[flang][OpenMP] Port OpenMP FIR tests for atomic update/capture to HLFIR #70627
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
NimishMishra
requested review from
kiranchandramohan,
harishch4,
chandankds and
kiranktp
llvmbot
added
flang
|
@llvm/pr-subscribers-flang-openmp @llvm/pr-subscribers-flang-fir-hlfir Author: None (NimishMishra) ChangesPort atomic update and capture tests to HLFIR. Full diff: https://github.com/llvm/llvm-project/pull/70627.diff 3 Files Affected:
diff --git a/flang/test/Lower/OpenMP/atomic-capture.f90 b/flang/test/Lower/OpenMP/atomic-capture.f90
new file mode 100644
index 000000000000000..cd4a93fd9c10aa3
--- /dev/null
+++ b/flang/test/Lower/OpenMP/atomic-capture.f90
@@ -0,0 +1,98 @@
+! This test checks the lowering of atomic capture
+
+! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-hlfir %s -o - | FileCheck %s
+! RUN: %flang_fc1 -mllvm --use-desc-for-alloc=false -emit-hlfir -fopenmp %s -o - | FileCheck %s
+
+
+
+program OmpAtomicCapture
+ use omp_lib
+
+!CHECK: %[[val_X_alloca:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
+!CHECK: %[[val_X_declare:.*]]:2 = hlfir.declare %[[val_X_alloca]] {{.*}}
+!CHECK: %[[val_Y_alloca:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
+!CHECK: %[[val_Y_declare:.*]]:2 = hlfir.declare %[[val_Y_alloca]] {{.*}}
+ integer :: x, y
+
+!CHECK: %[[val_Y_loaded:.*]] = fir.load %[[val_X_declare]]#0 : !fir.ref<i32>
+!CHECK: omp.atomic.capture hint(uncontended) {
+!CHECK: omp.atomic.update %[[val_Y_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.muli %[[val_Y_loaded]], %[[ARG]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+!CHECK: omp.atomic.read %[[val_X_declare]]#1 = %[[val_Y_declare]]#1 : !fir.ref<i32>, i32
+!CHECK: }
+ !$omp atomic hint(omp_sync_hint_uncontended) capture
+ y = x * y
+ x = y
+ !$omp end atomic
+
+!CHECK: %[[val_20:.*]] = arith.constant 20 : i32
+!CHECK: %[[val_8:.*]] = arith.constant 8 : i32
+!CHECK: %[[val_X_loaded:.*]] = fir.load %[[val_X_declare]]#0 : !fir.ref<i32>
+!CHECK: %[[sub:.*]] = arith.subi %[[val_8]], %[[val_X_loaded]] : i32
+!CHECK: %[[no_reassoc:.*]] = hlfir.no_reassoc %[[sub]] : i32
+!CHECK: %[[add:.*]] = arith.addi %[[val_20]], %[[no_reassoc]] : i32
+!CHECK: omp.atomic.capture memory_order(acquire) hint(nonspeculative) {
+!CHECK: omp.atomic.read %[[val_X_declare]]#1 = %[[val_Y_declare]]#1 : !fir.ref<i32>, i32
+!CHECK: omp.atomic.write %[[val_Y_declare]]#1 = %[[add]] : !fir.ref<i32>, i32
+!CHECK: }
+!CHECK: return
+!CHECK: }
+ !$omp atomic hint(omp_lock_hint_nonspeculative) capture acquire
+ x = y
+ y = 2 * 10 + (8 - x)
+ !$omp end atomic
+end program
+
+
+!CHECK: func.func @_QPpointers_in_atomic_capture() {
+subroutine pointers_in_atomic_capture()
+
+!CHECK: %[[val_A_alloca:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFpointers_in_atomic_captureEa"}
+!CHECK: %[[zero:.*]] = fir.zero_bits !fir.ptr<i32>
+!CHECK: %[[embox:.*]] = fir.embox %[[zero]] : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_A_alloca]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_A_declare:.*]]:2 = hlfir.declare %[[val_A_alloca]] {{.*}}
+!CHECK: %[[val_B_alloca:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "b", uniq_name = "_QFpointers_in_atomic_captureEb"}
+!CHECK: %[[zero:.*]] = fir.zero_bits !fir.ptr<i32>
+!CHECK: %[[embox:.*]] = fir.embox %[[zero]] : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_B_alloca]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_B_declare:.*]]:2 = hlfir.declare %[[val_B_alloca]] {{.*}}
+!CHECK: %[[val_C_alloca:.*]] = fir.alloca i32 {bindc_name = "c", fir.target, uniq_name = "_QFpointers_in_atomic_captureEc"}
