Capability System: Implicit capability upgrade warning/error (#4241)

* capability upgrade warning/error

adjusted implementation + tests to support a warning/error if capabilities are implicitly upgraded and test accordingly.

* add glsl profile caps

* add GLSL and HLSL capabilities to the associated capability

* syntax error in capdef

* only error if user explicitly enables capabilities

1. changed testing infrastructure to not set a `profile` explicitly,
2. Added tests to be sure this works as intended with user API and with slangc command line

* Change capability atom definitions and how Slang manages them to fix errors

1. most `glsl_spirv` version atoms have been removed from `.capdef`, instead we will translate `spirv` version atoms into `glsl_spirv` since there is no point in writing the same code twice in `.capdef` files to define `spirv` versions.
2. add spirv version, and hlsl sm version (and equivlent) capability dependencies
3. removed some stage requirments which were set on objects, keep the wrapper capabilities. I am keeping the wrapper capabilities since I am unaware on if there are stage limitations (spec says code in practice does not work).

* check internal version instead of version profile (_spirv_1_5 vs. spirv_1_5)

* remove unused OpCapability. adjust SPIRV version'ing again for glsl_spirv

* apply workaround for glslang bug with rayquery usage

* ensure capabilities targetted by a profile and added together by a user are valid

* remove additions to `spirv_1_*` wrapper

* spirv_* -> glsl_spirv fix

* fix bug where incompatable profiles would cause invalid target caps

* try to avoid joining invalid capabilities

* fix the warning/error & printing

* run through tests to fix capability system and test mistakes

many mistakes were mesh shaders doing `-profile glsl_450+spirv_1_4`. This is not allowed for a few reasons
1. the test tooling does not handle arguments the same as `slangc`
2. glsl_450 core profile does not support mesh shaders, nor does spirv_1_4. sm_6_5 does work in this senario

* set some sm_4_1 intrinsics to sm_4_0

* replace `GLSL_` defs with `glsl_`

* swap the unsupported render-test syntax for working syntax

* set d3d11/d3d12 profile defaults

this is required since sm version changes compiled code & behavior

* adjusted nvapi capabilities with atomics + d3d11 set to use sm_5_0 as per default

* cleanup

* address review

* incorrect styling

* change `bitscanForward` to work as intended on 32 bit targets

---------

Co-authored-by: Yong He <yonghe@outlook.com>

18 files changed, 88 insertions, 28 deletions

diff --git a/tests/bugs/sample-grad-clamp-lod.slang b/tests/bugs/sample-grad-clamp-lod.slang
index cdb48fba0..8d1d232c7 100644
--- a/tests/bugs/sample-grad-clamp-lod.slang
+++ b/tests/bugs/sample-grad-clamp-lod.slang
@@ -1,4 +1,4 @@
-//TEST:CROSS_COMPILE: -profile glsl_460+GL_EXT_ray_tracing -stage miss -entry main -target spirv-assembly
+//TEST:CROSS_COMPILE: -profile sm_6_3 -stage miss -entry main -target spirv-assembly
 
 Texture2DArray  t2D;
 SamplerState samplerState;
diff --git a/tests/compute/dynamic-dispatch-14.slang b/tests/compute/dynamic-dispatch-14.slang
index 0dc99b432..609d1a215 100644
--- a/tests/compute/dynamic-dispatch-14.slang
+++ b/tests/compute/dynamic-dispatch-14.slang
@@ -2,8 +2,8 @@
 
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -profile sm_6_0 -use-dxil
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx11 -profile sm_5_0
-//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -gl  -profile glsl440
-//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -vk -profile glsl440
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -gl  -profile glsl_440
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -vk -profile glsl_440
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -cpu
 
