llvm/test/CodeGen/AArch64/sme-agnostic-za.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mattr=+sme2 < %s | FileCheck %s --check-prefixes=CHECK-COMMON,CHECK
; RUN: llc -mattr=+sme2 < %s -aarch64-new-sme-abi | FileCheck %s --check-prefixes=CHECK-COMMON,CHECK-NEWLOWERING

target triple = "aarch64"

declare i64 @private_za_decl(i64)
declare void @private_za()
declare i64 @agnostic_decl(i64) "aarch64_za_state_agnostic"

; No calls. Test that no buffer is allocated.
define i64 @agnostic_caller_no_callees(ptr %ptr) nounwind "aarch64_za_state_agnostic" {
; CHECK-COMMON-LABEL: agnostic_caller_no_callees:
; CHECK-COMMON:       // %bb.0:
; CHECK-COMMON-NEXT:    ldr x0, [x0]
; CHECK-COMMON-NEXT:    ret
  %v = load i64, ptr %ptr
  ret i64 %v
}

; agnostic-ZA -> private-ZA
;
; Test that a buffer is allocated and that the appropriate save/restore calls are
; inserted for calls to non-agnostic functions and that the arg/result registers are
; preserved by the register allocator.
define i64 @agnostic_caller_private_za_callee(i64 %v) nounwind "aarch64_za_state_agnostic" {
; CHECK-LABEL: agnostic_caller_private_za_callee:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEXT:    mov x29, sp
; CHECK-NEXT:    mov x8, x0
; CHECK-NEXT:    bl __arm_sme_state_size
; CHECK-NEXT:    sub sp, sp, x0
; CHECK-NEXT:    mov x19, sp
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    mov x0, x8
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    mov sp, x29
; CHECK-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
;
; CHECK-NEWLOWERING-LABEL: agnostic_caller_private_za_callee:
; CHECK-NEWLOWERING:       // %bb.0:
; CHECK-NEWLOWERING-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    mov x29, sp
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_state_size
; CHECK-NEWLOWERING-NEXT:    sub sp, sp, x0
; CHECK-NEWLOWERING-NEXT:    mov x19, sp
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_save
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_restore
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    mov sp, x29
; CHECK-NEWLOWERING-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ret
  %res = call i64 @private_za_decl(i64 %v)
  %res2 = call i64 @private_za_decl(i64 %res)
  ret i64 %res2
}

; agnostic-ZA -> agnostic-ZA
;
; Should not result in save/restore code.
define i64 @agnostic_caller_agnostic_callee(i64 %v) nounwind "aarch64_za_state_agnostic" {
; CHECK-COMMON-LABEL: agnostic_caller_agnostic_callee:
; CHECK-COMMON:       // %bb.0:
; CHECK-COMMON-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; CHECK-COMMON-NEXT:    bl agnostic_decl
; CHECK-COMMON-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; CHECK-COMMON-NEXT:    ret
  %res = call i64 @agnostic_decl(i64 %v)
  ret i64 %res
}

; shared-ZA -> agnostic-ZA
;
; Should not result in lazy-save or save of ZT0
define i64 @shared_caller_agnostic_callee(i64 %v) nounwind "aarch64_inout_za" "aarch64_inout_zt0" {
; CHECK-COMMON-LABEL: shared_caller_agnostic_callee:
; CHECK-COMMON:       // %bb.0:
; CHECK-COMMON-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; CHECK-COMMON-NEXT:    bl agnostic_decl
; CHECK-COMMON-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; CHECK-COMMON-NEXT:    ret
  %res = call i64 @agnostic_decl(i64 %v)
  ret i64 %res
}