+!CHECK: %[[val_C_declare:.*]]:2 = hlfir.declare %[[val_C_alloca]] {{.*}}
+!CHECK: %[[val_D_alloca:.*]] = fir.alloca i32 {bindc_name = "d", fir.target, uniq_name = "_QFpointers_in_atomic_captureEd"}
+!CHECK: %[[val_D_declare:.*]]:2 = hlfir.declare %[[val_D_alloca]] {{.*}}
+ integer, pointer :: a, b
+ integer, target :: c, d
+
+!CHECK: %[[embox:.*]] = fir.embox %[[val_C_declare]]#1 : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_A_declare]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[embox:.*]] = fir.embox %[[val_D_declare]]#1 : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_B_declare]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+ a=>c
+ b=>d
+
+!CHECK: %[[val_A_loaded:.*]] = fir.load %[[val_A_declare]]#0 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_A_box_addr:.*]] = fir.box_addr %[[val_A_loaded]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
+!CHECK: %[[val_B_loaded:.*]] = fir.load %[[val_B_declare]]#0 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_B_box_addr:.*]] = fir.box_addr %[[val_B_loaded]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
+!CHECK: %[[val_B_loaded_2:.*]] = fir.load %[[val_B_declare]]#0 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_B_box_addr_2:.*]] = fir.box_addr %[[val_B_loaded_2]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
+!CHECK: %[[val_B:.*]] = fir.load %[[val_B_box_addr_2]] : !fir.ptr<i32>
+!CHECK: omp.atomic.capture {
+!CHECK: omp.atomic.update %[[val_A_box_addr]] : !fir.ptr<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.addi %[[ARG]], %[[val_B]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+!CHECK: omp.atomic.read %[[val_B_box_addr]] = %[[val_A_box_addr]] : !fir.ptr<i32>, i32
+!CHECK: }
+!CHECK: return
+!CHECK: }
+ !$omp atomic capture
+ a = a + b
+ b = a
+ !$omp end atomic
+end subroutine
diff --git a/flang/test/Lower/OpenMP/atomic-update-hlfir.f90 b/flang/test/Lower/OpenMP/atomic-update-hlfir.f90
deleted file mode 100644
index 329009ab8ef8e9b..000000000000000
--- a/flang/test/Lower/OpenMP/atomic-update-hlfir.f90
+++ /dev/null
@@ -1,23 +0,0 @@
-! This test checks lowering of atomic and atomic update constructs with HLFIR
-! RUN: bbc -hlfir -fopenmp -emit-hlfir %s -o - | FileCheck %s
-! RUN: %flang_fc1 -flang-experimental-hlfir -emit-hlfir -fopenmp %s -o - | FileCheck %s
-
-subroutine sb
- integer :: x, y
-
- !$omp atomic update
- x = x + y
-end subroutine
-
-!CHECK-LABEL: @_QPsb
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFsbEx"}
-!CHECK: %[[X_DECL:.*]]:2 = hlfir.declare %[[X_REF]] {uniq_name = "_QFsbEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: %[[Y_REF:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFsbEy"}
-!CHECK: %[[Y_DECL:.*]]:2 = hlfir.declare %[[Y_REF]] {uniq_name = "_QFsbEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: %[[Y_VAL:.*]] = fir.load %[[Y_DECL]]#0 : !fir.ref<i32>
-!CHECK: omp.atomic.update %[[X_DECL]]#1 : !fir.ref<i32> {
-!CHECK: ^bb0(%[[ARG_X:.*]]: i32):
-!CHECK: %[[X_UPDATE_VAL:.*]] = arith.addi %[[ARG_X]], %[[Y_VAL]] : i32
-!CHECK: omp.yield(%[[X_UPDATE_VAL]] : i32)
-!CHECK: }
-!CHECK: return
diff --git a/flang/test/Lower/OpenMP/atomic-update.f90 b/flang/test/Lower/OpenMP/atomic-update.f90
new file mode 100644
index 000000000000000..8934695056ffa7d
--- /dev/null
+++ b/flang/test/Lower/OpenMP/atomic-update.f90
@@ -0,0 +1,149 @@
+! This test checks lowering of atomic and atomic update constructs
+! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-hlfir %s -o - | FileCheck %s
+! RUN: %flang_fc1 -mllvm --use-desc-for-alloc=false -emit-hlfir -fopenmp %s -o - | FileCheck %s
