derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
226 lines
9.6 KiB
YAML
226 lines
9.6 KiB
YAML
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 6
|
|
# RUN: llc -mtriple=aarch64 -mattr=+sve -mattr=+sme -run-pass=aarch64-machine-sme-abi -verify-machineinstrs %s -o - | FileCheck %s
|
|
|
|
--- |
|
|
; Test moving a state change to be before a $nzcv def
|
|
define void @move_before_nzcv_def() "aarch64_inout_za" { ret void }
|
|
|
|
; Test moving a state change to a point where $x0 is live
|
|
define void @move_to_x0_live() "aarch64_inout_za" { ret void }
|
|
|
|
; Test we don't move before a previous state change.
|
|
define void @do_not_move_before_prior_state_change() "aarch64_za_state_agnostic" { ret void }
|
|
|
|
; Test we don't move into a call sequence.
|
|
define void @do_not_move_into_call() "aarch64_inout_za" { ret void }
|
|
|
|
declare void @clobber()
|
|
declare void @inout_call() "aarch64_inout_za"
|
|
...
|
|
---
|
|
name: move_before_nzcv_def
|
|
tracksRegLiveness: true
|
|
isSSA: true
|
|
noVRegs: false
|
|
body: |
|
|
bb.0:
|
|
|
|
; CHECK-LABEL: name: move_before_nzcv_def
|
|
; CHECK: [[RDSVLI_XI:%[0-9]+]]:gpr64 = RDSVLI_XI 1, implicit $vg
|
|
; CHECK-NEXT: [[COPY:%[0-9]+]]:gpr64 = COPY $sp
|
|
; CHECK-NEXT: [[MSUBXrrr:%[0-9]+]]:gpr64 = MSUBXrrr [[RDSVLI_XI]], [[RDSVLI_XI]], [[COPY]]
|
|
; CHECK-NEXT: $sp = COPY [[MSUBXrrr]]
|
|
; CHECK-NEXT: STPXi [[MSUBXrrr]], [[RDSVLI_XI]], %stack.0, 0
|
|
; CHECK-NEXT: [[ADDXri:%[0-9]+]]:gpr64sp = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: [[COPY1:%[0-9]+]]:gpr64 = COPY [[ADDXri]]
|
|
; CHECK-NEXT: MSR 56965, [[COPY1]]
|
|
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: RequiresZASavePseudo
|
|
; CHECK-NEXT: BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: MSRpstatesvcrImm1 2, 1, implicit-def $nzcv
|
|
; CHECK-NEXT: [[MRS:%[0-9]+]]:gpr64 = MRS 56965, implicit-def $nzcv
|
|
; CHECK-NEXT: $x0 = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: RestoreZAPseudo [[MRS]], $x0, &__arm_tpidr2_restore, csr_aarch64_sme_abi_support_routines_preservemost_from_x0
|
|
; CHECK-NEXT: MSR 56965, $xzr
|
|
; CHECK-NEXT: $nzcv = IMPLICIT_DEF
|
|
; CHECK-NEXT: $zab0 = IMPLICIT_DEF
|
|
; CHECK-NEXT: FAKE_USE $nzcv
|
|
; CHECK-NEXT: RET_ReallyLR
|
|
ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
RequiresZASavePseudo
|
|
BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
|
|
$nzcv = IMPLICIT_DEF
|
|
$zab0 = IMPLICIT_DEF
|
|
FAKE_USE $nzcv
|
|
|
|
RET_ReallyLR
|
|
...
