mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2025-02-01 05:01:59 +01:00
887a76b062
In general SVE intrinsics are considered predicated and merging with everything else having suitable decoration. For predicated zeroing operations (like the predicate logical instructions) we use the "_z" suffix. After this change all intrinsics use their expected names (i.e. orr instead of or and eor instead of xor). I've removed intrinsics and patterns for condition code setting instructions as that data is not returned as part of the intrinsic. The expectation is to ask for a cc flag explicitly. For example: a = and_z(pg, p1, p2) cc = ptest_<flag>(pg, a) With the code generator expected to use "s" variants of instructions when available. Differential Revision: https://reviews.llvm.org/D71715
180 lines
9.4 KiB
LLVM
180 lines
9.4 KiB
LLVM
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s
|
|
|
|
define <vscale x 16 x i8> @and_pred_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
|
|
; CHECK-LABEL: and_pred_i8:
|
|
; CHECK: and z0.b, p0/m, z0.b, z1.b
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.and.nxv2i8(<vscale x 16 x i1> %pg,
|
|
<vscale x 16 x i8> %a,
|
|
<vscale x 16 x i8> %b)
|
|
ret <vscale x 16 x i8> %out
|
|
}
|
|
|
|
define <vscale x 8 x i16> @and_pred_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
|
|
; CHECK-LABEL: and_pred_i16:
|
|
; CHECK: and z0.h, p0/m, z0.h, z1.h
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.and.nxv2i16(<vscale x 8 x i1> %pg,
|
|
<vscale x 8 x i16> %a,
|
|
<vscale x 8 x i16> %b)
|
|
ret <vscale x 8 x i16> %out
|
|
}
|
|
|
|
define <vscale x 4 x i32> @and_pred_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
|
|
; CHECK-LABEL: and_pred_i32:
|
|
; CHECK: and z0.s, p0/m, z0.s, z1.s
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.and.nxv2i32(<vscale x 4 x i1> %pg,
|
|
<vscale x 4 x i32> %a,
|
|
<vscale x 4 x i32> %b)
|
|
ret <vscale x 4 x i32> %out
|
|
}
|
|
|
|
define <vscale x 2 x i64> @and_pred_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
|
|
; CHECK-LABEL: and_pred_i64:
|
|
; CHECK: and z0.d, p0/m, z0.d, z1.d
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.and.nxv2i64(<vscale x 2 x i1> %pg,
|
|
<vscale x 2 x i64> %a,
|
|
<vscale x 2 x i64> %b)
|
|
ret <vscale x 2 x i64> %out
|
|
}
|
|
|
|
define <vscale x 16 x i8> @or_pred_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
|
|
; CHECK-LABEL: or_pred_i8:
|
|
; CHECK: orr z0.b, p0/m, z0.b, z1.b
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.orr.nxv2i8(<vscale x 16 x i1> %pg,
|
|
<vscale x 16 x i8> %a,
|
|
<vscale x 16 x i8> %b)
|
|
ret <vscale x 16 x i8> %out
|
|
}
|
|
|
|
define <vscale x 8 x i16> @or_pred_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
|
|
; CHECK-LABEL: or_pred_i16:
|
|
; CHECK: orr z0.h, p0/m, z0.h, z1.h
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.orr.nxv2i16(<vscale x 8 x i1> %pg,
