mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 03:33:20 +01:00
dbac87d1fc
Patch by Ana Pazos. - Completed implementation of instruction formats: AdvSIMD three same AdvSIMD modified immediate AdvSIMD scalar pairwise - Completed implementation of instruction classes (some of the instructions in these classes belong to yet unfinished instruction formats): Vector Arithmetic Vector Immediate Vector Pairwise Arithmetic - Initial implementation of instruction formats: AdvSIMD scalar two-reg misc AdvSIMD scalar three same - Intial implementation of instruction class: Scalar Arithmetic - Initial clang changes to support arm v8 intrinsics. Note: no clang changes for scalar intrinsics function name mangling yet. - Comprehensive test cases for added instructions To verify auto codegen, encoding, decoding, diagnosis, intrinsics. llvm-svn: 187567
55 lines
2.4 KiB
LLVM
55 lines
2.4 KiB
LLVM
; RUN: llc < %s -verify-machineinstrs -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s
|
|
|
|
; Set of tests for when the intrinsic is used.
|
|
|
|
declare <2 x float> @llvm.arm.neon.vrsqrts.v2f32(<2 x float>, <2 x float>)
|
|
declare <4 x float> @llvm.arm.neon.vrsqrts.v4f32(<4 x float>, <4 x float>)
|
|
declare <2 x double> @llvm.arm.neon.vrsqrts.v2f64(<2 x double>, <2 x double>)
|
|
|
|
define <2 x float> @frsqrts_from_intr_v2f32(<2 x float> %lhs, <2 x float> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frsqrts v0.2s, v0.2s, v1.2s
|
|
%val = call <2 x float> @llvm.arm.neon.vrsqrts.v2f32(<2 x float> %lhs, <2 x float> %rhs)
|
|
ret <2 x float> %val
|
|
}
|
|
|
|
define <4 x float> @frsqrts_from_intr_v4f32(<4 x float> %lhs, <4 x float> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frsqrts v0.4s, v0.4s, v1.4s
|
|
%val = call <4 x float> @llvm.arm.neon.vrsqrts.v4f32(<4 x float> %lhs, <4 x float> %rhs)
|
|
ret <4 x float> %val
|
|
}
|
|
|
|
define <2 x double> @frsqrts_from_intr_v2f64(<2 x double> %lhs, <2 x double> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frsqrts v0.2d, v0.2d, v1.2d
|
|
%val = call <2 x double> @llvm.arm.neon.vrsqrts.v2f64(<2 x double> %lhs, <2 x double> %rhs)
|
|
ret <2 x double> %val
|
|
}
|
|
|
|
declare <2 x float> @llvm.arm.neon.vrecps.v2f32(<2 x float>, <2 x float>)
|
|
declare <4 x float> @llvm.arm.neon.vrecps.v4f32(<4 x float>, <4 x float>)
|
|
declare <2 x double> @llvm.arm.neon.vrecps.v2f64(<2 x double>, <2 x double>)
|
|
|
|
define <2 x float> @frecps_from_intr_v2f32(<2 x float> %lhs, <2 x float> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frecps v0.2s, v0.2s, v1.2s
|
|
%val = call <2 x float> @llvm.arm.neon.vrecps.v2f32(<2 x float> %lhs, <2 x float> %rhs)
|
|
ret <2 x float> %val
|
|
}
|
|
|
|
define <4 x float> @frecps_from_intr_v4f32(<4 x float> %lhs, <4 x float> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frecps v0.4s, v0.4s, v1.4s
|
|
%val = call <4 x float> @llvm.arm.neon.vrecps.v4f32(<4 x float> %lhs, <4 x float> %rhs)
|
|
ret <4 x float> %val
|
|
}
|
|
|
|
define <2 x double> @frecps_from_intr_v2f64(<2 x double> %lhs, <2 x double> %rhs) {
|
|
; Using registers other than v0, v1 are possible, but would be odd.
|
|
; CHECK: frecps v0.2d, v0.2d, v1.2d
|
|
%val = call <2 x double> @llvm.arm.neon.vrecps.v2f64(<2 x double> %lhs, <2 x double> %rhs)
|
|
ret <2 x double> %val
|
|
}
|
|
|