|
|
---
|
|
name: move_to_x0_live
|
|
tracksRegLiveness: true
|
|
isSSA: true
|
|
noVRegs: false
|
|
body: |
|
|
bb.0:
|
|
|
|
; CHECK-LABEL: name: move_to_x0_live
|
|
; CHECK: [[RDSVLI_XI:%[0-9]+]]:gpr64 = RDSVLI_XI 1, implicit $vg
|
|
; CHECK-NEXT: [[COPY:%[0-9]+]]:gpr64 = COPY $sp
|
|
; CHECK-NEXT: [[MSUBXrrr:%[0-9]+]]:gpr64 = MSUBXrrr [[RDSVLI_XI]], [[RDSVLI_XI]], [[COPY]]
|
|
; CHECK-NEXT: $sp = COPY [[MSUBXrrr]]
|
|
; CHECK-NEXT: STPXi [[MSUBXrrr]], [[RDSVLI_XI]], %stack.0, 0
|
|
; CHECK-NEXT: [[ADDXri:%[0-9]+]]:gpr64sp = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: [[COPY1:%[0-9]+]]:gpr64 = COPY [[ADDXri]]
|
|
; CHECK-NEXT: MSR 56965, [[COPY1]]
|
|
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: RequiresZASavePseudo
|
|
; CHECK-NEXT: BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: MSRpstatesvcrImm1 2, 1, implicit-def $nzcv
|
|
; CHECK-NEXT: [[MRS:%[0-9]+]]:gpr64 = MRS 56965, implicit-def $nzcv
|
|
; CHECK-NEXT: $x0 = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: RestoreZAPseudo [[MRS]], $x0, &__arm_tpidr2_restore, csr_aarch64_sme_abi_support_routines_preservemost_from_x0
|
|
; CHECK-NEXT: MSR 56965, $xzr
|
|
; CHECK-NEXT: $x0 = IMPLICIT_DEF
|
|
; CHECK-NEXT: $nzcv = IMPLICIT_DEF
|
|
; CHECK-NEXT: FAKE_USE $x0
|
|
; CHECK-NEXT: $zab0 = IMPLICIT_DEF
|
|
; CHECK-NEXT: FAKE_USE $nzcv
|
|
; CHECK-NEXT: RET_ReallyLR
|
|
ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
RequiresZASavePseudo
|
|
BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
|
|
$x0 = IMPLICIT_DEF
|
|
|
|
$nzcv = IMPLICIT_DEF
|
|
FAKE_USE $x0
|
|
|
|
$zab0 = IMPLICIT_DEF
|
|
FAKE_USE $nzcv
|
|
|
|
RET_ReallyLR
|
|
...
|
|
---
|
|
name: do_not_move_before_prior_state_change
|
|
tracksRegLiveness: true
|
|
isSSA: true
|
|
noVRegs: false
|
|
body: |
|
|
; CHECK-LABEL: name: do_not_move_before_prior_state_change
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: BL &__arm_sme_state_size, csr_aarch64_sme_abi_support_routines_preservemost_from_x1, implicit-def $lr, implicit $sp, implicit-def $x0
|
|
; CHECK-NEXT: [[COPY:%[0-9]+]]:gpr64 = COPY $x0
|
|
; CHECK-NEXT: $sp = SUBXrx64 $sp, [[COPY]], 24
|
|
; CHECK-NEXT: [[COPY1:%[0-9]+]]:gpr64 = COPY $sp
|
|
; CHECK-NEXT: $nzcv = IMPLICIT_DEF
|
|
; CHECK-NEXT: $zab0 = IMPLICIT_DEF
|
|
; CHECK-NEXT: [[MRS:%[0-9]+]]:gpr64 = MRS 55824, implicit-def $nzcv, implicit $nzcv
|
|
; CHECK-NEXT: $x0 = COPY [[COPY1]]
|
|
; CHECK-NEXT: BL &__arm_sme_save, csr_aarch64_sme_abi_support_routines_preservemost_from_x1, implicit-def $lr, implicit $sp, implicit $x0
|
|
; CHECK-NEXT: MSR 55824, [[MRS]], implicit-def $nzcv
|
|
; CHECK-NEXT: Bcc 2, %bb.1, implicit $nzcv
|
|
; CHECK-NEXT: B %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: liveins: $nzcv
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: FAKE_USE $nzcv
|
|
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: RequiresZASavePseudo
|
|
; CHECK-NEXT: BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: $x0 = COPY [[COPY1]]
|
|
; CHECK-NEXT: BL &__arm_sme_restore, csr_aarch64_sme_abi_support_routines_preservemost_from_x1, implicit-def $lr, implicit $sp, implicit $x0
|
|
; CHECK-NEXT: RET_ReallyLR
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: RequiresZASavePseudo
|
|
; CHECK-NEXT: BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: $x0 = COPY [[COPY1]]
|
|
; CHECK-NEXT: BL &__arm_sme_restore, csr_aarch64_sme_abi_support_routines_preservemost_from_x1, implicit-def $lr, implicit $sp, implicit $x0
|
|
; CHECK-NEXT: RET_ReallyLR
|
|
bb.0:
|
|
successors: %bb.1, %bb.2
|
|
|
|
; The insertion point can move before the $nzcv def (as that would require
|
|
; moving before a $zab0 def -- that requires the ACTIVE state).
