Files
clang-p2996/llvm/test/CodeGen/AArch64/GlobalISel/arm64-callingconv.ll
Amara Emerson 8bde5e58c0 Delay outgoing register assignments to last.
The delayed stack protector feature which is currently used for SDAG (and thus
allows for more commonly generating tail calls) depends on being able to extract
the tail call into a separate return block. To do this it also has to extract
the vreg->physreg copies that set up the call's arguments, since if it doesn't
then the call inst ends up using undefined physregs in it's new spliced block.

SelectionDAG implementations can do this because they delay emitting register
copies until  *after* the stack arguments are set up. GISel however just
processes and emits the arguments in IR order, so stack arguments always end up
last, and thus this breaks the code that looks for any register arg copies that
precede the call instruction.

This patch adds a thunk argument to the assignValueToReg() and custom assignment
hooks. For outgoing arguments, register assignments use this return param to
return a thunk that does the actual generating of the copies. We collect these
until all the outgoing stack assignments have been done and then execute them,
so that the copies (and perhaps some artifacts like G_SEXTs) are placed after
any stores.

Differential Revision: https://reviews.llvm.org/D110610
2021-10-04 12:33:20 -07:00

297 lines
14 KiB
LLVM

; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
; RUN: llc -O0 -stop-after=irtranslator -global-isel -global-isel-abort=1 -verify-machineinstrs %s -o - | FileCheck %s
target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
target triple = "aarch64-linux-gnu"
define i32 @args_i32(i32 %w0, i32 %w1, i32 %w2, i32 %w3,
; CHECK-LABEL: name: args_i32
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $w0, $w1, $w2, $w3, $w4, $w5, $w6, $w7
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $w0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $w1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $w2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s32) = COPY $w3
; CHECK-NEXT: [[COPY4:%[0-9]+]]:_(s32) = COPY $w4
; CHECK-NEXT: [[COPY5:%[0-9]+]]:_(s32) = COPY $w5
; CHECK-NEXT: [[COPY6:%[0-9]+]]:_(s32) = COPY $w6
; CHECK-NEXT: [[COPY7:%[0-9]+]]:_(s32) = COPY $w7
; CHECK-NEXT: $w0 = COPY [[COPY]](s32)
; CHECK-NEXT: RET_ReallyLR implicit $w0
i32 %w4, i32 %w5, i32 %w6, i32 %w7) {
ret i32 %w0
}
define i64 @args_i64(i64 %x0, i64 %x1, i64 %x2, i64 %x3,
; CHECK-LABEL: name: args_i64
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $x0, $x1, $x2, $x3, $x4, $x5, $x6, $x7
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s64) = COPY $x1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s64) = COPY $x3
; CHECK-NEXT: [[COPY4:%[0-9]+]]:_(s64) = COPY $x4
; CHECK-NEXT: [[COPY5:%[0-9]+]]:_(s64) = COPY $x5
; CHECK-NEXT: [[COPY6:%[0-9]+]]:_(s64) = COPY $x6
; CHECK-NEXT: [[COPY7:%[0-9]+]]:_(s64) = COPY $x7
; CHECK-NEXT: $x0 = COPY [[COPY]](s64)
; CHECK-NEXT: RET_ReallyLR implicit $x0
i64 %x4, i64 %x5, i64 %x6, i64 %x7) {
ret i64 %x0
}
define i8* @args_ptrs(i8* %x0, i16* %x1, <2 x i8>* %x2, {i8, i16, i32}* %x3,
; CHECK-LABEL: name: args_ptrs
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $x0, $x1, $x2, $x3, $x4, $x5, $x6, $x7
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(p0) = COPY $x1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(p0) = COPY $x2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(p0) = COPY $x3
; CHECK-NEXT: [[COPY4:%[0-9]+]]:_(p0) = COPY $x4
; CHECK-NEXT: [[COPY5:%[0-9]+]]:_(p0) = COPY $x5
; CHECK-NEXT: [[COPY6:%[0-9]+]]:_(p0) = COPY $x6
; CHECK-NEXT: [[COPY7:%[0-9]+]]:_(p0) = COPY $x7
; CHECK-NEXT: $x0 = COPY [[COPY]](p0)
; CHECK-NEXT: RET_ReallyLR implicit $x0
[3 x float]* %x4, double* %x5, i8* %x6, i8* %x7) {
ret i8* %x0
}
define [1 x double] @args_arr([1 x double] %d0) {
; CHECK-LABEL: name: args_arr
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $d0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY $d0
; CHECK-NEXT: $d0 = COPY [[COPY]](s64)
; CHECK-NEXT: RET_ReallyLR implicit $d0
ret [1 x double] %d0
}
declare void @varargs(i32, double, i64, ...)
