Change log for October 25, 2024 Vulkan 1.3.300 spec update:

Github Issues * Add SHADY author ID (public PR 2448). * Migrate antora/Makefile targets into top-level Makefile (public PR 2450). Internal Issues * Update XML feature schema and generators to express more dependency cases, allowing replacing more hand-written feature specification language (internal issues 3949, 3947, 3950): ** Allows multiple versions of the same feature to be required in the xml (e.g. the same feature in two structs), without duplicating autogenerated text in the feature requirements section of the spec. ** Allows definition of "this or that feature" as a requirement. ** Adds features to depends attributes on require/remove blocks with features in them, enabling more complex feature dependencies. ** Enables more complex dependencies for feature requirements (and/or lists with no parentheses). * Fix typo in common draw dispatch validity VU 08607 (internal issue 4056). * Clarify device-host concurrent resource access language for vkFlushMappedMemoryRanges and the <<resources-memory-overlap, Resource Memory Overlap>> section (internal MR 5771). * Improve readability of the <<features, Features>> chapter introduction (internal MR 6888). * Fix typo in VkGeneratedCommandsInfoEXT VU 11072 (internal MR 6904). * Move VkRayTracingPipelineCreateInfoNV VU 11008 from the common ray tracking pipeline create info validity section to be specific to that structure (internal MR 6905). * Add missing VkGeneratedCommandsInfoEXT VU mirroring VkGeneratedCommandsMemoryRequirementsInfoEXT VU 11011 (internal MR 6948). * Remove duplicate VkSwapchainCreateInfoKHR VU 10156 (internal MR 6942). * Combine common draw validity discard rectangle VUs for vkCmdSetDiscardRectangleEnableEXT and vkCmdSetDiscardRectangleModeEXT (internal MR 6953). * Combine common draw validity sample location VUs for VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_ENABLE_EXT and vkCmdSetSampleLocationsEnableEXT (internal MR 6955). * Combine common draw validity line stipple dynamic state VUs for shader objects and pipelines (internal MR 6956). * Remove vkCmdDrawMeshTasksIndirectCountNV, vkCmdDrawMeshTasksIndirectCountEXT, vkCmdDrawIndirectCount, and vkCmdDrawIndexedIndirectCount VUs since drawCount is already capped by maxDrawCount (internal MR 6969). * Fix typo in description of vkCmdCopyMemoryToMicromapEXT::pInfo (internal MR 6974). New Extensions * VK_NV_cooperative_matrix2
KhronosGroup · Oct 25, 2024 · c9304f8 · c9304f8
1 parent f3a114c
commit c9304f8
Show file tree

Hide file tree

Showing 28 changed files with 1,159 additions and 402 deletions.
diff --git a/ChangeLog.adoc b/ChangeLog.adoc
@@ -14,6 +14,63 @@ appears frequently in the change log.
 
 '''
 
