[llvm] r359431 - [ARM] Add v4f16 and v8f16 types to the CallingConv
Diogo N. Sampaio via llvm-commits
llvm-commits at lists.llvm.org
Mon Apr 29 03:10:37 PDT 2019
Author: dnsampaio
Date: Mon Apr 29 03:10:37 2019
New Revision: 359431
URL: http://llvm.org/viewvc/llvm-project?rev=359431&view=rev
Log:
[ARM] Add v4f16 and v8f16 types to the CallingConv
Summary:
The Procedure Call Standard for the Arm Architecture
states that float16x4_t and float16x8_t behave just
as uint16x4_t and uint16x8_t for argument passing.
This patch adds the fp16 vectors to the
ARMCallingConv.td file.
Reviewers: miyuki, ostannard
Reviewed By: ostannard
Subscribers: ostannard, javed.absar, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D60720
Added:
llvm/trunk/test/CodeGen/ARM/fp16-vector-argument.ll
Modified:
llvm/trunk/lib/Target/ARM/ARMCallingConv.td
Modified: llvm/trunk/lib/Target/ARM/ARMCallingConv.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/ARM/ARMCallingConv.td?rev=359431&r1=359430&r2=359431&view=diff
==============================================================================
--- llvm/trunk/lib/Target/ARM/ARMCallingConv.td (original)
+++ llvm/trunk/lib/Target/ARM/ARMCallingConv.td Mon Apr 29 03:10:37 2019
@@ -30,8 +30,8 @@ def CC_ARM_APCS : CallingConv<[
CCIfSwiftError<CCIfType<[i32], CCAssignToReg<[R8]>>>,
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
// f64 and v2f64 are passed in adjacent GPRs, possibly split onto the stack
CCIfType<[f64, v2f64], CCCustom<"CC_ARM_APCS_Custom_f64">>,
@@ -56,8 +56,8 @@ def RetCC_ARM_APCS : CallingConv<[
CCIfSwiftError<CCIfType<[i32], CCAssignToReg<[R8]>>>,
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
CCIfType<[f64, v2f64], CCCustom<"RetCC_ARM_APCS_Custom_f64">>,
@@ -71,8 +71,8 @@ def RetCC_ARM_APCS : CallingConv<[
let Entry = 1 in
def FastCC_ARM_APCS : CallingConv<[
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
CCIfType<[v2f64], CCAssignToReg<[Q0, Q1, Q2, Q3]>>,
CCIfType<[f64], CCAssignToReg<[D0, D1, D2, D3, D4, D5, D6, D7]>>,
@@ -91,8 +91,8 @@ def FastCC_ARM_APCS : CallingConv<[
let Entry = 1 in
def RetFastCC_ARM_APCS : CallingConv<[
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
CCIfType<[v2f64], CCAssignToReg<[Q0, Q1, Q2, Q3]>>,
CCIfType<[f64], CCAssignToReg<[D0, D1, D2, D3, D4, D5, D6, D7]>>,
@@ -108,8 +108,8 @@ def RetFastCC_ARM_APCS : CallingConv<[
let Entry = 1 in
def CC_ARM_APCS_GHC : CallingConv<[
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
CCIfType<[v2f64], CCAssignToReg<[Q4, Q5]>>,
CCIfType<[f64], CCAssignToReg<[D8, D9, D10, D11]>>,
@@ -165,8 +165,8 @@ def CC_ARM_AAPCS : CallingConv<[
CCIfNest<CCAssignToReg<[R12]>>,
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
// Pass SwiftSelf in a callee saved register.
CCIfSwiftSelf<CCIfType<[i32], CCAssignToReg<[R10]>>>,
@@ -182,8 +182,8 @@ def CC_ARM_AAPCS : CallingConv<[
let Entry = 1 in
def RetCC_ARM_AAPCS : CallingConv<[
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v8f16,v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v8f16,v16i8, v4f32], CCBitConvertToType<v2f64>>,
// Pass SwiftSelf in a callee saved register.
CCIfSwiftSelf<CCIfType<[i32], CCAssignToReg<[R10]>>>,
@@ -208,8 +208,8 @@ def CC_ARM_AAPCS_VFP : CallingConv<[
CCIfByVal<CCPassByVal<4, 4>>,
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
// Pass SwiftSelf in a callee saved register.
CCIfSwiftSelf<CCIfType<[i32], CCAssignToReg<[R10]>>>,
@@ -230,8 +230,8 @@ def CC_ARM_AAPCS_VFP : CallingConv<[
let Entry = 1 in
def RetCC_ARM_AAPCS_VFP : CallingConv<[
// Handle all vector types as either f64 or v2f64.