define void @test_varargs() {
; CHECK-LABEL: name: test_varargs
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_FCONSTANT double 1.000000e+00
; CHECK-NEXT: [[C2:%[0-9]+]]:_(s64) = G_CONSTANT i64 12
; CHECK-NEXT: [[C3:%[0-9]+]]:_(s8) = G_CONSTANT i8 3
; CHECK-NEXT: [[C4:%[0-9]+]]:_(s16) = G_CONSTANT i16 1
; CHECK-NEXT: [[C5:%[0-9]+]]:_(s32) = G_CONSTANT i32 4
; CHECK-NEXT: [[C6:%[0-9]+]]:_(s32) = G_FCONSTANT float 1.000000e+00
; CHECK-NEXT: [[C7:%[0-9]+]]:_(s64) = G_FCONSTANT double 2.000000e+00
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s32) = G_ANYEXT [[C3]](s8)
; CHECK-NEXT: [[ANYEXT1:%[0-9]+]]:_(s32) = G_ANYEXT [[C4]](s16)
; CHECK-NEXT: $w0 = COPY [[C]](s32)
; CHECK-NEXT: $d0 = COPY [[C1]](s64)
; CHECK-NEXT: $x1 = COPY [[C2]](s64)
; CHECK-NEXT: $w2 = COPY [[ANYEXT]](s32)
; CHECK-NEXT: $w3 = COPY [[ANYEXT1]](s32)
; CHECK-NEXT: $w4 = COPY [[C5]](s32)
; CHECK-NEXT: $s1 = COPY [[C6]](s32)
; CHECK-NEXT: $d2 = COPY [[C7]](s64)
; CHECK-NEXT: BL @varargs, csr_aarch64_aapcs, implicit-def $lr, implicit $sp, implicit $w0, implicit $d0, implicit $x1, implicit $w2, implicit $w3, implicit $w4, implicit $s1, implicit $d2
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: RET_ReallyLR
call void(i32, double, i64, ...) @varargs(i32 42, double 1.0, i64 12, i8 3, i16 1, i32 4, float 1.0, double 2.0)
ret void
}
; signext/zeroext parameters on the stack: not part of any real ABI as far as I
; know, but ELF currently allocates 8 bytes for a signext parameter on the
; stack. The ADJCALLSTACK ops should reflect this, even if the difference is
; theoretical.
declare void @stack_ext_needed([8 x i64], i8 signext %in)
define void @test_stack_ext_needed() {
; CHECK-LABEL: name: test_stack_ext_needed
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s64) = G_IMPLICIT_DEF
; CHECK-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 42
; CHECK-NEXT: ADJCALLSTACKDOWN 8, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $sp
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C1]](s64)
; CHECK-NEXT: G_STORE [[C]](s8), [[PTR_ADD]](p0) :: (store (s8) into stack)
; CHECK-NEXT: $x0 = COPY [[DEF]](s64)
; CHECK-NEXT: $x1 = COPY [[DEF]](s64)
; CHECK-NEXT: $x2 = COPY [[DEF]](s64)
; CHECK-NEXT: $x3 = COPY [[DEF]](s64)
; CHECK-NEXT: $x4 = COPY [[DEF]](s64)
; CHECK-NEXT: $x5 = COPY [[DEF]](s64)
; CHECK-NEXT: $x6 = COPY [[DEF]](s64)
; CHECK-NEXT: $x7 = COPY [[DEF]](s64)
; CHECK-NEXT: BL @stack_ext_needed, csr_aarch64_aapcs, implicit-def $lr, implicit $sp, implicit $x0, implicit $x1, implicit $x2, implicit $x3, implicit $x4, implicit $x5, implicit $x6, implicit $x7
; CHECK-NEXT: ADJCALLSTACKUP 8, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: RET_ReallyLR
call void @stack_ext_needed([8 x i64] undef, i8 signext 42)
ret void
}
; Check that we can lower incoming i128 types into constituent s64 gprs.
define void @callee_s128(i128 %a, i128 %b, i128 *%ptr) {
; CHECK-LABEL: name: callee_s128
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $x0, $x1, $x2, $x3, $x4
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s64) = COPY $x1
; CHECK-NEXT: [[MV:%[0-9]+]]:_(s128) = G_MERGE_VALUES [[COPY]](s64), [[COPY1]](s64)
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s64) = COPY $x3
; CHECK-NEXT: [[MV1:%[0-9]+]]:_(s128) = G_MERGE_VALUES [[COPY2]](s64), [[COPY3]](s64)
; CHECK-NEXT: [[COPY4:%[0-9]+]]:_(p0) = COPY $x4
; CHECK-NEXT: G_STORE [[MV1]](s128), [[COPY4]](p0) :: (store (s128) into %ir.ptr)
; CHECK-NEXT: RET_ReallyLR
store i128 %b, i128 *%ptr
ret void
}
; Check we can lower outgoing s128 arguments into s64 gprs.