|
|
<vscale x 8 x i16> %a,
|
|
<vscale x 8 x i16> %b)
|
|
ret <vscale x 8 x i16> %out
|
|
}
|
|
|
|
define <vscale x 4 x i32> @or_pred_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
|
|
; CHECK-LABEL: or_pred_i32:
|
|
; CHECK: orr z0.s, p0/m, z0.s, z1.s
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.orr.nxv2i32(<vscale x 4 x i1> %pg,
|
|
<vscale x 4 x i32> %a,
|
|
<vscale x 4 x i32> %b)
|
|
ret <vscale x 4 x i32> %out
|
|
}
|
|
|
|
define <vscale x 2 x i64> @or_pred_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
|
|
; CHECK-LABEL: or_pred_i64:
|
|
; CHECK: orr z0.d, p0/m, z0.d, z1.d
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.orr.nxv2i64(<vscale x 2 x i1> %pg,
|
|
<vscale x 2 x i64> %a,
|
|
<vscale x 2 x i64> %b)
|
|
ret <vscale x 2 x i64> %out
|
|
}
|
|
|
|
define <vscale x 16 x i8> @xor_pred_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
|
|
; CHECK-LABEL: xor_pred_i8:
|
|
; CHECK: eor z0.b, p0/m, z0.b, z1.b
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.eor.nxv2i8(<vscale x 16 x i1> %pg,
|
|
<vscale x 16 x i8> %a,
|
|
<vscale x 16 x i8> %b)
|
|
ret <vscale x 16 x i8> %out
|
|
}
|
|
|
|
define <vscale x 8 x i16> @xor_pred_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
|
|
; CHECK-LABEL: xor_pred_i16:
|
|
; CHECK: eor z0.h, p0/m, z0.h, z1.h
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.eor.nxv2i16(<vscale x 8 x i1> %pg,
|
|
<vscale x 8 x i16> %a,
|
|
<vscale x 8 x i16> %b)
|
|
ret <vscale x 8 x i16> %out
|
|
}
|
|
|
|
define <vscale x 4 x i32> @xor_pred_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
|
|
; CHECK-LABEL: xor_pred_i32:
|
|
; CHECK: eor z0.s, p0/m, z0.s, z1.s
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.eor.nxv2i32(<vscale x 4 x i1> %pg,
|
|
<vscale x 4 x i32> %a,
|
|
<vscale x 4 x i32> %b)
|
|
ret <vscale x 4 x i32> %out
|
|
}
|
|
|
|
define <vscale x 2 x i64> @xor_pred_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
|
|
; CHECK-LABEL: xor_pred_i64:
|
|
; CHECK: eor z0.d, p0/m, z0.d, z1.d
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.eor.nxv2i64(<vscale x 2 x i1> %pg,
|
|
<vscale x 2 x i64> %a,
|
|
<vscale x 2 x i64> %b)
|
|
ret <vscale x 2 x i64> %out
|
|
}
|
|
|
|
define <vscale x 16 x i8> @bic_pred_i8(<vscale x 16 x i1> %pg, <vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
|
|
; CHECK-LABEL: bic_pred_i8:
|
|
; CHECK: bic z0.b, p0/m, z0.b, z1.b
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 16 x i8> @llvm.aarch64.sve.bic.nxv2i8(<vscale x 16 x i1> %pg,
|
|
<vscale x 16 x i8> %a,
|
|
<vscale x 16 x i8> %b)
|
|
ret <vscale x 16 x i8> %out
|
|
}
|
|
|
|
define <vscale x 8 x i16> @bic_pred_i16(<vscale x 8 x i1> %pg, <vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
|
|
; CHECK-LABEL: bic_pred_i16:
|
|
; CHECK: bic z0.h, p0/m, z0.h, z1.h