+Change log for October 25, 2024 Vulkan 1.3.300 spec update:
+
+Github Issues
+
+  * Add SHADY author ID (public PR 2448).
+  * Migrate antora/Makefile targets into top-level Makefile (public PR
+    2450).
+
+Internal Issues
+
+  * Update XML feature schema and generators to express more dependency
+    cases, allowing replacing more hand-written feature specification
+    language (internal issues 3949, 3947, 3950):
+  ** Allows multiple versions of the same feature to be required in the xml
+     (e.g. the same feature in two structs), without duplicating
+     autogenerated text in the feature requirements section of the spec.
+  ** Allows definition of "this or that feature" as a requirement.
+  ** Adds features to depends attributes on require/remove blocks with
+     features in them, enabling more complex feature dependencies.
+  ** Enables more complex dependencies for feature requirements (and/or
+     lists with no parentheses).
+  * Fix typo in common draw dispatch validity VU 08607 (internal issue
+    4056).
+  * Clarify device-host concurrent resource access language for
+    vkFlushMappedMemoryRanges and the <<resources-memory-overlap, Resource
+    Memory Overlap>> section (internal MR 5771).
+  * Improve readability of the <<features, Features>> chapter introduction
+    (internal MR 6888).
+  * Fix typo in VkGeneratedCommandsInfoEXT VU 11072 (internal MR 6904).
+  * Move VkRayTracingPipelineCreateInfoNV VU 11008 from the common ray
+    tracking pipeline create info validity section to be specific to that
+    structure (internal MR 6905).
+  * Add missing VkGeneratedCommandsInfoEXT VU mirroring
+    VkGeneratedCommandsMemoryRequirementsInfoEXT VU 11011 (internal MR
+    6948).
+  * Remove duplicate VkSwapchainCreateInfoKHR VU 10156 (internal MR 6942).
+  * Combine common draw validity discard rectangle VUs for
+    vkCmdSetDiscardRectangleEnableEXT and vkCmdSetDiscardRectangleModeEXT
+    (internal MR 6953).
+  * Combine common draw validity sample location VUs for
+    VK_DYNAMIC_STATE_SAMPLE_LOCATIONS_ENABLE_EXT and
+    vkCmdSetSampleLocationsEnableEXT (internal MR 6955).
+  * Combine common draw validity line stipple dynamic state VUs for shader
+    objects and pipelines (internal MR 6956).
+  * Remove vkCmdDrawMeshTasksIndirectCountNV,
+    vkCmdDrawMeshTasksIndirectCountEXT, vkCmdDrawIndirectCount, and
+    vkCmdDrawIndexedIndirectCount VUs since drawCount is already capped by
+    maxDrawCount (internal MR 6969).
+  * Fix typo in description of vkCmdCopyMemoryToMicromapEXT::pInfo (internal
+    MR 6974).
+
+New Extensions
+
+  * VK_NV_cooperative_matrix2
+
+'''
+
 Change log for October 18, 2024 Vulkan 1.3.299 spec update:
 
 Internal Issues

diff --git a/Makefile b/Makefile
@@ -142,7 +142,7 @@ VERBOSE =
 # ADOCOPTS options for asciidoc->HTML5 output
 
 NOTEOPTS     = -a editing-notes -a implementation-guide
-PATCHVERSION = 299
+PATCHVERSION = 300
 BASEOPTS     =
 
 ifneq (,$(findstring VKSC_VERSION_1_0,$(VERSIONS)))

