Add missing v16bf16 handling

Author: mikolaj-pirog

llvm/lib/Target/X86/X86CallingConv.td

@@ -267,19 +267,19 @@ def RetCC_X86Common : CallingConv<[
   // Vector types are returned in XMM0 and XMM1, when they fit.  XMM2 and XMM3
   // can only be used by ABI non-compliant code. If the target doesn't have XMM
   // registers, it won't have vector types.
-  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
             CCAssignToReg<[XMM0,XMM1,XMM2,XMM3]>>,
 
   // 256-bit vectors are returned in YMM0 and XMM1, when they fit. YMM2 and YMM3
   // can only be used by ABI non-compliant code. This vector type is only
   // supported while using the AVX target feature.
-  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
             CCAssignToReg<[YMM0,YMM1,YMM2,YMM3]>>,
 
   // 512-bit vectors are returned in ZMM0 and ZMM1, when they fit. ZMM2 and ZMM3
   // can only be used by ABI non-compliant code. This vector type is only
   // supported while using the AVX-512 target feature.
-  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
             CCAssignToReg<[ZMM0,ZMM1,ZMM2,ZMM3]>>,
 
   // Long double types are always returned in FP0 (even with SSE),
@@ -565,7 +565,7 @@ def CC_X86_64_C : CallingConv<[
   CCIfType<[v64i1], CCPromoteToType<v64i8>>,
 
   // The first 8 FP/Vector arguments are passed in XMM registers.
-  CCIfType<[f16, f32, f64, f128, v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfType<[f16, f32, f64, f128, v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
             CCIfSubtarget<"hasSSE1()",
             CCAssignToReg<[XMM0, XMM1, XMM2, XMM3, XMM4, XMM5, XMM6, XMM7]>>>,
 
@@ -574,13 +574,13 @@ def CC_X86_64_C : CallingConv<[
   // FIXME: This isn't precisely correct; the x86-64 ABI document says that
   // fixed arguments to vararg functions are supposed to be passed in
   // registers.  Actually modeling that would be a lot of work, though.
-  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
                           CCIfSubtarget<"hasAVX()",
                           CCAssignToReg<[YMM0, YMM1, YMM2, YMM3,
                                          YMM4, YMM5, YMM6, YMM7]>>>>,
 
   // The first 8 512-bit vector arguments are passed in ZMM registers.
-  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
             CCIfSubtarget<"hasAVX512()",
             CCAssignToReg<[ZMM0, ZMM1, ZMM2, ZMM3, ZMM4, ZMM5, ZMM6, ZMM7]>>>>,
 
@@ -593,14 +593,14 @@ def CC_X86_64_C : CallingConv<[
   CCIfType<[f80, f128], CCAssignToStack<0, 0>>,
 
   // Vectors get 16-byte stack slots that are 16-byte aligned.
-  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64], CCAssignToStack<16, 16>>,
+  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64], CCAssignToStack<16, 16>>,
 
   // 256-bit vectors get 32-byte stack slots that are 32-byte aligned.
-  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
            CCAssignToStack<32, 32>>,
 
   // 512-bit vectors get 64-byte stack slots that are 64-byte aligned.
-  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
            CCAssignToStack<64, 64>>
 ]>;
 
@@ -631,13 +631,13 @@ def CC_X86_Win64_C : CallingConv<[
   CCIfCFGuardTarget<CCAssignToReg<[RAX]>>,
 
   // 128 bit vectors are passed by pointer
-  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64], CCPassIndirect<i64>>,
+  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64], CCPassIndirect<i64>>,
 
   // 256 bit vectors are passed by pointer
-  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64], CCPassIndirect<i64>>,
+  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64], CCPassIndirect<i64>>,
 
   // 512 bit vectors are passed by pointer
-  CCIfType<[v64i8, v32i16, v16i32, v32f16, v16f32, v8f64, v8i64], CCPassIndirect<i64>>,
+  CCIfType<[v64i8, v32i16, v16i32, v32f16, v32bf16, v16f32, v8f64, v8i64], CCPassIndirect<i64>>,
 
   // Long doubles are passed by pointer
   CCIfType<[f80], CCPassIndirect<i64>>,
@@ -734,48 +734,48 @@ def CC_X86_64_AnyReg : CallingConv<[
 /// values are spilled on the stack.
 def CC_X86_32_Vector_Common : CallingConv<[
   // Other SSE vectors get 16-byte stack slots that are 16-byte aligned.
-  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
            CCAssignToStack<16, 16>>,
 
   // 256-bit AVX vectors get 32-byte stack slots that are 32-byte aligned.
-  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
            CCAssignToStack<32, 32>>,
 
   // 512-bit AVX 512-bit vectors get 64-byte stack slots that are 64-byte aligned.
-  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
            CCAssignToStack<64, 64>>
 ]>;
 
 /// CC_X86_Win32_Vector - In X86 Win32 calling conventions, extra vector
 /// values are spilled on the stack.
 def CC_X86_Win32_Vector : CallingConv<[
   // Other SSE vectors get 16-byte stack slots that are 4-byte aligned.
-  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
            CCAssignToStack<16, 4>>,
 
   // 256-bit AVX vectors get 32-byte stack slots that are 4-byte aligned.
-  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
            CCAssignToStack<32, 4>>,
 
   // 512-bit AVX 512-bit vectors get 64-byte stack slots that are 4-byte aligned.
-  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
            CCAssignToStack<64, 4>>
 ]>;
 
 // CC_X86_32_Vector_Standard - The first 3 vector arguments are passed in
 // vector registers
 def CC_X86_32_Vector_Standard : CallingConv<[
   // SSE vector arguments are passed in XMM registers.
-  CCIfNotVarArg<CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfNotVarArg<CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
                 CCAssignToReg<[XMM0, XMM1, XMM2]>>>,
 
   // AVX 256-bit vector arguments are passed in YMM registers.
-  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
                 CCIfSubtarget<"hasAVX()",
                 CCAssignToReg<[YMM0, YMM1, YMM2]>>>>,
 
   // AVX 512-bit vector arguments are passed in ZMM registers.
-  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
                 CCAssignToReg<[ZMM0, ZMM1, ZMM2]>>>,
 
   CCIfIsVarArgOnWin<CCDelegateTo<CC_X86_Win32_Vector>>,
@@ -786,16 +786,16 @@ def CC_X86_32_Vector_Standard : CallingConv<[
 // vector registers.
 def CC_X86_32_Vector_Darwin : CallingConv<[
   // SSE vector arguments are passed in XMM registers.
-  CCIfNotVarArg<CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v4f32, v2f64],
+  CCIfNotVarArg<CCIfType<[v16i8, v8i16, v4i32, v2i64, v8f16, v8bf16, v4f32, v2f64],
                 CCAssignToReg<[XMM0, XMM1, XMM2, XMM3]>>>,
 
   // AVX 256-bit vector arguments are passed in YMM registers.
-  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v8f32, v4f64],
+  CCIfNotVarArg<CCIfType<[v32i8, v16i16, v8i32, v4i64, v16f16, v16bf16, v8f32, v4f64],
                 CCIfSubtarget<"hasAVX()",
                 CCAssignToReg<[YMM0, YMM1, YMM2, YMM3]>>>>,
 
   // AVX 512-bit vector arguments are passed in ZMM registers.
-  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v16f32, v8f64],
+  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v32f16, v32bf16, v16f32, v8f64],
                 CCAssignToReg<[ZMM0, ZMM1, ZMM2, ZMM3]>>>,
 
   CCDelegateTo<CC_X86_32_Vector_Common>