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 8 x i16> @llvm.aarch64.sve.bic.nxv2i16(<vscale x 8 x i1> %pg,
|
|
<vscale x 8 x i16> %a,
|
|
<vscale x 8 x i16> %b)
|
|
ret <vscale x 8 x i16> %out
|
|
}
|
|
|
|
|
|
define <vscale x 4 x i32> @bic_pred_i32(<vscale x 4 x i1> %pg, <vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
|
|
; CHECK-LABEL: bic_pred_i32:
|
|
; CHECK: bic z0.s, p0/m, z0.s, z1.s
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 4 x i32> @llvm.aarch64.sve.bic.nxv2i32(<vscale x 4 x i1> %pg,
|
|
<vscale x 4 x i32> %a,
|
|
<vscale x 4 x i32> %b)
|
|
ret <vscale x 4 x i32> %out
|
|
}
|
|
|
|
define <vscale x 2 x i64> @bic_pred_i64(<vscale x 2 x i1> %pg, <vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
|
|
; CHECK-LABEL: bic_pred_i64:
|
|
; CHECK: bic z0.d, p0/m, z0.d, z1.d
|
|
; CHECK-NEXT: ret
|
|
%out = call <vscale x 2 x i64> @llvm.aarch64.sve.bic.nxv2i64(<vscale x 2 x i1> %pg,
|
|
<vscale x 2 x i64> %a,
|
|
<vscale x 2 x i64> %b)
|
|
ret <vscale x 2 x i64> %out
|
|
}
|
|
|
|
declare <vscale x 16 x i8> @llvm.aarch64.sve.and.nxv2i8(<vscale x 16 x i1>,<vscale x 16 x i8>,<vscale x 16 x i8>)
|
|
declare <vscale x 8 x i16> @llvm.aarch64.sve.and.nxv2i16(<vscale x 8 x i1>,<vscale x 8 x i16>,<vscale x 8 x i16>)
|
|
declare <vscale x 4 x i32> @llvm.aarch64.sve.and.nxv2i32(<vscale x 4 x i1>,<vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare <vscale x 2 x i64> @llvm.aarch64.sve.and.nxv2i64(<vscale x 2 x i1>,<vscale x 2 x i64>,<vscale x 2 x i64>)
|
|
declare <vscale x 16 x i8> @llvm.aarch64.sve.orr.nxv2i8(<vscale x 16 x i1>,<vscale x 16 x i8>,<vscale x 16 x i8>)
|
|
declare <vscale x 8 x i16> @llvm.aarch64.sve.orr.nxv2i16(<vscale x 8 x i1>,<vscale x 8 x i16>,<vscale x 8 x i16>)
|
|
declare <vscale x 4 x i32> @llvm.aarch64.sve.orr.nxv2i32(<vscale x 4 x i1>,<vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare <vscale x 2 x i64> @llvm.aarch64.sve.orr.nxv2i64(<vscale x 2 x i1>,<vscale x 2 x i64>,<vscale x 2 x i64>)
|
|
declare <vscale x 16 x i8> @llvm.aarch64.sve.eor.nxv2i8(<vscale x 16 x i1>,<vscale x 16 x i8>,<vscale x 16 x i8>)
|
|
declare <vscale x 8 x i16> @llvm.aarch64.sve.eor.nxv2i16(<vscale x 8 x i1>,<vscale x 8 x i16>,<vscale x 8 x i16>)
|
|
declare <vscale x 4 x i32> @llvm.aarch64.sve.eor.nxv2i32(<vscale x 4 x i1>,<vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare <vscale x 2 x i64> @llvm.aarch64.sve.eor.nxv2i64(<vscale x 2 x i1>,<vscale x 2 x i64>,<vscale x 2 x i64>)
|
|
declare <vscale x 16 x i8> @llvm.aarch64.sve.bic.nxv2i8(<vscale x 16 x i1>,<vscale x 16 x i8>,<vscale x 16 x i8>)
|
|
declare <vscale x 8 x i16> @llvm.aarch64.sve.bic.nxv2i16(<vscale x 8 x i1>,<vscale x 8 x i16>,<vscale x 8 x i16>)
|
|
declare <vscale x 4 x i32> @llvm.aarch64.sve.bic.nxv2i32(<vscale x 4 x i1>,<vscale x 4 x i32>,<vscale x 4 x i32>)
|
|
declare <vscale x 2 x i64> @llvm.aarch64.sve.bic.nxv2i64(<vscale x 2 x i1>,<vscale x 2 x i64>,<vscale x 2 x i64>)
|