[x86] Fix a few more tiny patterns with the new vector shuffle lowering

[oota-llvm.git] / lib / Target / X86 / X86CallingConv.td

diff --git a/lib/Target/X86/X86CallingConv.td b/lib/Target/X86/X86CallingConv.td
index ef8caa0290e9cab11e871a1774320e9c607e99d5..dec73eac606ca811629dd021dfa7120f4f079155 100644 (file)
--- a/lib/Target/X86/X86CallingConv.td
+++ b/lib/Target/X86/X86CallingConv.td
@@ -14,7 +14,9 @@
 
 /// CCIfSubtarget - Match if the current subtarget has a feature F.
 class CCIfSubtarget<string F, CCAction A>
- : CCIf<!strconcat("State.getTarget().getSubtarget<X86Subtarget>().", F), A>;
+    : CCIf<!strconcat("static_cast<const X86Subtarget&>"
+                       "(State.getMachineFunction().getSubtarget()).", F),
+           A>;
 
 //===----------------------------------------------------------------------===//
 // Return Value Calling Conventions
@@ -52,27 +54,27 @@ def RetCC_X86Common : CallingConv<[
   // 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<[v16i32, v8i64, v16f32, v8f64],
+  CCIfType<[v64i8, v32i16, v16i32, v8i64, v16f32, v8f64],
             CCAssignToReg<[ZMM0,ZMM1,ZMM2,ZMM3]>>,
 
   // MMX vector types are always returned in MM0. If the target doesn't have
   // MM0, it doesn't support these vector types.
   CCIfType<[x86mmx], CCAssignToReg<[MM0]>>,
 
-  // Long double types are always returned in ST0 (even with SSE).
-  CCIfType<[f80], CCAssignToReg<[ST0, ST1]>>
+  // Long double types are always returned in FP0 (even with SSE).
+  CCIfType<[f80], CCAssignToReg<[FP0, FP1]>>
 ]>;
 
 // X86-32 C return-value convention.
 def RetCC_X86_32_C : CallingConv<[
-  // The X86-32 calling convention returns FP values in ST0, unless marked
+  // The X86-32 calling convention returns FP values in FP0, unless marked
   // with "inreg" (used here to distinguish one kind of reg from another,
   // weirdly; this is really the sse-regparm calling convention) in which
   // case they use XMM0, otherwise it is the same as the common X86 calling
   // conv.
   CCIfInReg<CCIfSubtarget<"hasSSE2()",
     CCIfType<[f32, f64], CCAssignToReg<[XMM0,XMM1,XMM2]>>>>,
-  CCIfType<[f32,f64], CCAssignToReg<[ST0, ST1]>>,
+  CCIfType<[f32,f64], CCAssignToReg<[FP0, FP1]>>,
   CCDelegateTo<RetCC_X86Common>
 ]>;
 
@@ -224,6 +226,7 @@ def CC_X86_64_C : CallingConv<[
   CCIfType<[i8, i16], CCPromoteToType<i32>>,
 
   // The 'nest' parameter, if any, is passed in R10.
+  CCIfNest<CCIfSubtarget<"isTarget64BitILP32()", CCAssignToReg<[R10D]>>>,
   CCIfNest<CCAssignToReg<[R10]>>,
 
   // The first 6 integer arguments are passed in integer registers.
@@ -252,7 +255,7 @@ def CC_X86_64_C : CallingConv<[
                                          YMM4, YMM5, YMM6, YMM7]>>>>,
 
   // The first 8 512-bit vector arguments are passed in ZMM registers.
-  CCIfNotVarArg<CCIfType<[v16i32, v8i64, v16f32, v8f64],
+  CCIfNotVarArg<CCIfType<[v64i8, v32i16, v16i32, v8i64, v16f32, v8f64],
             CCIfSubtarget<"hasAVX512()",
             CCAssignToReg<[ZMM0, ZMM1, ZMM2, ZMM3, ZMM4, ZMM5, ZMM6, ZMM7]>>>>,
 
@@ -358,12 +361,15 @@ def CC_X86_64_WebKit_JS : CallingConv<[
   CCIfType<[i8, i16], CCPromoteToType<i32>>,
 