diff --git a/appendices/VK_NV_cooperative_matrix2.adoc b/appendices/VK_NV_cooperative_matrix2.adoc
@@ -0,0 +1,56 @@
+// Copyright 2024 The Khronos Group Inc.
+//
+// SPDX-License-Identifier: CC-BY-4.0
+
+include::{generated}/meta/{refprefix}VK_NV_cooperative_matrix2.adoc[]
+
+=== Other Extension Metadata
+
+*Last Modified Date*::
+    2024-08-01
+*Interactions and External Dependencies*::
+  - This extension provides API support for
+    {GLSLregistry}/nv/GLSL_NV_cooperative_matrix2.txt[`GLSL_NV_cooperative_matrix2`]
+*Contributors*::
+  - Jeff Bolz, NVIDIA
+  - Karthik Vaidyanathan, NVIDIA
+
+=== Description
+
+This extension adds several new features building on the cooperative matrix
+types added in VK_KHR_cooperative_matrix.
+The goal is to add and accelerate features beyond just simple GEMM kernels,
+including adding support for type/use conversions, reductions, per-element
+operations, and tensor addressing, and also to improve usability and
+out-of-the-box performance by adding support for more flexible matrix sizes,
+and workgroup scope matrices with compiler-managed staging through shared
+memory.
+
+The new functionality is defined by the
+{spirv}/NV/SPV_NV_tensor_addressing.html[`SPV_NV_tensor_addressing`] and
+{spirv}/NV/SPV_NV_cooperative_matrix2.html[`SPV_NV_cooperative_matrix2`]
+SPIR-V extensions and can be used with the
+{GLSLregistry}/nv/GLSL_NV_cooperative_matrix2.txt[`GLSL_NV_cooperative_matrix2`]
+GLSL extension.
+
+This extension includes support for enumerating the matrix types and
+dimensions that are supported by the implementation, and which specific
+features are supported.
+
+include::{generated}/interfaces/VK_NV_cooperative_matrix2.adoc[]
+
+=== New SPIR-V Capabilities
+
+  * <<spirvenv-capabilities-table-TensorAddressingNV,TensorAddressingNV>>
+  * <<spirvenv-capabilities-table-CooperativeMatrixReductionsNV,CooperativeMatrixReductionsNV>>
+  * <<spirvenv-capabilities-table-CooperativeMatrixConversionsNV,CooperativeMatrixConversionsNV>>
+  * <<spirvenv-capabilities-table-CooperativeMatrixPerElementOperationsNV,CooperativeMatrixPerElementOperationsNV>>
+  * <<spirvenv-capabilities-table-CooperativeMatrixTensorAddressingNV,CooperativeMatrixTensorAddressingNV>>
+  * <<spirvenv-capabilities-table-CooperativeMatrixBlockLoadsNV,CooperativeMatrixBlockLoadsNV>>
+
+=== Issues
+
+=== Version History
+
+  * Revision 1, 2024-08-01 (Jeff Bolz)
+  ** Initial revision
diff --git a/appendices/memorymodel.adoc b/appendices/memorymodel.adoc
@@ -1189,13 +1189,15 @@ ifdef::VK_NV_cooperative_matrix,VK_KHR_cooperative_matrix[]
 For each dynamic instance of a cooperative matrix load instruction
 (code:OpCooperativeMatrixLoadKHR
 ifdef::VK_NV_cooperative_matrix[, code:OpCooperativeMatrixLoadNV]
+ifdef::VK_NV_cooperative_matrix2[, code:OpCooperativeMatrixLoadTensorNV]
 ), some implementation-dependent invocation(s) within the instance of the
 matrix's scope perform a non-atomic load from each memory location that is
 defined to be accessed by the instruction.
 