 
diff --git a/tests/feature/source-embed/source-embed-1.slang b/tests/feature/source-embed/source-embed-1.slang
index ecacaed9b..b595cc173 100644
--- a/tests/feature/source-embed/source-embed-1.slang
+++ b/tests/feature/source-embed/source-embed-1.slang
@@ -1,4 +1,4 @@
-//TEST:SIMPLE(filecheck=CHECK):-target spirv -entry computeMain -profile cs_6_3 -line-directive-mode none -source-embed-style default -emit-spirv-via-glsl
+//TEST:SIMPLE(filecheck=CHECK):-target spirv -entry computeMain -profile cs_5_1 -line-directive-mode none -source-embed-style default -emit-spirv-via-glsl
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0], stride=4):out,name outputBuffer
 RWStructuredBuffer<int> outputBuffer;
diff --git a/tests/language-feature/capability/capability5.slang b/tests/language-feature/capability/capability5.slang
new file mode 100644
index 000000000..6199f045a
--- /dev/null
+++ b/tests/language-feature/capability/capability5.slang
@@ -0,0 +1,29 @@
+//TEST:SIMPLE(filecheck=PASS): -target glsl -entry main -stage compute -allow-glsl
+//TEST:SIMPLE(filecheck=PASS): -target glsl -entry main -stage compute -allow-glsl -capability shader5_sm_5_0
+//TEST:SIMPLE(filecheck=PASS): -target glsl -entry main -stage compute -allow-glsl -profile sm_6_0
+//TEST:SIMPLE(filecheck=WARN): -target glsl -entry main -stage compute -allow-glsl -capability GLSL_130
+//TEST:SIMPLE(filecheck=ERROR): -target glsl -entry main -stage compute -allow-glsl -capability GLSL_130 -restrictive-capability-check
+// CHECK_IGNORE_CAPS-NOT: error 36104
+
+// Check that a non-static member method implictly requires capabilities
+// defined in ThisType.
+
+//PASS-NOT: warning 41012
+//PASS-NOT: error 41012
+
+//WARN: warning 41012
+//WARN-NOT: error 41012
+
+//ERROR-NOT: warning 41012
+//ERROR: error 41012
+
+buffer MyBlockName
+{
+    int data;
+} inputBuffer;
+
+[numthreads(1,1,1)]
+void main()
+{
+    inputBuffer.data = firstbithigh(1);
+}
diff --git a/tests/language-feature/capability/specializeTargetSwitch.slang b/tests/language-feature/capability/specializeTargetSwitch.slang
index 251adfaf8..e1e5d4225 100644
--- a/tests/language-feature/capability/specializeTargetSwitch.slang
+++ b/tests/language-feature/capability/specializeTargetSwitch.slang
@@ -1,6 +1,8 @@
 //TEST:SIMPLE(filecheck=CHECK_HLSL): -target hlsl -entry main -stage compute -capability _sm_5_1
 //TEST:SIMPLE(filecheck=CHECK_GLSL1): -target glsl -entry main -stage compute -capability _GLSL_420
+//TEST:SIMPLE(filecheck=CHECK_GLSL1): -target glsl -entry main -stage compute -capability _GLSL_420
 //TEST:SIMPLE(filecheck=CHECK_GLSL2): -target glsl -entry main -stage compute -capability _GLSL_330
+//TEST:SIMPLE(filecheck=CHECK_GLSL2_NO_UPGRADE): -target glsl -entry main -stage compute -capability _GLSL_330 -DTURN_OFF_LARGER_GLSL_TARGETS 
 //TEST:SIMPLE(filecheck=CHECK_METAL): -target cpp -entry main -stage compute -capability image_loadstore
 //TEST:SIMPLE(filecheck=CHECK_WILL_ERROR1): -target glsl -entry main -stage compute -capability image_loadstore -DWILL_ERROR1
 //TEST:SIMPLE(filecheck=CHECK_WILL_ERROR2): -target glsl -entry main -stage compute -capability _GLSL_130 -DWILL_ERROR2
@@ -9,11 +11,19 @@
 RWTexture1D<int> tex;
 
 //CHECK_HLSL: {{.*}}21{{.*}};
+
 //CHECK_GLSL1: {{.*}}13{{.*}}
+
 //CHECK_GLSL2: {{.*}}11{{.*}}
+
+//CHECK_GLSL2_NO_UPGRADE-NOT: warning 41012
+//CHECK_GLSL2_NO_UPGRADE-NOT: error 41012
+
 //CHECK_METAL: {{.*}}30{{.*}}
+
 //CHECK_WILL_ERROR1: error 36109
 //CHECK_WILL_ERROR2: error 41011
+
 //CHECK_GLSL3: {{.*}}30{{.*}}
 