; agnostic-ZA + streaming -> private-ZA + non-streaming
define i64 @streaming_agnostic_caller_nonstreaming_private_za_callee(i64 %v) nounwind "aarch64_za_state_agnostic" "aarch64_pstate_sm_enabled" {
; CHECK-LABEL: streaming_agnostic_caller_nonstreaming_private_za_callee:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp d15, d14, [sp, #-96]! // 16-byte Folded Spill
; CHECK-NEXT:    stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT:    mov x8, x0
; CHECK-NEXT:    stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT:    stp d9, d8, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT:    stp x29, x30, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT:    add x29, sp, #64
; CHECK-NEXT:    stp x20, x19, [sp, #80] // 16-byte Folded Spill
; CHECK-NEXT:    bl __arm_sme_state_size
; CHECK-NEXT:    sub sp, sp, x0
; CHECK-NEXT:    mov x20, sp
; CHECK-NEXT:    mov x0, x20
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    smstop sm
; CHECK-NEXT:    mov x0, x8
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    smstart sm
; CHECK-NEXT:    mov x0, x20
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x20
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    smstop sm
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    smstart sm
; CHECK-NEXT:    mov x0, x20
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    sub sp, x29, #64
; CHECK-NEXT:    ldp x20, x19, [sp, #80] // 16-byte Folded Reload
; CHECK-NEXT:    ldp x29, x30, [sp, #64] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d9, d8, [sp, #48] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d11, d10, [sp, #32] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d13, d12, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d15, d14, [sp], #96 // 16-byte Folded Reload
; CHECK-NEXT:    ret
;
; CHECK-NEWLOWERING-LABEL: streaming_agnostic_caller_nonstreaming_private_za_callee:
; CHECK-NEWLOWERING:       // %bb.0:
; CHECK-NEWLOWERING-NEXT:    stp d15, d14, [sp, #-96]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp d9, d8, [sp, #48] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp x29, x30, [sp, #64] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    add x29, sp, #64
; CHECK-NEWLOWERING-NEXT:    stp x20, x19, [sp, #80] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_state_size
; CHECK-NEWLOWERING-NEXT:    sub sp, sp, x0
; CHECK-NEWLOWERING-NEXT:    mov x20, sp
; CHECK-NEWLOWERING-NEXT:    mov x0, x20
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_save
; CHECK-NEWLOWERING-NEXT:    smstop sm
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    smstart sm
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    mov x0, x20
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_restore
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    sub sp, x29, #64
; CHECK-NEWLOWERING-NEXT:    ldp x20, x19, [sp, #80] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp x29, x30, [sp, #64] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d9, d8, [sp, #48] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d11, d10, [sp, #32] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d13, d12, [sp, #16] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d15, d14, [sp], #96 // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ret
  %res = call i64 @private_za_decl(i64 %v)
  %res2 = call i64 @private_za_decl(i64 %res)
  ret i64 %res2
}