 For each memory location accessed by a dynamic instance of a cooperative
 matrix store instruction (code:OpCooperativeMatrixStoreKHR
 ifdef::VK_NV_cooperative_matrix[, code:OpCooperativeMatrixStoreNV]
+ifdef::VK_NV_cooperative_matrix2[, code:OpCooperativeMatrixStoreTensorNV]
 ), a single implementation-dependent invocation within the instance of the
 matrix's scope performs a non-atomic store to that memory location.
 endif::VK_NV_cooperative_matrix,VK_KHR_cooperative_matrix[]
diff --git a/appendices/spirvenv.adoc b/appendices/spirvenv.adoc
@@ -1328,13 +1328,39 @@ ifdef::VK_NV_cooperative_matrix[]
     slink:VkPhysicalDeviceCooperativeMatrixPropertiesNV::pname:cooperativeMatrixSupportedStages
 endif::VK_NV_cooperative_matrix[]
 ifdef::VK_KHR_cooperative_matrix[]
-  * [[VUID-{refpage}-OpTypeCooperativeMatrixKHR-08974]]
-    For code:OpTypeCooperativeMatrixKHR, the component type, scope, number
-    of rows, and number of columns must: match one of the matrices in any of
-    the supported slink:VkCooperativeMatrixPropertiesKHR
+  * [[VUID-{refpage}-OpTypeCooperativeMatrixKHR-10163]]
+    For code:OpTypeCooperativeMatrixKHR,
+ifdef::VK_NV_cooperative_matrix2[]
+    if <<features-cooperativeMatrixFlexibleDimensions,
+    pname:cooperativeMatrixFlexibleDimensions>> is not enabled,
+endif::VK_NV_cooperative_matrix2[]
+    the component type, scope, number of rows, and number of columns must:
+    match one of the matrices in any of the supported
+    slink:VkCooperativeMatrixPropertiesKHR, where
+  ** for code:Use of code:MatrixA, the number of rows must match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:MSize and the number of
+     columns must match slink:VkCooperativeMatrixPropertiesKHR::pname:KSize
+     and the type must: match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:AType
+  ** for code:Use of code:MatrixB, the number of rows must match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:KSize and the number of
+     columns must match slink:VkCooperativeMatrixPropertiesKHR::pname:NSize
+     and the type must: match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:BType
+  ** for code:Use of code:MatrixAccumulator, the number of rows must match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:MSize and the number of
+     columns must match slink:VkCooperativeMatrixPropertiesKHR::pname:NSize
+     and the type must: match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:CType or
+     slink:VkCooperativeMatrixPropertiesKHR::pname:ResultType
   * [[VUID-{refpage}-OpCooperativeMatrixMulAddKHR-10060]]
-    For code:OpCooperativeMatrixMulAddKHR, the operands must: match a
-    supported slink:VkCooperativeMatrixPropertiesKHR, such that:
+    For code:OpCooperativeMatrixMulAddKHR,
+ifdef::VK_NV_cooperative_matrix2[]
+    if <<features-cooperativeMatrixFlexibleDimensions,
+    pname:cooperativeMatrixFlexibleDimensions>> is not enabled,
+endif::VK_NV_cooperative_matrix2[]
+    the operands must: match a supported
+    slink:VkCooperativeMatrixPropertiesKHR, such that:
   ** The type of code:A must: have code:Rows match
      slink:VkCooperativeMatrixPropertiesKHR::pname:MSize, code:Columns match
      slink:VkCooperativeMatrixPropertiesKHR::pname:KSize, code:Use be
@@ -1370,8 +1396,113 @@ ifdef::VK_KHR_cooperative_matrix[]
   ** If and only if
      slink:VkCooperativeMatrixPropertiesKHR::pname:saturatingAccumulation is
      ename:VK_FALSE, code:SaturatingAccumulationKHR must: not be used.
-  ** The scope of all cooperative matrix operands must: be
-     elink:VkScopeKHR::ename:VK_SCOPE_SUBGROUP_KHR
+  ** The scope of all cooperative matrix operands must: match
+     slink:VkCooperativeMatrixPropertiesKHR::pname:scope.
+ifdef::VK_NV_cooperative_matrix2[]
+  * [[VUID-{refpage}-cooperativeMatrixWorkgroupScope-10164]]
+    If <<features-cooperativeMatrixWorkgroupScope,
+    pname:cooperativeMatrixWorkgroupScope>> is not enabled, the scope of all
+    code:OpTypeCooperativeMatrixKHR must: not be
+    elink:VkScopeKHR::ename:VK_SCOPE_WORKGROUP_KHR
+  * [[VUID-{refpage}-cooperativeMatrixFlexibleDimensions-10165]]
+    For code:OpTypeCooperativeMatrixKHR, if
+    <<features-cooperativeMatrixFlexibleDimensions,
+    pname:cooperativeMatrixFlexibleDimensions>> is enabled, the component
+    type, scope, number of rows, and number of columns must: match either
+    one of the matrices in one of the supported
+    slink:VkCooperativeMatrixPropertiesKHR as described above, or one of the
+    supported slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV, where
+  ** for code:MatrixA, the number of rows must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:MGranularity
+     and the number of columns must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:KGranularity
+     and the type must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:AType
+  ** for code:MatrixB, the number of rows must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:KGranularity
+     and the number of columns must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:NGranularity
+     and the type must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:BType