 int specialize()
@@ -31,10 +41,12 @@ int specialize()
         return 10;
     case _GLSL_330:
         return 11;
+#ifndef TURN_OFF_LARGER_GLSL_TARGETS
     case _GLSL_400:
         return 12;
     case _GLSL_410:
         return 13;
+#endif
 #ifdef WILL_ERROR1
     case image_loadstore:
         return 14;
diff --git a/tests/language-feature/capability/testing-framework-with-profiles.slang b/tests/language-feature/capability/testing-framework-with-profiles.slang
new file mode 100644
index 000000000..215ba887e
--- /dev/null
+++ b/tests/language-feature/capability/testing-framework-with-profiles.slang
@@ -0,0 +1,21 @@
+
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=BUF):-vk -compute
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=BUF):-vk -compute -profile sm_6_1
+//TEST:SIMPLE(filecheck=CHECK): -target hlsl  -entry computeMain -stage compute
+//TEST:SIMPLE(filecheck=CHECK): -target hlsl  -entry computeMain -stage compute -capability sm_6_1
+//TEST:SIMPLE(filecheck=CHECK): -target hlsl -entry computeMain -stage compute -profile sm_6_1
+//TEST:SIMPLE(filecheck=CHECK_WARN): -target hlsl -entry computeMain -stage compute -capability sm_4_0 
+
+//CHECK-NOT: warning 41012
+//CHECK_WARN: warning 41012
+//CHECK: computeMain
+
+//TEST_INPUT:ubuffer(data=[1], stride=4):out,name outputBuffer
+RWStructuredBuffer<uint> outputBuffer;
+
+[numthreads(1, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    // BUF: 1
+    outputBuffer[0] = WaveMaskSum(0xFF, 1);
+}
diff --git a/tests/pipeline/rasterization/get-attribute-at-vertex.slang b/tests/pipeline/rasterization/get-attribute-at-vertex.slang
index a88cf8d24..9ae347a3a 100644
--- a/tests/pipeline/rasterization/get-attribute-at-vertex.slang
+++ b/tests/pipeline/rasterization/get-attribute-at-vertex.slang
@@ -3,8 +3,8 @@
 // Basic test for `GetAttributeAtVertex` function
 
 //TEST:CROSS_COMPILE:-target dxil -entry main -stage fragment -profile sm_6_1
-//TEST:CROSS_COMPILE:-target spirv -entry main -stage fragment -profile glsl_450
-//TEST:SIMPLE(filecheck=CHECK):-emit-spirv-directly -target spirv -entry main -stage fragment -profile glsl_450
+//TEST:CROSS_COMPILE:-target spirv -entry main -stage fragment -profile glsl_450+GL_EXT_fragment_shader_barycentric
+//TEST:SIMPLE(filecheck=CHECK):-emit-spirv-directly -target spirv -entry main -stage fragment -profile glsl_450+GL_EXT_fragment_shader_barycentric
 
 // CHECK: OpCapability FragmentBarycentricKHR
 // CHECK: OpExtension "SPV_KHR_fragment_shader_barycentric"
diff --git a/tests/pipeline/rasterization/mesh/component-write.slang b/tests/pipeline/rasterization/mesh/component-write.slang
index afa5cd291..ed2d8e9a3 100644
--- a/tests/pipeline/rasterization/mesh/component-write.slang
+++ b/tests/pipeline/rasterization/mesh/component-write.slang
@@ -2,7 +2,7 @@
 
 // This tests that writing to individual components of the output struct works
 
-//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+spirv_1_4 -entry main -stage mesh
+//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+GL_EXT_mesh_shader -entry main -stage mesh
 
 const static float2 positions[3] = {
   float2(0.0, -0.5),
diff --git a/tests/pipeline/rasterization/mesh/dead-loop.slang b/tests/pipeline/rasterization/mesh/dead-loop.slang
index 875a12db5..987463104 100644
--- a/tests/pipeline/rasterization/mesh/dead-loop.slang
+++ b/tests/pipeline/rasterization/mesh/dead-loop.slang
@@ -1,5 +1,5 @@
 //TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -dx12 -use-dxil -profile sm_6_6 -render-features mesh-shader
-//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -vk -profile glsl_450+spirv_1_4 -render-features mesh-shader
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -vk -profile sm_6_5 -render-features mesh-shader
 
 // See https://github.com/shader-slang/slang/issues/3401
 
diff --git a/tests/pipeline/rasterization/mesh/hlsl-syntax.slang b/tests/pipeline/rasterization/mesh/hlsl-syntax.slang
index eaa853a04..76ac3a598 100644
--- a/tests/pipeline/rasterization/mesh/hlsl-syntax.slang
+++ b/tests/pipeline/rasterization/mesh/hlsl-syntax.slang
@@ -2,7 +2,7 @@
 