   // Only the first integer argument is passed in register.
-  CCIfType<[i32], CCAssignToReg<[EDI]>>,
-  CCIfType<[i64], CCAssignToReg<[RDI]>>,
+  CCIfType<[i32], CCAssignToReg<[EAX]>>,
+  CCIfType<[i64], CCAssignToReg<[RAX]>>,
 
-  // Integer/FP values are always stored in stack slots that are 8 bytes in size
-  // and 8-byte aligned.
-  CCIfType<[i32, i64, f32, f64], CCAssignToStack<8, 8>>
+  // The remaining integer arguments are passed on the stack. 32bit integer and
+  // floating-point arguments are aligned to 4 byte and stored in 4 byte slots.
+  // 64bit integer and floating-point arguments are aligned to 8 byte and stored
+  // in 8 byte stack slots.
+  CCIfType<[i32, f32], CCAssignToStack<4, 4>>,
+  CCIfType<[i64, f64], CCAssignToStack<8, 8>>
 ]>;
 
 // No explicit register is specified for the AnyReg calling convention. The
@@ -617,14 +623,25 @@ def CSR_64EHRet : CalleeSavedRegs<(add RAX, RDX, CSR_64)>;
 def CSR_Win64 : CalleeSavedRegs<(add RBX, RBP, RDI, RSI, R12, R13, R14, R15,
                                      (sequence "XMM%u", 6, 15))>;
 
-def CSR_MostRegs_64 : CalleeSavedRegs<(add RBX, RCX, RDX, RSI, RDI, R8, R9, R10,
+// All GPRs - except r11
+def CSR_64_RT_MostRegs : CalleeSavedRegs<(add CSR_64, RAX, RCX, RDX, RSI, RDI,
+                                              R8, R9, R10, RSP)>;
+
+// All registers - except r11
+def CSR_64_RT_AllRegs     : CalleeSavedRegs<(add CSR_64_RT_MostRegs,
+                                                 (sequence "XMM%u", 0, 15))>;
+def CSR_64_RT_AllRegs_AVX : CalleeSavedRegs<(add CSR_64_RT_MostRegs,
+                                                 (sequence "YMM%u", 0, 15))>;
+
+def CSR_64_MostRegs : CalleeSavedRegs<(add RBX, RCX, RDX, RSI, RDI, R8, R9, R10,
                                            R11, R12, R13, R14, R15, RBP,
                                            (sequence "XMM%u", 0, 15))>;
 
-def CSR_AllRegs_64 : CalleeSavedRegs<(add CSR_MostRegs_64, RAX, RSP,
-                                      (sequence "XMM%u", 16, 31),
-                                      (sequence "YMM%u", 0, 31),
-                                      (sequence "ZMM%u", 0, 31))>;
+def CSR_64_AllRegs     : CalleeSavedRegs<(add CSR_64_MostRegs, RAX, RSP,
+                                              (sequence "XMM%u", 16, 31))>;
+def CSR_64_AllRegs_AVX : CalleeSavedRegs<(sub (add CSR_64_MostRegs, RAX, RSP,
+                                                   (sequence "YMM%u", 0, 31)),
+                                              (sequence "XMM%u", 0, 15))>;
 
 // Standard C + YMM6-15
 def CSR_Win64_Intel_OCL_BI_AVX : CalleeSavedRegs<(add RBX, RBP, RDI, RSI, R12,
@@ -643,6 +660,6 @@ def CSR_64_Intel_OCL_BI       : CalleeSavedRegs<(add CSR_64,
 def CSR_64_Intel_OCL_BI_AVX    : CalleeSavedRegs<(add CSR_64,
                                                   (sequence "YMM%u", 8, 15))>;
 
-def CSR_64_Intel_OCL_BI_AVX512    : CalleeSavedRegs<(add CSR_64,
+def CSR_64_Intel_OCL_BI_AVX512 : CalleeSavedRegs<(add RBX, RDI, RSI, R14, R15,
                                                   (sequence "ZMM%u", 16, 31),
                                                   K4, K5, K6, K7)>;