- CCIfType<[v1i64, v2i32, v4i16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
- CCIfType<[v2i64, v4i32, v8i16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
+ CCIfType<[v1i64, v2i32, v4i16, v4f16, v4f16, v8i8, v2f32], CCBitConvertToType<f64>>,
+ CCIfType<[v2i64, v4i32, v8i16, v8f16, v8f16, v16i8, v4f32], CCBitConvertToType<v2f64>>,
// Pass SwiftSelf in a callee saved register.
CCIfSwiftSelf<CCIfType<[i32], CCAssignToReg<[R10]>>>,
Added: llvm/trunk/test/CodeGen/ARM/fp16-vector-argument.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/ARM/fp16-vector-argument.ll?rev=359431&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/ARM/fp16-vector-argument.ll (added)
+++ llvm/trunk/test/CodeGen/ARM/fp16-vector-argument.ll Mon Apr 29 03:10:37 2019
@@ -0,0 +1,226 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=armv8a -mattr=+armv8.2-a,+fullfp16,+neon -target-abi=aapcs-gnu -float-abi=soft -o - %s | FileCheck %s --check-prefix=SOFT
+; RUN: llc -mtriple=armv8a -mattr=+armv8.2-a,+fullfp16,+neon -target-abi=aapcs-gnu -float-abi=hard -o - %s | FileCheck %s --check-prefix=HARD
+; RUNTOADD: llc -mtriple=armeb-eabi -mattr=+armv8.2-a,+fullfp16,+neon -target-abi=aapcs-gnu -float-abi=soft -o - %s | FileCheck %s --check-prefix=SOFTEB
+; RUNTOADD: llc -mtriple=armeb-eabi -mattr=+armv8.2-a,+fullfp16,+neon -target-abi=aapcs-gnu -float-abi=hard -o - %s | FileCheck %s --check-prefix=HARDEB
+
+declare <4 x half> @llvm.fabs.v4f16(<4 x half>)
+declare <8 x half> @llvm.fabs.v8f16(<8 x half>)
+declare void @use(double, float, <4 x half>, i16, <8 x half>)
+define <4 x half> @test_vabs_f16(<4 x half> %a) {
+; SOFT-LABEL: test_vabs_f16:
+; SOFT: @ %bb.0: @ %entry
+; SOFT-NEXT: vmov d16, r0, r1
+; SOFT-NEXT: vabs.f16 d16, d16
+; SOFT-NEXT: vmov r0, r1, d16
+; SOFT-NEXT: bx lr
+;
+; HARD-LABEL: test_vabs_f16:
+; HARD: @ %bb.0: @ %entry
+; HARD-NEXT: vabs.f16 d0, d0
+; HARD-NEXT: bx lr
+;
+; SOFTEB-LABEL: test_vabs_f16:
+; SOFTEB: @ %bb.0: @ %entry
+; SOFTEB-NEXT: vmov d16, r1, r0
+; SOFTEB-NEXT: vrev64.16 d16, d16
+; SOFTEB-NEXT: vabs.f16 d16, d16
+; SOFTEB-NEXT: vrev64.16 d16, d16
+; SOFTEB-NEXT: vmov r1, r0, d16
+; SOFTEB-NEXT: bx lr
+;
+; HARDEB-LABEL: test_vabs_f16:
+; HARDEB: @ %bb.0: @ %entry
+; HARDEB-NEXT: vrev64.16 d16, d0
+; HARDEB-NEXT: vabs.f16 d16, d16
+; HARDEB-NEXT: vrev64.16 d0, d16
+; HARDEB-NEXT: bx lr
+entry:
+ %vabs1.i = tail call <4 x half> @llvm.fabs.v4f16(<4 x half> %a)
+ ret <4 x half> %vabs1.i
+}
+
+
+define <8 x half> @test2_vabs_f16(<8 x half> %a) {
+; SOFT-LABEL: test2_vabs_f16:
+; SOFT: @ %bb.0: @ %entry
+; SOFT-NEXT: vmov d17, r2, r3
+; SOFT-NEXT: vmov d16, r0, r1
+; SOFT-NEXT: vabs.f16 q8, q8
+; SOFT-NEXT: vmov r0, r1, d16