define void @caller_s128(i128 *%ptr) {
; CHECK-LABEL: name: caller_s128
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $x0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
; CHECK-NEXT: [[LOAD:%[0-9]+]]:_(s128) = G_LOAD [[COPY]](p0) :: (load (s128) from %ir.ptr)
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s64), [[UV1:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[LOAD]](s128)
; CHECK-NEXT: [[UV2:%[0-9]+]]:_(s64), [[UV3:%[0-9]+]]:_(s64) = G_UNMERGE_VALUES [[LOAD]](s128)
; CHECK-NEXT: $x0 = COPY [[UV]](s64)
; CHECK-NEXT: $x1 = COPY [[UV1]](s64)
; CHECK-NEXT: $x2 = COPY [[UV2]](s64)
; CHECK-NEXT: $x3 = COPY [[UV3]](s64)
; CHECK-NEXT: $x4 = COPY [[COPY]](p0)
; CHECK-NEXT: BL @callee_s128, csr_aarch64_aapcs, implicit-def $lr, implicit $sp, implicit $x0, implicit $x1, implicit $x2, implicit $x3, implicit $x4
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: RET_ReallyLR
%v = load i128, i128 *%ptr
call void @callee_s128(i128 %v, i128 %v, i128 *%ptr)
ret void
}
declare i64 @i8i16callee(i64 %a1, i64 %a2, i64 %a3, i8 signext %a4, i16 signext %a5, i64 %a6, i64 %a7, i64 %a8, i8 signext %b1, i16 signext %b2, i8 signext %b3, i8 signext %b4) nounwind readnone noinline
define i32 @i8i16caller() nounwind readnone {
; CHECK-LABEL: name: i8i16caller
; CHECK: bb.1.entry:
; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 1
; CHECK-NEXT: [[C2:%[0-9]+]]:_(s64) = G_CONSTANT i64 2
; CHECK-NEXT: [[C3:%[0-9]+]]:_(s8) = G_CONSTANT i8 3
; CHECK-NEXT: [[C4:%[0-9]+]]:_(s16) = G_CONSTANT i16 4
; CHECK-NEXT: [[C5:%[0-9]+]]:_(s64) = G_CONSTANT i64 5
; CHECK-NEXT: [[C6:%[0-9]+]]:_(s64) = G_CONSTANT i64 6
; CHECK-NEXT: [[C7:%[0-9]+]]:_(s64) = G_CONSTANT i64 7
; CHECK-NEXT: [[C8:%[0-9]+]]:_(s8) = G_CONSTANT i8 97
; CHECK-NEXT: [[C9:%[0-9]+]]:_(s16) = G_CONSTANT i16 98
; CHECK-NEXT: [[C10:%[0-9]+]]:_(s8) = G_CONSTANT i8 99
; CHECK-NEXT: [[C11:%[0-9]+]]:_(s8) = G_CONSTANT i8 100
; CHECK-NEXT: ADJCALLSTACKDOWN 32, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $sp
; CHECK-NEXT: [[C12:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C12]](s64)
; CHECK-NEXT: G_STORE [[C8]](s8), [[PTR_ADD]](p0) :: (store (s8) into stack)
; CHECK-NEXT: [[C13:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
; CHECK-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C13]](s64)
; CHECK-NEXT: G_STORE [[C9]](s16), [[PTR_ADD1]](p0) :: (store (s16) into stack + 8, align 1)
; CHECK-NEXT: [[C14:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
; CHECK-NEXT: [[PTR_ADD2:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C14]](s64)
; CHECK-NEXT: G_STORE [[C10]](s8), [[PTR_ADD2]](p0) :: (store (s8) into stack + 16)
; CHECK-NEXT: [[C15:%[0-9]+]]:_(s64) = G_CONSTANT i64 24
; CHECK-NEXT: [[PTR_ADD3:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C15]](s64)
; CHECK-NEXT: G_STORE [[C11]](s8), [[PTR_ADD3]](p0) :: (store (s8) into stack + 24)
; CHECK-NEXT: $x0 = COPY [[C]](s64)
; CHECK-NEXT: $x1 = COPY [[C1]](s64)
; CHECK-NEXT: $x2 = COPY [[C2]](s64)
; CHECK-NEXT: [[SEXT:%[0-9]+]]:_(s32) = G_SEXT [[C3]](s8)
; CHECK-NEXT: $w3 = COPY [[SEXT]](s32)
; CHECK-NEXT: [[SEXT1:%[0-9]+]]:_(s32) = G_SEXT [[C4]](s16)
; CHECK-NEXT: $w4 = COPY [[SEXT1]](s32)
; CHECK-NEXT: $x5 = COPY [[C5]](s64)
; CHECK-NEXT: $x6 = COPY [[C6]](s64)
; CHECK-NEXT: $x7 = COPY [[C7]](s64)