+  ** for code:MatrixAccumulator, the number of rows must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:MGranularity
+     and the number of columns must be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:NGranularity
+     and the type must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:CType or
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:ResultType
+  ** if the scope is ename:VK_SCOPE_WORKGROUP_KHR, the number of invocations
+     in the local workgroup must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:workgroupInvocations
+  * [[VUID-{refpage}-cooperativeMatrixFlexibleDimensions-10166]]
+    For code:OpCooperativeMatrixMulAddKHR, if
+    <<features-cooperativeMatrixFlexibleDimensions,
+    pname:cooperativeMatrixFlexibleDimensions>> is enabled, the operands
+    must: match either one of the supported
+    slink:VkCooperativeMatrixPropertiesKHR as described above, or one of the
+    supported slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV, such
+    that:
+  ** The type of code:A must: have code:Rows be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:MGranularity,
+     code:Columns be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:KGranularity,
+     code:Use be code:MatrixAKHR, and code:ComponentType match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:AType.
+  ** The type of code:B must: have code:Rows be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:KGranularity,
+     code:Columns be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:NGranularity,
+     code:Use be code:MatrixBKHR, and code:ComponentType match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:BType.
+  ** The type of code:C must: have code:Rows be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:MGranularity,
+     code:Columns be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:NGranularity,
+     code:Use be code:MatrixAccumulatorKHR, and code:ComponentType match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:CType.
+  ** The type of code:Result must: have code:Rows be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:MGranularity,
+     code:Columns be a multiple of
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:NGranularity,
+     code:Use be code:MatrixAccumulatorKHR, and code:ComponentType match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:ResultType.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:AType is
+     a signed integer type, code:MatrixASignedComponents must: be used.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:BType is
+     a signed integer type, code:MatrixBSignedComponents must: be used.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:CType is
+     a signed integer type, code:MatrixCSignedComponents must: be used.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:ResultType
+     is a signed integer type, code:MatrixResultSignedComponents must: be
+     used.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:saturatingAccumulation
+     is ename:VK_TRUE, code:SaturatingAccumulationKHR must: be used.
+  ** If and only if
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:saturatingAccumulation
+     is ename:VK_FALSE, code:SaturatingAccumulationKHR must: not be used.
+  ** The scope of all cooperative matrix operands must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:scope.
+  ** If the scope is ename:VK_SCOPE_WORKGROUP_KHR, the number of invocations
+     in the local workgroup must: match
+     slink:VkCooperativeMatrixFlexibleDimensionsPropertiesNV::pname:workgroupInvocations
+  * [[VUID-{refpage}-cooperativeMatrixFlexibleDimensionsMaxDimension-10167]]
+    All code:OpTypeCooperativeMatrixKHR dimensions must: be less than or
+    equal to
+    <<limits-cooperativeMatrixFlexibleDimensionsMaxDimension,pname:cooperativeMatrixFlexibleDimensionsMaxDimension>>
+  * [[VUID-{refpage}-maxComputeSharedMemorySize-10168]]
+    If the module uses code:OpTypeCooperativeMatrixKHR with code:Scope equal
+    to code:Workgroup, the sum of size in bytes for variables and
+    <<workgroup-padding, padding>> in the code:Workgroup {StorageClass} in
+    the code:GLCompute {ExecutionModel} must: be less than or equal to
+    <<limits-maxComputeSharedMemorySize, pname:maxComputeSharedMemorySize>>
+    minus
+    <<limits-cooperativeMatrixWorkgroupScopeReservedSharedMemory,pname:cooperativeMatrixWorkgroupScopeReservedSharedMemory>>
+endif::VK_NV_cooperative_matrix2[]
   * [[VUID-{refpage}-cooperativeMatrixSupportedStages-08985]]
     code:OpTypeCooperativeMatrixKHR and code:OpCooperativeMatrix*
     instructions must: not be used in shader stages not included in
@@ -2317,6 +2448,10 @@ endif::VK_KHR_cooperative_matrix[]
 ifdef::VK_NV_cooperative_matrix[]
     code:OpCooperativeMatrixLoadNV, code:OpCooperativeMatrixStoreNV,
 endif::VK_NV_cooperative_matrix[]
+ifdef::VK_NV_cooperative_matrix2[]
+    code:OpCooperativeMatrixLoadTensorNV,
+    code:OpCooperativeMatrixStoreTensorNV,
+endif::VK_NV_cooperative_matrix2[]
     code:OpAtomicLoad, code:OpAtomicStore, code:OpAtomicExchange,
     code:OpStore, and code:OpLoad are referred to as _bit-preserving
     operations_.

