<T>()
     return __alignOf_intrinsic_impl<T>(__default<T>());
 }
 
+[__unsafeForceInlineEarly]
+int32_t __elemToByteOffset<T>(int32_t elemOffset)
+{
+    return elemOffset * __naturalStrideOf<T>();
+}
+
+[__unsafeForceInlineEarly]
+int32_t __byteToElemOffset<T>(int32_t byteOffset)
+{
+    return byteOffset / __naturalStrideOf<T>();
+}
+
 __intrinsic_op($(kIROp_TreatAsDynamicUniform))
 T asDynamicUniform<T>(T v);