+; SOFT-NEXT: vmov r2, r3, d17
+; SOFT-NEXT: bx lr
+;
+; HARD-LABEL: test2_vabs_f16:
+; HARD: @ %bb.0: @ %entry
+; HARD-NEXT: vabs.f16 q0, q0
+; HARD-NEXT: bx lr
+;
+; SOFTEB-LABEL: test2_vabs_f16:
+; SOFTEB: @ %bb.0: @ %entry
+; SOFTEB-NEXT: vmov d17, r3, r2
+; SOFTEB-NEXT: vmov d16, r1, r0
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vabs.f16 q8, q8
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vmov r1, r0, d16
+; SOFTEB-NEXT: vmov r3, r2, d17
+; SOFTEB-NEXT: bx lr
+;
+; HARDEB-LABEL: test2_vabs_f16:
+; HARDEB: @ %bb.0: @ %entry
+; HARDEB-NEXT: vrev64.16 q8, q0
+; HARDEB-NEXT: vabs.f16 q8, q8
+; HARDEB-NEXT: vrev64.16 q0, q8
+; HARDEB-NEXT: bx lr
+entry:
+ %vabs1.i = tail call <8 x half> @llvm.fabs.v8f16(<8 x half> %a)
+ ret <8 x half> %vabs1.i
+}
+
+define void @test(double, float, i16, <4 x half>, <8 x half>) {
+; SOFT-LABEL: test:
+; SOFT: @ %bb.0: @ %entry
+; SOFT-NEXT: push {r11, lr}
+; SOFT-NEXT: sub sp, sp, #32
+; SOFT-NEXT: vldr d16, [sp, #40]
+; SOFT-NEXT: mov r12, #16
+; SOFT-NEXT: vabs.f16 d16, d16
+; SOFT-NEXT: mov lr, sp
+; SOFT-NEXT: vst1.16 {d16}, [lr:64], r12
+; SOFT-NEXT: add r12, sp, #48
+; SOFT-NEXT: vld1.64 {d16, d17}, [r12]
+; SOFT-NEXT: vabs.f16 q8, q8
+; SOFT-NEXT: str r3, [sp, #8]
+; SOFT-NEXT: vst1.64 {d16, d17}, [lr]
+; SOFT-NEXT: bl use
+; SOFT-NEXT: add sp, sp, #32
+; SOFT-NEXT: pop {r11, pc}
+;
+; HARD-LABEL: test:
+; HARD: @ %bb.0: @ %entry
+; HARD-NEXT: vabs.f16 q2, q2
+; HARD-NEXT: vabs.f16 d2, d2
+; HARD-NEXT: b use
+;
+; SOFTEB-LABEL: test:
+; SOFTEB: @ %bb.0: @ %entry
+; SOFTEB-NEXT: .save {r11, lr}
+; SOFTEB-NEXT: push {r11, lr}
+; SOFTEB-NEXT: .pad #32
+; SOFTEB-NEXT: sub sp, sp, #32
+; SOFTEB-NEXT: vldr d16, [sp, #40]
+; SOFTEB-NEXT: mov r12, #16
+; SOFTEB-NEXT: mov lr, sp
+; SOFTEB-NEXT: str r3, [sp, #8]
+; SOFTEB-NEXT: vrev64.16 d16, d16
+; SOFTEB-NEXT: vabs.f16 d16, d16
+; SOFTEB-NEXT: vst1.16 {d16}, [lr:64], r12
+; SOFTEB-NEXT: add r12, sp, #48
+; SOFTEB-NEXT: vld1.64 {d16, d17}, [r12]
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vabs.f16 q8, q8
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vst1.64 {d16, d17}, [lr]
+; SOFTEB-NEXT: bl use
+; SOFTEB-NEXT: add sp, sp, #32
+; SOFTEB-NEXT: pop {r11, pc}
+;
+; HARDEB-LABEL: test:
+; HARDEB: @ %bb.0: @ %entry
+; HARDEB-NEXT: vrev64.16 d16, d2
+; HARDEB-NEXT: vabs.f16 d16, d16
+; HARDEB-NEXT: vrev64.16 d2, d16
+; HARDEB-NEXT: vrev64.16 q8, q2
+; HARDEB-NEXT: vabs.f16 q8, q8
+; HARDEB-NEXT: vrev64.16 q2, q8
+; HARDEB-NEXT: b use
+entry:
+ %5 = tail call <4 x half> @llvm.fabs.v4f16(<4 x half> %3)
+ %6 = tail call <8 x half> @llvm.fabs.v8f16(<8 x half> %4)
+ tail call void @use(double %0, float %1, <4 x half> %5, i16 %2, <8 x half> %6)
+ ret void