 // Test that we can ingest hlsl mesh output syntax
 
-//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+spirv_1_4 -entry main -stage mesh
+//TEST:CROSS_COMPILE:-target spirv -profile sm_6_5 -entry main -stage mesh
 
 const static float2 positions[3] = {
   float2(0.0, -0.5),
diff --git a/tests/pipeline/rasterization/mesh/nested-component-write.slang b/tests/pipeline/rasterization/mesh/nested-component-write.slang
index f55d5365e..681331a51 100644
--- a/tests/pipeline/rasterization/mesh/nested-component-write.slang
+++ b/tests/pipeline/rasterization/mesh/nested-component-write.slang
@@ -3,7 +3,7 @@
 // This tests that writing to individual components nested structs works for
 // mesh shader outputs.
 
-//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+spirv_1_4 -entry main -stage mesh
+//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+GL_EXT_mesh_shader -entry main -stage mesh
 
 struct Foo
 {
diff --git a/tests/pipeline/rasterization/mesh/simple.slang b/tests/pipeline/rasterization/mesh/simple.slang
index de1195d71..70dbfb73a 100644
--- a/tests/pipeline/rasterization/mesh/simple.slang
+++ b/tests/pipeline/rasterization/mesh/simple.slang
@@ -1,5 +1,6 @@
-//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -dx12 -use-dxil -profile sm_6_6 -render-features mesh-shader
-//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -vk -profile glsl_450+spirv_1_4 -render-features mesh-shader
+//T-EST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -dx12 -use-dxil -profile sm_6_6 -render-features mesh-shader
+//T-EST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -vk -render-features mesh-shader
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -mesh -output-using-type -vk -profile sm_6_5 -render-features mesh-shader
 
 // To test a simple mesh shader, we'll generate 4 triangles, the vertices of
 // each one will hold the triangle index and a value (the square). The fragment
diff --git a/tests/pipeline/rasterization/mesh/swizzled-store.slang b/tests/pipeline/rasterization/mesh/swizzled-store.slang
index 713416430..6636ef31d 100644
--- a/tests/pipeline/rasterization/mesh/swizzled-store.slang
+++ b/tests/pipeline/rasterization/mesh/swizzled-store.slang
@@ -2,7 +2,7 @@
 
 // This tests that writing to individual components of the output struct works
 
-//TEST:CROSS_COMPILE:-target spirv -profile glsl_450+spirv_1_4 -entry main -stage mesh
+//TEST:CROSS_COMPILE:-target spirv -profile sm_6_5 -entry main -stage mesh
 
 const static uint MAX_VERTS = 3;
 const static uint MAX_PRIMS = 1;
diff --git a/tests/pipeline/rasterization/mesh/task-groupshared.slang b/tests/pipeline/rasterization/mesh/task-groupshared.slang
index 518d94bfb..46334bf3e 100644
--- a/tests/pipeline/rasterization/mesh/task-groupshared.slang
+++ b/tests/pipeline/rasterization/mesh/task-groupshared.slang
@@ -1,5 +1,5 @@
 //TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -dx12 -use-dxil -profile sm_6_6 -render-features mesh-shader
-//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -vk -profile glsl_450+spirv_1_4 -render-features mesh-shader
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -vk -profile sm_6_5 -render-features mesh-shader
 
 // Similar to task-simple, except that the payload is declared as a groupshared
 // variable. During lowerin to GLSL and SPIR-V we'll have to identify this as
diff --git a/tests/pipeline/rasterization/mesh/task-simple.slang b/tests/pipeline/rasterization/mesh/task-simple.slang
index 2b2f3d186..053b24045 100644
--- a/tests/pipeline/rasterization/mesh/task-simple.slang
+++ b/tests/pipeline/rasterization/mesh/task-simple.slang
@@ -1,5 +1,5 @@
 //TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -dx12 -use-dxil -profile sm_6_6 -render-features mesh-shader
-//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -vk -profile glsl_450+spirv_1_4 -render-features mesh-shader
+//TEST(compute):COMPARE_COMPUTE_EX(filecheck-buffer=CHECK): -task -output-using-type -vk -profile sm_6_5 -render-features mesh-shader
 //TEST:SIMPLE(filecheck=HLSL):-target hlsl -entry meshMain -stage mesh
 //TEST:SIMPLE(filecheck=CHECK_SPV):-target spirv -entry taskMain -stage amplification
 
diff --git a/tests/pipeline/ray-tracing/trace-ray-inline.slang b/tests/pipeline/ray-tracing/trace-ray-inline.slang
index 23ec90885..009dffbc7 100644
--- a/tests/pipeline/ray-tracing/trace-ray-inline.slang
+++ b/tests/pipeline/ray-tracing/trace-ray-inline.slang
@@ -1,7 +1,14 @@
 // trace-ray-inline.slang
 