; agnostic-ZA + streaming-compatible -> private-ZA + non-streaming
define i64 @streaming_compatible_agnostic_caller_nonstreaming_private_za_callee(i64 %v) nounwind "aarch64_za_state_agnostic" "aarch64_pstate_sm_compatible" {
; CHECK-LABEL: streaming_compatible_agnostic_caller_nonstreaming_private_za_callee:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp d15, d14, [sp, #-96]! // 16-byte Folded Spill
; CHECK-NEXT:    stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT:    mov x8, x0
; CHECK-NEXT:    stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT:    stp d9, d8, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT:    stp x29, x30, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT:    add x29, sp, #64
; CHECK-NEXT:    stp x20, x19, [sp, #80] // 16-byte Folded Spill
; CHECK-NEXT:    mrs x20, SVCR
; CHECK-NEXT:    bl __arm_sme_state_size
; CHECK-NEXT:    sub sp, sp, x0
; CHECK-NEXT:    mov x19, sp
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    tbz w20, #0, .LBB5_2
; CHECK-NEXT:  // %bb.1:
; CHECK-NEXT:    smstop sm
; CHECK-NEXT:  .LBB5_2:
; CHECK-NEXT:    mov x0, x8
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    tbz w20, #0, .LBB5_4
; CHECK-NEXT:  // %bb.3:
; CHECK-NEXT:    smstart sm
; CHECK-NEXT:  .LBB5_4:
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    tbz w20, #0, .LBB5_6
; CHECK-NEXT:  // %bb.5:
; CHECK-NEXT:    smstop sm
; CHECK-NEXT:  .LBB5_6:
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    bl private_za_decl
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    tbz w20, #0, .LBB5_8
; CHECK-NEXT:  // %bb.7:
; CHECK-NEXT:    smstart sm
; CHECK-NEXT:  .LBB5_8:
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    sub sp, x29, #64
; CHECK-NEXT:    ldp x20, x19, [sp, #80] // 16-byte Folded Reload
; CHECK-NEXT:    ldp x29, x30, [sp, #64] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d9, d8, [sp, #48] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d11, d10, [sp, #32] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d13, d12, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT:    ldp d15, d14, [sp], #96 // 16-byte Folded Reload
; CHECK-NEXT:    ret
;
; CHECK-NEWLOWERING-LABEL: streaming_compatible_agnostic_caller_nonstreaming_private_za_callee:
; CHECK-NEWLOWERING:       // %bb.0:
; CHECK-NEWLOWERING-NEXT:    stp d15, d14, [sp, #-96]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp d9, d8, [sp, #48] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    stp x29, x30, [sp, #64] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    add x29, sp, #64
; CHECK-NEWLOWERING-NEXT:    stp x20, x19, [sp, #80] // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_state_size
; CHECK-NEWLOWERING-NEXT:    sub sp, sp, x0
; CHECK-NEWLOWERING-NEXT:    mov x19, sp
; CHECK-NEWLOWERING-NEXT:    mrs x20, SVCR
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_save
; CHECK-NEWLOWERING-NEXT:    tbz w20, #0, .LBB5_2
; CHECK-NEWLOWERING-NEXT:  // %bb.1:
; CHECK-NEWLOWERING-NEXT:    smstop sm
; CHECK-NEWLOWERING-NEXT:  .LBB5_2:
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    bl private_za_decl
; CHECK-NEWLOWERING-NEXT:    tbz w20, #0, .LBB5_4
; CHECK-NEWLOWERING-NEXT:  // %bb.3:
; CHECK-NEWLOWERING-NEXT:    smstart sm
; CHECK-NEWLOWERING-NEXT:  .LBB5_4:
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_restore
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    sub sp, x29, #64
; CHECK-NEWLOWERING-NEXT:    ldp x20, x19, [sp, #80] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp x29, x30, [sp, #64] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d9, d8, [sp, #48] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d11, d10, [sp, #32] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d13, d12, [sp, #16] // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp d15, d14, [sp], #96 // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ret
  %res = call i64 @private_za_decl(i64 %v)
  %res2 = call i64 @private_za_decl(i64 %res)
  ret i64 %res2
}

declare i64 @many_args_private_za_callee(
  i64, i64, i64, i64, i64, i64, i64, i64, i64, i64)

; In this example some arguments are passed on the stack, which decrements the
; stack pointer before the call -- in this test the call to __arm_sme_save
; should occur _before_ the stack decrement.
define i64  @test_many_callee_arguments(
; CHECK-LABEL: test_many_callee_arguments:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEXT:    mov x29, sp
; CHECK-NEXT:    mov x8, x0
; CHECK-NEXT:    bl __arm_sme_state_size
; CHECK-NEXT:    sub sp, sp, x0
; CHECK-NEXT:    ldp x9, x10, [x29, #32]
; CHECK-NEXT:    mov x19, sp
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    stp x9, x10, [sp, #-16]!
; CHECK-NEXT:    mov x0, x8
; CHECK-NEXT:    bl many_args_private_za_callee
; CHECK-NEXT:    add sp, sp, #16
; CHECK-NEXT:    mov x1, x0
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov x0, x1
; CHECK-NEXT:    mov sp, x29
; CHECK-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
;
; CHECK-NEWLOWERING-LABEL: test_many_callee_arguments:
; CHECK-NEWLOWERING:       // %bb.0:
; CHECK-NEWLOWERING-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    mov x29, sp
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_state_size
; CHECK-NEWLOWERING-NEXT:    sub sp, sp, x0
; CHECK-NEWLOWERING-NEXT:    mov x19, sp
; CHECK-NEWLOWERING-NEXT:    ldp x9, x10, [x29, #32]
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_save
; CHECK-NEWLOWERING-NEXT:    stp x9, x10, [sp, #-16]!
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    bl many_args_private_za_callee
; CHECK-NEWLOWERING-NEXT:    add sp, sp, #16
; CHECK-NEWLOWERING-NEXT:    mov x8, x0
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_restore
; CHECK-NEWLOWERING-NEXT:    mov x0, x8
; CHECK-NEWLOWERING-NEXT:    mov sp, x29
; CHECK-NEWLOWERING-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ret
  i64 %0, i64 %1, i64 %2, i64 %3, i64 %4, i64 %5, i64 %6, i64 %7, i64 %8, i64 %9
) nounwind "aarch64_za_state_agnostic" {
  %ret = call i64 @many_args_private_za_callee(
    i64 %0, i64 %1, i64 %2, i64 %3, i64 %4, i64 %5, i64 %6, i64 %7, i64 %8, i64 %9)
  ret i64 %ret
}