|
|
$nzcv = IMPLICIT_DEF
|
|
$zab0 = IMPLICIT_DEF
|
|
Bcc 2, %bb.1, implicit $nzcv
|
|
B %bb.2
|
|
; bb.1 and bb.2 both require ZA saved on entry (to force bb.0's exit bundle to
|
|
; pick the LOCAL_SAVED state).
|
|
bb.1:
|
|
liveins: $nzcv
|
|
FAKE_USE $nzcv
|
|
|
|
ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
RequiresZASavePseudo
|
|
BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
|
|
RET_ReallyLR
|
|
bb.2:
|
|
ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
RequiresZASavePseudo
|
|
BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
|
|
RET_ReallyLR
|
|
...
|
|
---
|
|
name: do_not_move_into_call
|
|
tracksRegLiveness: true
|
|
isSSA: true
|
|
noVRegs: false
|
|
body: |
|
|
bb.0:
|
|
|
|
; CHECK-LABEL: name: do_not_move_into_call
|
|
; CHECK: [[RDSVLI_XI:%[0-9]+]]:gpr64 = RDSVLI_XI 1, implicit $vg
|
|
; CHECK-NEXT: [[COPY:%[0-9]+]]:gpr64 = COPY $sp
|
|
; CHECK-NEXT: [[MSUBXrrr:%[0-9]+]]:gpr64 = MSUBXrrr [[RDSVLI_XI]], [[RDSVLI_XI]], [[COPY]]
|
|
; CHECK-NEXT: $sp = COPY [[MSUBXrrr]]
|
|
; CHECK-NEXT: STPXi [[MSUBXrrr]], [[RDSVLI_XI]], %stack.0, 0
|
|
; CHECK-NEXT: [[ADDXri:%[0-9]+]]:gpr64sp = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: [[COPY1:%[0-9]+]]:gpr64 = COPY [[ADDXri]]
|
|
; CHECK-NEXT: MSR 56965, [[COPY1]]
|
|
; CHECK-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: RequiresZASavePseudo
|
|
; CHECK-NEXT: BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
; CHECK-NEXT: $nzcv = IMPLICIT_DEF
|
|
; CHECK-NEXT: ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
; CHECK-NEXT: [[MRS:%[0-9]+]]:gpr64 = MRS 55824, implicit-def $nzcv, implicit $nzcv
|
|
; CHECK-NEXT: MSRpstatesvcrImm1 2, 1, implicit-def $nzcv
|
|
; CHECK-NEXT: [[MRS1:%[0-9]+]]:gpr64 = MRS 56965, implicit-def $nzcv
|
|
; CHECK-NEXT: $x0 = ADDXri %stack.0, 0, 0
|
|
; CHECK-NEXT: RestoreZAPseudo [[MRS1]], $x0, &__arm_tpidr2_restore, csr_aarch64_sme_abi_support_routines_preservemost_from_x0
|
|
; CHECK-NEXT: MSR 56965, $xzr
|
|
; CHECK-NEXT: MSR 55824, [[MRS]], implicit-def $nzcv
|
|
; CHECK-NEXT: $zab0 = IMPLICIT_DEF
|
|
; CHECK-NEXT: FAKE_USE $nzcv
|
|
; CHECK-NEXT: RET_ReallyLR
|
|
ADJCALLSTACKDOWN 0, 0, implicit-def dead $sp, implicit $sp
|
|
RequiresZASavePseudo
|
|
BL @clobber, csr_aarch64_aapcs, implicit-def dead $lr, implicit $sp, implicit-def $sp
|
|
|
|
; This is artificial test where NZCV is def'd inside a call, so we can't
|
|
; move the insert point before it's definition.
|
|
$nzcv = IMPLICIT_DEF
|
|
ADJCALLSTACKUP 0, 0, implicit-def dead $sp, implicit $sp
|
|
|
|
$zab0 = IMPLICIT_DEF
|
|
FAKE_USE $nzcv
|
|
|
|
RET_ReallyLR
|
|
...
|