+
+program OmpAtomicUpdate
+ use omp_lib
+!CHECK: %[[val_A:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFEa"}
+!CHECK: %[[zero:.*]] = fir.zero_bits !fir.ptr<i32>
+!CHECK: %[[embox:.*]] = fir.embox %[[zero]] : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_A]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_A_declare:.*]]:2 = hlfir.declare %[[val_A]] {{.*}}
+!CHECK: %[[val_B:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "b", uniq_name = "_QFEb"}
+!CHECK: %[[zero:.*]] = fir.zero_bits !fir.ptr<i32>
+!CHECK: %[[embox:.*]] = fir.embox %[[zero]] : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_B]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_B_declare:.*]]:2 = hlfir.declare %[[val_B]] {{.*}}
+!CHECK: %[[val_C_address:.*]] = fir.address_of(@_QFEc) : !fir.ref<i32>
+!CHECK: %[[val_C_declare:.*]]:2 = hlfir.declare %[[val_C_address]] {{.*}}
+!CHECK: %[[val_D_address:.*]] = fir.address_of(@_QFEd) : !fir.ref<i32>
+!CHECK: %[[val_D_declare:.*]]:2 = hlfir.declare %[[val_D_address]] {{.}}
+!CHECK: %[[val_i1_alloca:.*]] = fir.alloca i8 {bindc_name = "i1", uniq_name = "_QFEi1"}
+!CHECK: %[[val_i1_declare:.*]]:2 = hlfir.declare %[[val_i1_alloca]] {{.*}}
+!CHECK: %[[val_c5:.*]] = arith.constant 5 : index
+!CHECK: %[[val_k_alloca:.*]] = fir.alloca !fir.array<5xi32> {bindc_name = "k", uniq_name = "_QFEk"}
+!CHECK: %[[val_k_shaped:.*]] = fir.shape %[[val_c5]] : (index) -> !fir.shape<1>
+!CHECK: %[[val_K_declare:.*]]:2 = hlfir.declare %[[val_k_alloca]](%[[val_k_shaped]]) {{.*}}
+
+!CHECK: %[[val_X_alloca:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
+!CHECK: %[[val_X_declare:.*]]:2 = hlfir.declare %[[val_X_alloca]] {uniq_name = "_QFEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK: %[[val_Y_alloca:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
+!CHECK: %[[val_Y_declare:.*]]:2 = hlfir.declare %[[val_Y_alloca]] {uniq_name = "_QFEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK: %[[val_Z_alloca:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFEz"}
+!CHECK: %[[val_Z_declare:.*]]:2 = hlfir.declare %[[val_Z_alloca]] {uniq_name = "_QFEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+ integer :: x, y, z
+ integer, pointer :: a, b
+ integer, target :: c, d
+ integer(1) :: i1
+ integer, dimension(5) :: k
+
+!CHECK: %[[embox:.*]] = fir.embox %[[val_C_declare]]#1 : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_A_declare]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[embox:.*]] = fir.embox %[[val_D_declare]]#1 : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
+!CHECK: fir.store %[[embox]] to %[[val_B_declare]]#1 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+ a=>c
+ b=>d
+
+!CHECK: %[[val_c3:.*]] = arith.constant 3 : index
+!CHECK: %[[val_K_designate:.*]] = hlfir.designate %[[val_K_declare]]#0 (%[[val_c3]]) : (!fir.ref<!fir.array<5xi32>>, index) -> !fir.ref<i32>
+!CHECK: %[[loaded_Z:.*]] = fir.load %[[val_Z_declare]]#0 : !fir.ref<i32>
+!CHECK: omp.atomic.update %[[val_K_designate]] : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.muli %[[loaded_Z]], %[[ARG]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic update
+ k(3) = z * k(3)
+
+!CHECK: %[[val_A_loaded:.*]] = fir.load %[[val_A_declare]]#0 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_A_box_addr:.*]] = fir.box_addr %[[val_A_loaded]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
+!CHECK: %[[val_B_loaded:.*]] = fir.load %[[val_B_declare]]#0 : !fir.ref<!fir.box<!fir.ptr<i32>>>
+!CHECK: %[[val_B_box_addr:.*]] = fir.box_addr %[[val_B_loaded]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