 //TEST:CROSS_COMPILE:-target dxil-asm -stage compute -profile sm_6_5 -entry main -line-directive-mode none
-//TEST:CROSS_COMPILE:-target spirv-asm -stage compute -profile glsl_460+spirv_1_4 -entry main -line-directive-mode none
+//TEST:SIMPLE(filecheck=CHECK):-target spirv-asm -stage compute -profile glsl_460+GL_EXT_ray_query -entry main -line-directive-mode none
+
+// CHECK: OpCapability RayQueryKHR
+// CHECK: OpExtension "SPV_KHR_ray_query"
+// CHECK: OpRayQueryInitializeKHR
+// CHECK: OpRayQueryProceedKHR
+// CHECK: OpRayQueryGetIntersectionTypeKHR
+// CHECK: OpRayQueryConfirmIntersectionKHR
 
 // The goal of this shader is to use all the main pieces
 // of functionality in DXR 1.1's `TraceRayInline` feature,
diff --git a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
index c3204a37c..4d26317b7 100644
--- a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
+++ b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
@@ -1,7 +1,8 @@
 // atomic-float-byte-address-buffer-cross.slang
 
 //TEST:SIMPLE(filecheck=CHECK_SPV): -profile cs_6_5 -entry computeMain -target spirv-assembly
-//TEST:SIMPLE(filecheck=CHECK): -profile cs_6_5 -entry computeMain -target spirv-assembly -emit-spirv-via-glsl
+//TEST:SIMPLE(filecheck=CHECK_SPV): -profile cs_6_5 -entry computeMain -target spirv-assembly -emit-spirv-via-glsl
+//TEST:SIMPLE(filecheck=CHECK_GLSL): -profile cs_6_5 -entry computeMain -target glsl
 // We can't do this test, because it relies on nvAPI
 //DISABLE_TEST:CROSS_COMPILE: -profile cs_6_5 -entry computeMain -target dxil
 
@@ -19,16 +20,6 @@ RWStructuredBuffer<float> anotherBuffer;
 [numthreads(16, 1, 1)]
 void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
 {
-    // CHECK-DAG: OpDecorate %[[V1:[a-zA-Z0-9_]+]] Binding 1
-    // CHECK-DAG: OpDecorate %[[V2:[a-zA-Z0-9_]+]] Binding 0
-    // CHECK-DAG: %[[P1:[a-zA-Z0-9_]+]] = OpTypePointer Uniform %float
-    // CHECK-DAG: %[[P2:[a-zA-Z0-9_]+]] = OpTypePointer Input %uint
-    // CHECK: OpAccessChain %[[P2]]
-    // CHECK: OpAccessChain %[[P1]] %[[V1]]
-    // CHECK: OpAccessChain %[[P1]] %[[V2]]
-    // CHECK: OpAtomicFAddEXT
-    // CHECK: OpAccessChain %[[P1]] %[[V2]]
-    // CHECK: OpAtomicFAddEXT
     uint tid = dispatchThreadID.x;
     int idx = int((tid & 3) ^ (tid >> 2)); 
 
diff --git a/tests/vkray/rayquery-closesthit.slang b/tests/vkray/rayquery-closesthit.slang
index 6576e09bd..33e1c9c9a 100644
--- a/tests/vkray/rayquery-closesthit.slang
+++ b/tests/vkray/rayquery-closesthit.slang
@@ -25,7 +25,6 @@ void main(
 }
 
 // CHECK: OpCapability RayQueryKHR
-// CHECK: OpCapability RayTracingPositionFetchKHR
 // CHECK: OpCapability RayQueryPositionFetchKHR
 // CHECK: OpExtension "SPV_KHR_ray_query"
 // CHECK: OpExtension "SPV_KHR_ray_tracing_position_fetch"