+}
+
+define void @many_args_test(double, float, i16, <4 x half>, <8 x half>, <8 x half>, <8 x half>) {
+; SOFT-LABEL: many_args_test:
+; SOFT: @ %bb.0: @ %entry
+; SOFT-NEXT: push {r11, lr}
+; SOFT-NEXT: sub sp, sp, #32
+; SOFT-NEXT: add r12, sp, #80
+; SOFT-NEXT: mov lr, sp
+; SOFT-NEXT: vld1.64 {d16, d17}, [r12]
+; SOFT-NEXT: add r12, sp, #48
+; SOFT-NEXT: vabs.f16 q8, q8
+; SOFT-NEXT: vld1.64 {d18, d19}, [r12]
+; SOFT-NEXT: add r12, sp, #64
+; SOFT-NEXT: str r3, [sp, #8]
+; SOFT-NEXT: vadd.f16 q8, q8, q9
+; SOFT-NEXT: vld1.64 {d18, d19}, [r12]
+; SOFT-NEXT: mov r12, #16
+; SOFT-NEXT: vmul.f16 q8, q9, q8
+; SOFT-NEXT: vldr d18, [sp, #40]
+; SOFT-NEXT: vst1.16 {d18}, [lr:64], r12
+; SOFT-NEXT: vst1.64 {d16, d17}, [lr]
+; SOFT-NEXT: bl use
+; SOFT-NEXT: add sp, sp, #32
+; SOFT-NEXT: pop {r11, pc}
+;
+; HARD-LABEL: many_args_test:
+; HARD: @ %bb.0: @ %entry
+; HARD-NEXT: mov r1, sp
+; HARD-NEXT: vld1.64 {d16, d17}, [r1]
+; HARD-NEXT: vabs.f16 q8, q8
+; HARD-NEXT: vadd.f16 q8, q8, q2
+; HARD-NEXT: vmul.f16 q2, q3, q8
+; HARD-NEXT: b use
+;
+; SOFTEB-LABEL: many_args_test:
+; SOFTEB: @ %bb.0: @ %entry
+; SOFTEB-NEXT: .save {r11, lr}
+; SOFTEB-NEXT: push {r11, lr}
+; SOFTEB-NEXT: .pad #32
+; SOFTEB-NEXT: sub sp, sp, #32
+; SOFTEB-NEXT: vldr d16, [sp, #40]
+; SOFTEB-NEXT: mov r12, #16
+; SOFTEB-NEXT: mov lr, sp
+; SOFTEB-NEXT: str r3, [sp, #8]
+; SOFTEB-NEXT: vrev64.16 d16, d16
+; SOFTEB-NEXT: vst1.16 {d16}, [lr:64], r12
+; SOFTEB-NEXT: add r12, sp, #80
+; SOFTEB-NEXT: vld1.64 {d16, d17}, [r12]
+; SOFTEB-NEXT: add r12, sp, #48
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vabs.f16 q8, q8
+; SOFTEB-NEXT: vld1.64 {d18, d19}, [r12]
+; SOFTEB-NEXT: add r12, sp, #64
+; SOFTEB-NEXT: vrev64.16 q9, q9
+; SOFTEB-NEXT: vadd.f16 q8, q8, q9
+; SOFTEB-NEXT: vld1.64 {d18, d19}, [r12]
+; SOFTEB-NEXT: vrev64.16 q9, q9
+; SOFTEB-NEXT: vmul.f16 q8, q9, q8
+; SOFTEB-NEXT: vrev64.16 q8, q8
+; SOFTEB-NEXT: vst1.64 {d16, d17}, [lr]
+; SOFTEB-NEXT: bl use
+; SOFTEB-NEXT: add sp, sp, #32
+; SOFTEB-NEXT: pop {r11, pc}
+;
+; HARDEB-LABEL: many_args_test:
+; HARDEB: @ %bb.0: @ %entry
+; HARDEB-NEXT: mov r1, sp
+; HARDEB-NEXT: vld1.64 {d16, d17}, [r1]
+; HARDEB-NEXT: vrev64.16 q8, q8
+; HARDEB-NEXT: vabs.f16 q8, q8
+; HARDEB-NEXT: vrev64.16 q9, q2
+; HARDEB-NEXT: vadd.f16 q8, q8, q9
+; HARDEB-NEXT: vrev64.16 q9, q3
+; HARDEB-NEXT: vmul.f16 q8, q9, q8
+; HARDEB-NEXT: vrev64.16 q2, q8
+; HARDEB-NEXT: b use
+entry:
+ %7 = tail call <8 x half> @llvm.fabs.v8f16(<8 x half> %6)
+ %8 = fadd <8 x half> %7, %4
+ %9 = fmul <8 x half> %5, %8
+ tail call void @use(double %0, float %1, <4 x half> %3, i16 %2, <8 x half> %9)
+ ret void
+}
More information about the llvm-commits
mailing list