; CHECK-NEXT: BL @i8i16callee, csr_aarch64_aapcs, implicit-def $lr, implicit $sp, implicit $x0, implicit $x1, implicit $x2, implicit $w3, implicit $w4, implicit $x5, implicit $x6, implicit $x7, implicit-def $x0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: ADJCALLSTACKUP 32, 0, implicit-def $sp, implicit $sp
; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(s32) = G_TRUNC [[COPY1]](s64)
; CHECK-NEXT: $w0 = COPY [[TRUNC]](s32)
; CHECK-NEXT: RET_ReallyLR implicit $w0
entry:
%call = tail call i64 @i8i16callee(i64 0, i64 1, i64 2, i8 signext 3, i16 signext 4, i64 5, i64 6, i64 7, i8 97, i16 98, i8 99, i8 100)
%conv = trunc i64 %call to i32
ret i32 %conv
}
define void @arg_v2i64(<2 x i64> %arg) {
; CHECK-LABEL: name: arg_v2i64
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $q0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
; CHECK-NEXT: G_STORE [[COPY]](<2 x s64>), [[DEF]](p0) :: (store (<2 x s64>) into `<2 x i64>* undef`)
; CHECK-NEXT: RET_ReallyLR
store <2 x i64> %arg, <2 x i64>* undef
ret void
}
define void @arg_v8i64(<8 x i64> %arg) {
; CHECK-LABEL: name: arg_v8i64
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $q0, $q1, $q2, $q3
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(<2 x s64>) = COPY $q2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(<2 x s64>) = COPY $q3
; CHECK-NEXT: [[CONCAT_VECTORS:%[0-9]+]]:_(<8 x s64>) = G_CONCAT_VECTORS [[COPY]](<2 x s64>), [[COPY1]](<2 x s64>), [[COPY2]](<2 x s64>), [[COPY3]](<2 x s64>)
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
; CHECK-NEXT: G_STORE [[CONCAT_VECTORS]](<8 x s64>), [[DEF]](p0) :: (store (<8 x s64>) into `<8 x i64>* undef`)
; CHECK-NEXT: RET_ReallyLR
store <8 x i64> %arg, <8 x i64>* undef
ret void
}
define void @arg_v4f32(<4 x float> %arg) {
; CHECK-LABEL: name: arg_v4f32
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $q0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0
; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY]](<2 x s64>)
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
; CHECK-NEXT: G_STORE [[BITCAST]](<4 x s32>), [[DEF]](p0) :: (store (<4 x s32>) into `<4 x float>* undef`)
; CHECK-NEXT: RET_ReallyLR
store <4 x float> %arg, <4 x float>* undef
ret void
}
define void @ret_arg_v16f32(<16 x float> %arg) {
; CHECK-LABEL: name: ret_arg_v16f32
; CHECK: bb.1 (%ir-block.0):
; CHECK-NEXT: liveins: $q0, $q1, $q2, $q3
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(<2 x s64>) = COPY $q2
; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(<2 x s64>) = COPY $q3
; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY]](<2 x s64>)
; CHECK-NEXT: [[BITCAST1:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY1]](<2 x s64>)
; CHECK-NEXT: [[BITCAST2:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY2]](<2 x s64>)
; CHECK-NEXT: [[BITCAST3:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY3]](<2 x s64>)
; CHECK-NEXT: [[CONCAT_VECTORS:%[0-9]+]]:_(<16 x s32>) = G_CONCAT_VECTORS [[BITCAST]](<4 x s32>), [[BITCAST1]](<4 x s32>), [[BITCAST2]](<4 x s32>), [[BITCAST3]](<4 x s32>)
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
; CHECK-NEXT: G_STORE [[CONCAT_VECTORS]](<16 x s32>), [[DEF]](p0) :: (store (<16 x s32>) into `<16 x float>* undef`)
; CHECK-NEXT: RET_ReallyLR
store <16 x float> %arg, <16 x float>* undef
ret void
}