mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-19 11:02:59 +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
49 lines
2.1 KiB
ArmAsm
49 lines
2.1 KiB
ArmAsm
// RUN: llvm-mc -triple aarch64-none-linux-gnu -mattr=+neon,+fullfp16 -show-encoding < %s | FileCheck %s
|
|
|
|
//------------------------------------------------------------------------------
|
|
// Floating Point fused multiply-add (scalar, by element)
|
|
//------------------------------------------------------------------------------
|
|
fmla h0, h1, v1.h[5]
|
|
fmla s0, s1, v1.s[0]
|
|
fmla s30, s11, v1.s[1]
|
|
fmla s4, s5, v7.s[2]
|
|
fmla s16, s22, v16.s[3]
|
|
fmla d0, d1, v1.d[0]
|
|
fmla d30, d11, v1.d[1]
|
|
|
|
// CHECK: fmla h0, h1, v1.h[5] // encoding: [0x20,0x18,0x11,0x5f]
|
|
// CHECK: fmla s0, s1, v1.s[0] // encoding: [0x20,0x10,0x81,0x5f]
|
|
// CHECK: fmla s30, s11, v1.s[1] // encoding: [0x7e,0x11,0xa1,0x5f]
|
|
// CHECK: fmla s4, s5, v7.s[2] // encoding: [0xa4,0x18,0x87,0x5f]
|
|
// CHECK: fmla s16, s22, v16.s[3] // encoding: [0xd0,0x1a,0xb0,0x5f]
|
|
// CHECK: fmla d0, d1, v1.d[0] // encoding: [0x20,0x10,0xc1,0x5f]
|
|
// CHECK: fmla d30, d11, v1.d[1] // encoding: [0x7e,0x19,0xc1,0x5f]
|
|
|
|
//------------------------------------------------------------------------------
|
|
// Floating Point fused multiply-subtract (scalar, by element)
|
|
//------------------------------------------------------------------------------
|
|
|
|
fmls h2, h3, v4.h[5]
|
|
fmls s2, s3, v4.s[0]
|
|
fmls s29, s10, v28.s[1]
|
|
fmls s5, s12, v23.s[2]
|
|
fmls s7, s17, v26.s[3]
|
|
fmls d0, d1, v1.d[0]
|
|
fmls d30, d11, v1.d[1]
|
|
|
|
// CHECK: fmls h2, h3, v4.h[5] // encoding: [0x62,0x58,0x14,0x5f]
|
|
// CHECK: fmls s2, s3, v4.s[0] // encoding: [0x62,0x50,0x84,0x5f]
|
|
// CHECK: fmls s29, s10, v28.s[1] // encoding: [0x5d,0x51,0xbc,0x5f]
|
|
// CHECK: fmls s5, s12, v23.s[2] // encoding: [0x85,0x59,0x97,0x5f]
|
|
// CHECK: fmls s7, s17, v26.s[3] // encoding: [0x27,0x5a,0xba,0x5f]
|
|
// CHECK: fmls d0, d1, v1.d[0] // encoding: [0x20,0x50,0xc1,0x5f]
|
|
// CHECK: fmls d30, d11, v1.d[1] // encoding: [0x7e,0x59,0xc1,0x5f]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|