diff --git a/chapters/VK_EXT_opacity_micromap/micromaps.adoc b/chapters/VK_EXT_opacity_micromap/micromaps.adoc
@@ -667,7 +667,7 @@ include::{generated}/api/protos/vkCmdCopyMemoryToMicromapEXT.adoc[]
 
   * pname:commandBuffer is the command buffer into which the command will be
     recorded.
-  * pname:pInfo is a pointer to a slink:VkCopyMicromapToMemoryInfoEXT
+  * pname:pInfo is a pointer to a slink:VkCopyMemoryToMicromapInfoEXT
     structure defining the copy operation.
 
 Accesses to pname:pInfo->dst must: be <<synchronization-dependencies,

diff --git a/chapters/VK_KHR_swapchain/wsi.adoc b/chapters/VK_KHR_swapchain/wsi.adoc
@@ -403,11 +403,8 @@ ifdef::VK_EXT_swapchain_maintenance1[]
     If the <<features-swapchainMaintenance1,pname:swapchainMaintenance1>>
     feature is not enabled, then the pname:pNext chain must: not include a
     slink:VkSwapchainPresentModesCreateInfoEXT structure
-  * [[VUID-VkSwapchainCreateInfoKHR-swapchainMaintenance1-10156]]
-    If the <<features-swapchainMaintenance1,pname:swapchainMaintenance1>>
-    feature is not enabled, then the pname:pNext chain must: not include a
-    slink:VkSwapchainPresentScalingCreateInfoEXT structure
 endif::VK_EXT_swapchain_maintenance1[]
+  * [[VUID-VkSwapchainCreateInfoKHR-presentMode-02839]]
 ifdef::VK_KHR_shared_presentable_image[]
     If pname:presentMode is not
     ename:VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor

diff --git a/chapters/VK_NV_mesh_shader/drawing.adoc b/chapters/VK_NV_mesh_shader/drawing.adoc
@@ -165,16 +165,6 @@ include::{chapters}/commonvalidity/draw_indirect_count_common.adoc[]
     [eq]#(pname:stride {times} (pname:maxDrawCount - 1) {plus} pname:offset
     {plus} code:sizeof(sname:VkDrawMeshTasksIndirectCommandNV))# must: be
     less than or equal to the size of pname:buffer
-  * [[VUID-vkCmdDrawMeshTasksIndirectCountNV-countBuffer-02191]]
-    If the count stored in pname:countBuffer is equal to `1`,
-    [eq]#(pname:offset {plus}
-    code:sizeof(sname:VkDrawMeshTasksIndirectCommandNV))# must: be less than
-    or equal to the size of pname:buffer
-  * [[VUID-vkCmdDrawMeshTasksIndirectCountNV-countBuffer-02192]]
-    If the count stored in pname:countBuffer is greater than `1`,
-    [eq]#(pname:stride {times} (pname:drawCount - 1) {plus} pname:offset
-    {plus} code:sizeof(sname:VkDrawMeshTasksIndirectCommandNV))# must: be
-    less than or equal to the size of pname:buffer
   * [[VUID-vkCmdDrawMeshTasksIndirectCountNV-MeshNV-07082]]
     The current pipeline bound to ename:VK_PIPELINE_BIND_POINT_GRAPHICS
     must: contain a shader stage using the code:MeshNV {ExecutionModel}
@@ -343,16 +333,6 @@ include::{chapters}/commonvalidity/draw_indirect_count_common.adoc[]
     [eq]#(pname:stride {times} (pname:maxDrawCount - 1) {plus} pname:offset
     {plus} code:sizeof(sname:VkDrawMeshTasksIndirectCommandEXT))# must: be
     less than or equal to the size of pname:buffer
-  * [[VUID-vkCmdDrawMeshTasksIndirectCountEXT-countBuffer-07098]]
-    If the count stored in pname:countBuffer is equal to `1`,
-    [eq]#(pname:offset {plus}
-    code:sizeof(sname:VkDrawMeshTasksIndirectCommandEXT))# must: be less
-    than or equal to the size of pname:buffer
-  * [[VUID-vkCmdDrawMeshTasksIndirectCountEXT-countBuffer-07099]]
-    If the count stored in pname:countBuffer is greater than `1`,
-    [eq]#(pname:stride {times} (pname:drawCount - 1) {plus} pname:offset
-    {plus} code:sizeof(sname:VkDrawMeshTasksIndirectCommandEXT))# must: be
-    less than or equal to the size of pname:buffer
   * [[VUID-vkCmdDrawMeshTasksIndirectCountEXT-MeshEXT-07100]]
     The current pipeline bound to ename:VK_PIPELINE_BIND_POINT_GRAPHICS
     must: contain a shader stage using the code:MeshEXT {ExecutionModel}