mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-19 19:12:56 +02:00
421ddae8b6
ARMv8.2-A adds 16-bit floating point versions of all existing SIMD floating-point instructions. This is an optional extension, so all of these instructions require the FeatureFullFP16 subtarget feature. Note that VFP without SIMD is not a valid combination for any version of ARMv8-A, but I have ensured that these instructions all depend on both FeatureNEON and FeatureFullFP16 for consistency. The ".2h" vector type specifier is now legal (for the scalar pairwise reduction instructions), so some unrelated tests have been modified as different error messages are emitted. This is not a problem as the invalid operands are still caught. llvm-svn: 255010
42 lines
2.0 KiB
ArmAsm
42 lines
2.0 KiB
ArmAsm
// RUN: llvm-mc -triple aarch64-none-linux-gnu -mattr=+neon,+fullfp16 -show-encoding < %s | FileCheck %s
|
|
|
|
//------------------------------------------------------------------------------
|
|
// Floating Point multiply (scalar, by element)
|
|
//------------------------------------------------------------------------------
|
|
fmul h0, h1, v1.h[5]
|
|
fmul s0, s1, v1.s[0]
|
|
fmul s30, s11, v1.s[1]
|
|
fmul s4, s5, v7.s[2]
|
|
fmul s16, s22, v16.s[3]
|
|
fmul d0, d1, v1.d[0]
|
|
fmul d30, d11, v1.d[1]
|
|
|
|
// CHECK: fmul h0, h1, v1.h[5] // encoding: [0x20,0x98,0x11,0x5f]
|
|
// CHECK: fmul s0, s1, v1.s[0] // encoding: [0x20,0x90,0x81,0x5f]
|
|
// CHECK: fmul s30, s11, v1.s[1] // encoding: [0x7e,0x91,0xa1,0x5f]
|
|
// CHECK: fmul s4, s5, v7.s[2] // encoding: [0xa4,0x98,0x87,0x5f]
|
|
// CHECK: fmul s16, s22, v16.s[3] // encoding: [0xd0,0x9a,0xb0,0x5f]
|
|
// CHECK: fmul d0, d1, v1.d[0] // encoding: [0x20,0x90,0xc1,0x5f]
|
|
// CHECK: fmul d30, d11, v1.d[1] // encoding: [0x7e,0x99,0xc1,0x5f]
|
|
|
|
|
|
//------------------------------------------------------------------------------
|
|
// Floating Point multiply extended (scalar, by element)
|
|
//------------------------------------------------------------------------------
|
|
fmulx h6, h2, v8.h[5]
|
|
fmulx s6, s2, v8.s[0]
|
|
fmulx s7, s3, v13.s[1]
|
|
fmulx s9, s7, v9.s[2]
|
|
fmulx s13, s21, v10.s[3]
|
|
fmulx d15, d9, v7.d[0]
|
|
fmulx d13, d12, v11.d[1]
|
|
|
|
// CHECK: fmulx h6, h2, v8.h[5] // encoding: [0x46,0x98,0x18,0x7f]
|
|
// CHECK: fmulx s6, s2, v8.s[0] // encoding: [0x46,0x90,0x88,0x7f]
|
|
// CHECK: fmulx s7, s3, v13.s[1] // encoding: [0x67,0x90,0xad,0x7f]
|
|
// CHECK: fmulx s9, s7, v9.s[2] // encoding: [0xe9,0x98,0x89,0x7f]
|
|
// CHECK: fmulx s13, s21, v10.s[3] // encoding: [0xad,0x9a,0xaa,0x7f]
|
|
// CHECK: fmulx d15, d9, v7.d[0] // encoding: [0x2f,0x91,0xc7,0x7f]
|
|
// CHECK: fmulx d13, d12, v11.d[1] // encoding: [0x8d,0x99,0xcb,0x7f]
|
|
|