43222bd309
This pull request aims to remove any dependency on OpenCL/SPIR-V type information in LLVM IR metadata. While, using metadata might simplify and prettify the resulting SPIR-V output (and restore some of the information missed in the transformation to opaque pointers), the overall methodology for resolving kernel parameter types is highly inefficient. The high-level strategy is to assign kernel parameter types in this order: 1. Resolving the types using builtin function calls as mangled names must contain type information or by looking up builtin definition in SPIRVBuiltins.td. Then: - Assigning the type temporarily using an intrinsic and later setting the right SPIR-V type in SPIRVGlobalRegistry after IRTranslation - Inserting a bitcast 2. Defaulting to LLVM IR types (in case of pointers the generic i8* type or types from byval/byref attributes) In case of type incompatibility (e.g. parameter defined initially as sampler_t and later used as image_t) the error will be found early on before IRTranslation (in the SPIRVEmitIntrinsics pass).
52 lines
2.4 KiB
LLVM
52 lines
2.4 KiB
LLVM
; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
|
|
|
|
; CHECK-DAG: OpEntryPoint Kernel %[[#test_func:]] "test"
|
|
; CHECK-DAG: OpDecorate %[[#f2_decl:]] LinkageAttributes "BuiltInGlobalOffset" Import
|
|
; CHECK-DAG: %[[#int32_ty:]] = OpTypeInt 32 0
|
|
; CHECK-DAG: %[[#i32ptr_ty:]] = OpTypePointer CrossWorkgroup %[[#int32_ty]]
|
|
; CHECK-DAG: %[[#void_ty:]] = OpTypeVoid
|
|
; CHECK-DAG: %[[#func_ty:]] = OpTypeFunction %[[#void_ty]] %[[#i32ptr_ty]]
|
|
; CHECK-DAG: %[[#int64_ty:]] = OpTypeInt 64 0
|
|
; CHECK-DAG: %[[#vec_ty:]] = OpTypeVector %[[#int64_ty]] 3
|
|
; CHECK-DAG: %[[#func2_ty:]] = OpTypeFunction %[[#vec_ty]]
|
|
;; TODO: add 64-bit constant defs
|
|
; CHECK-DAG: %[[#f2_decl]] = OpFunction %[[#vec_ty]] Pure %[[#func2_ty]]
|
|
; CHECK: OpFunctionEnd
|
|
;; Check that the function register name does not match other registers
|
|
; CHECK-NOT: %[[#void_ty]] = OpFunction
|
|
; CHECK-NOT: %[[#func_ty]] = OpFunction
|
|
; CHECK-NOT: %[[#int64_ty]] = OpFunction
|
|
; CHECK-NOT: %[[#vec_ty]] = OpFunction
|
|
; CHECK-NOT: %[[#func2_ty]] = OpFunction
|
|
; CHECK-NOT: %[[#f2_decl]] = OpFunction
|
|
|
|
define spir_kernel void @test(i32 addrspace(1)* %outOffsets) {
|
|
entry:
|
|
%0 = call spir_func <3 x i64> @BuiltInGlobalOffset() #1
|
|
%call = extractelement <3 x i64> %0, i32 0
|
|
%conv = trunc i64 %call to i32
|
|
; CHECK: %[[#i1:]] = OpInBoundsPtrAccessChain %[[#i32ptr_ty]] %[[#outOffsets:]]
|
|
; CHECK: OpStore %[[#i1:]] %[[#]] Aligned 4
|
|
%arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %outOffsets, i64 0
|
|
store i32 %conv, i32 addrspace(1)* %arrayidx, align 4
|
|
%1 = call spir_func <3 x i64> @BuiltInGlobalOffset() #1
|
|
%call1 = extractelement <3 x i64> %1, i32 1
|
|
%conv2 = trunc i64 %call1 to i32
|
|
; CHECK: %[[#i2:]] = OpInBoundsPtrAccessChain %[[#i32ptr_ty]] %[[#outOffsets]]
|
|
; CHECK: OpStore %[[#i2:]] %[[#]] Aligned 4
|
|
%arrayidx3 = getelementptr inbounds i32, i32 addrspace(1)* %outOffsets, i64 1
|
|
store i32 %conv2, i32 addrspace(1)* %arrayidx3, align 4
|
|
%2 = call spir_func <3 x i64> @BuiltInGlobalOffset() #1
|
|
%call4 = extractelement <3 x i64> %2, i32 2
|
|
%conv5 = trunc i64 %call4 to i32
|
|
; CHECK: %[[#i3:]] = OpInBoundsPtrAccessChain %[[#i32ptr_ty]] %[[#outOffsets]]
|
|
; CHECK: OpStore %[[#i3:]] %[[#]] Aligned 4
|
|
%arrayidx6 = getelementptr inbounds i32, i32 addrspace(1)* %outOffsets, i64 2
|
|
store i32 %conv5, i32 addrspace(1)* %arrayidx6, align 4
|
|
ret void
|
|
}
|
|
|
|
declare spir_func <3 x i64> @BuiltInGlobalOffset() #1
|
|
|
|
attributes #1 = { nounwind readnone }
|