define void @agnostic_za_buffer_alloc_with_stack_probes() nounwind "aarch64_za_state_agnostic" "probe-stack"="inline-asm" "stack-probe-size"="65536"{
; CHECK-LABEL: agnostic_za_buffer_alloc_with_stack_probes:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEXT:    mov x29, sp
; CHECK-NEXT:    bl __arm_sme_state_size
; CHECK-NEXT:    mov x8, sp
; CHECK-NEXT:    sub x19, x8, x0
; CHECK-NEXT:  .LBB7_1: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT:    sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEXT:    cmp sp, x19
; CHECK-NEXT:    b.le .LBB7_3
; CHECK-NEXT:  // %bb.2: // in Loop: Header=BB7_1 Depth=1
; CHECK-NEXT:    str xzr, [sp]
; CHECK-NEXT:    b .LBB7_1
; CHECK-NEXT:  .LBB7_3:
; CHECK-NEXT:    mov sp, x19
; CHECK-NEXT:    ldr xzr, [sp]
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_save
; CHECK-NEXT:    bl private_za
; CHECK-NEXT:    mov x0, x19
; CHECK-NEXT:    bl __arm_sme_restore
; CHECK-NEXT:    mov sp, x29
; CHECK-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
;
; CHECK-NEWLOWERING-LABEL: agnostic_za_buffer_alloc_with_stack_probes:
; CHECK-NEWLOWERING:       // %bb.0:
; CHECK-NEWLOWERING-NEXT:    stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    str x19, [sp, #16] // 8-byte Folded Spill
; CHECK-NEWLOWERING-NEXT:    mov x29, sp
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_state_size
; CHECK-NEWLOWERING-NEXT:    mov x8, sp
; CHECK-NEWLOWERING-NEXT:    sub x19, x8, x0
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_save
; CHECK-NEWLOWERING-NEXT:  .LBB7_1: // =>This Inner Loop Header: Depth=1
; CHECK-NEWLOWERING-NEXT:    sub sp, sp, #16, lsl #12 // =65536
; CHECK-NEWLOWERING-NEXT:    cmp sp, x19
; CHECK-NEWLOWERING-NEXT:    b.le .LBB7_3
; CHECK-NEWLOWERING-NEXT:  // %bb.2: // in Loop: Header=BB7_1 Depth=1
; CHECK-NEWLOWERING-NEXT:    str xzr, [sp]
; CHECK-NEWLOWERING-NEXT:    b .LBB7_1
; CHECK-NEWLOWERING-NEXT:  .LBB7_3:
; CHECK-NEWLOWERING-NEXT:    mov sp, x19
; CHECK-NEWLOWERING-NEXT:    ldr xzr, [sp]
; CHECK-NEWLOWERING-NEXT:    bl private_za
; CHECK-NEWLOWERING-NEXT:    mov x0, x19
; CHECK-NEWLOWERING-NEXT:    bl __arm_sme_restore
; CHECK-NEWLOWERING-NEXT:    mov sp, x29
; CHECK-NEWLOWERING-NEXT:    ldr x19, [sp, #16] // 8-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEWLOWERING-NEXT:    ret
  call void @private_za()
  ret void
}