The assembly string for the VMOVPQIto64rr instruction incorrectly lacked the 'v'
prefix, resulting in mis-assembly of the vanilla movd instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162963
91177308-0d34-0410-b5e6-
96231b3b80d8
// Move Packed Doubleword Int first element to Doubleword Int
//
def VMOVPQIto64rr : I<0x7E, MRMDestReg, (outs GR64:$dst), (ins VR128:$src),
- "mov{d|q}\t{$src, $dst|$dst, $src}",
+ "vmov{d|q}\t{$src, $dst|$dst, $src}",
[(set GR64:$dst, (vector_extract (v2i64 VR128:$src),
(iPTR 0)))],
IIC_SSE_MOVD_ToGP>,
// CHECK: encoding: [0x66,0x48,0x0f,0x6e,0xc7]
movd %rdi,%xmm0
+// CHECK: movd %xmm0, %rax
+// CHECK: encoding: [0x66,0x48,0x0f,0x7e,0xc0]
+ movd %xmm0, %rax
+
// CHECK: movntil %eax, (%rdi)
// CHECK: encoding: [0x0f,0xc3,0x07]
// CHECK: movntil