SPIR-V image operations (#3163)

* Add __truncate and __sampledType for spirv_asm

Allows some texture tests to start passing

* add __isVector

Currently unused

* Add 1-vector legalization pass (WIP)

* Add capabilities for image types

* neaten instruction dumping

* add 1-vector test

* Add a couple of cases to vec1 legalization

* Remove texture tests from expected failures

* comment

* regenerate vs projects

* Remove redundant define form synchapi emulation

* refactoring image methods

* All sample functions refactored

* Remove incorrect glsl intrinsics

Partially addresses https://github.com/shader-slang/slang/issues/3174

* __subscript image ops via writing funcs

* Extract texture struct writing from core.meta.slang

* Abstract out cuda intrinsic

* Remvoe erroneous call to opDecorateIndex

* spirv asm IR utils

* Correct position of loads for SPIR-V asm inst operands

* Raise constructors to global scope during spir-v legalization

* Correct snippet output

* Implement most texture sampling ops for SPIR-V

* Legalize 1-vectors for glsl too

* Make SPIR-V inst operands non-hoistable

* Better 1-vector legalization

* Put textures in ptrs for spirv

* insert missing break

* Add vec1 legalization test

* Add some missing pieces to slang-ir-insts

* Greatly neaten vec1 legalization

* a

* Neaten vec1 legalization

* Add image read and write intrinsics for spir-v

* Squash warnings

* regenerate vs projects

* Drop redundant guards

* Drop 5 tests from expected failure list

* Inst numbering changes to cross compile tests

* vec1 legalization tests only on vk

* Correct location of asm op emit

* Inline constant in spirv-asm

* Correct signedness for lane in wave intrinsics

* Extract element from float1 for cuda

* squash warnings

* Neaten spirv-emit

* dedupe more capabilities

* warnings

* neaten assert

* comments

* comments

2 files changed, 81 insertions, 0 deletions

diff --git a/tests/language-feature/1-vector.slang b/tests/language-feature/1-vector.slang
new file mode 100644
index 000000000..bb8cedf3c
--- /dev/null
+++ b/tests/language-feature/1-vector.slang
@@ -0,0 +1,19 @@
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK):-shaderobj
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK):-dx12 -use-dxil -shaderobj
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK):-cpu -shaderobj
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK):-vk -shaderobj
+
+//TEST_INPUT:ubuffer(data=[1 2 3 4], stride=4):out,name=outputBuffer
+RWStructuredBuffer<int> outputBuffer;
+
+// CHECK:      0
+// CHECK-NEXT: 1
+// CHECK-NEXT: 2
+// CHECK-NEXT: 3
+
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    vector<int,1> i = vector<int,1>(dispatchThreadID.x);
+    outputBuffer[i.x] = i.x;
+}
diff --git a/tests/language-feature/spirv-asm/truncate.slang b/tests/language-feature/spirv-asm/truncate.slang
new file mode 100644
index 000000000..9837890f5
--- /dev/null
+++ b/tests/language-feature/spirv-asm/truncate.slang
@@ -0,0 +1,62 @@
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK):-vk -shaderobj -emit-spirv-directly -output-using-type
+
+//TEST_INPUT:ubuffer(data=[1 2 3 4], stride=4):out,name=outputBuffer
+RWStructuredBuffer<int> outputBuffer;
+
+// CHECK:      8
+// CHECK-NEXT: 13
+// CHECK-NEXT: 18
+// CHECK-NEXT: 23
+
+//
+// This test tests the __truncate operator
+//
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    int i = dispatchThreadID.x;
+    int n = outputBuffer[i];
+
+    int scalar = n;
+
+    // 1-vectors are not valid in SPIR-V
+    // vector<int, 1> vector1 = vector<int, 1>(n);
+
+    vector<int, 4> vector4 = n + vector<int, 4>(0,1,2,3);
+    //int expected = 0 + n + n + (n + (n+1) + (n+2));
+
+    int r = 0;
+    spirv_asm
+    {
+        // scalar to scalar
+        __truncate $$int %a1 $$int $scalar;
+        %r1 : $$int = OpIAdd %a1 $r;
+
+        // scalar to 1-vector
+        // __truncate $$vector<int,1> %a2 $$int $scalar;
+        // %x1 : $$int = OpCompositeExtract %a2 0;
+        // %r2 : $$int = OpIAdd %x1 %r1;
+        %r2 : $$int = OpCopyObject %r1;
+
+        // 1-vector to scalar
+        // __truncate $$int %a3 $$vector<int,1> $vector1;
+        // %r3 : $$int = OpIAdd %a3 %r2;
+        %r3 : $$int = OpCopyObject %r2;
+
+        // n-vector to scalar
+        __truncate $$int %a4 $$vector<int,4> $vector4;
+        %r4 : $$int = OpIAdd %a4 %r3;
+
+        // n-vector to m-vector
+        __truncate $$vector<int,3> %a5 $$vector<int,4> $vector4;
+        %x2 : $$int = OpCompositeExtract %a5 0;
+        %x3 : $$int = OpCompositeExtract %a5 1;
+        %x4 : $$int = OpCompositeExtract %a5 2;
+        %r5 : $$int = OpIAdd %x2 %r4;
+        %r6 : $$int = OpIAdd %x3 %r5;
+        %r7 : $$int = OpIAdd %x4 %r6;
+
+        OpStore &r %r7
+    };
+    outputBuffer[i] = r;
+}