+!CHECK: %[[val_B:.*]] = fir.load %[[val_B_box_addr]] : !fir.ptr<i32>
+!CHECK: omp.atomic.update %[[val_A_box_addr]] : !fir.ptr<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.addi %[[ARG]], %[[val_B]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic update
+ a = a + b
+
+!CHECK: %[[val_c1:.*]] = arith.constant 1 : i32
+!CHECK: omp.atomic.update %[[val_Y_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.addi %[[ARG]], %[[val_c1]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic
+ y = y + 1
+
+!CHECK: %[[val_X_loaded:.*]] = fir.load %[[val_X_declare]]#0 : !fir.ref<i32>
+!CHECK: omp.atomic.update %[[val_Z_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.muli %[[val_X_loaded]], %[[ARG]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic update
+ z = x * z
+
+!CHECK: %[[val_c1:.*]] = arith.constant 1 : i32
+!CHECK: omp.atomic.update memory_order(relaxed) hint(uncontended) %[[val_X_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.subi %[[ARG]], %[[val_c1]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic relaxed update hint(omp_sync_hint_uncontended)
+ x = x - 1
+
+!CHECK: omp.atomic.update memory_order(relaxed) %[[val_Y_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[val_C_loaded:.*]] = fir.load %[[val_C_declare]]#0 : !fir.ref<i32>
+!CHECK: %[[val_D_loaded:.*]] = fir.load %[[val_D_declare]]#0 : !fir.ref<i32>
+!CHECK: {{.*}} = arith.cmpi sgt, %[[ARG]], {{.*}} : i32
+!CHECK: {{.*}} = arith.select {{.*}}, %[[ARG]], {{.*}} : i32
+!CHECK: {{.*}} = arith.cmpi sgt, {{.*}}
+!CHECK: %[[temp:.*]] = arith.select {{.*}} : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic update relaxed
+ y = max(y, c, d)
+
+!CHECK: %[[val_X_loaded:.*]] = fir.load %[[val_X_declare]]#0 : !fir.ref<i32>
+!CHECK: omp.atomic.update memory_order(relaxed) hint(contended) %[[val_Z_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.addi %[[ARG]], %[[val_X_loaded]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic relaxed hint(omp_sync_hint_contended)
+ z = z + x
+
+!CHECK: %[[val_c10:.*]] = arith.constant 10 : i32
+!CHECK: omp.atomic.update memory_order(release) hint(contended) %[[val_Z_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.muli %[[val_c10]], %[[ARG]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic release update hint(omp_lock_hint_contended)
+ z = z * 10
+
+!CHECK: %[[val_Z_loaded:.*]] = fir.load %[[val_Z_declare]]#0 : !fir.ref<i32>
+!CHECK: omp.atomic.update memory_order(release) hint(speculative) %[[val_X_declare]]#1 : !fir.ref<i32> {
+!CHECK: ^bb0(%[[ARG:.*]]: i32):
+!CHECK: %[[temp:.*]] = arith.divsi %[[ARG]], %[[val_Z_loaded]] : i32
+!CHECK: omp.yield(%[[temp]] : i32)
+!CHECK: }
+ !$omp atomic hint(omp_lock_hint_speculative) update release
+ x = x / z
+
+!CHECK: %[[val_c1:.*]] = arith.constant 1 : i32
+!CHECK: omp.atomic.update %[[val_i1_declare]]#1 : !fir.ref<i8> {
+!CHECK: ^bb0(%[[ARG:.*]]: i8):
+!CHECK: %[[convert:.*]] = fir.convert %[[ARG]] : (i8) -> i32
+!CHECK: %[[add:.*]] = arith.addi %[[convert]], %[[val_c1]] : i32
+!CHECK: %[[temp:.*]] = fir.convert %[[add]] : (i32) -> i8
+!CHECK: omp.yield(%[[temp]] : i8)
+!CHECK: }
+ !$omp atomic
+ i1 = i1 + 1
+ !$omp end atomic
+end program OmpAtomicUpdate
|
kiranchandramohan
approved these changes
Oct 30, 2023
| @@ -0,0 +1,98 @@ | |||
| ! This test checks the lowering of atomic capture | |||
|
|
|||
| ! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-hlfir %s -o - | FileCheck %s | |||
There was a problem hiding this comment.
Do you remember why --use-desc-for-alloc=false was used for these tests?
There was a problem hiding this comment.
You can go ahead and merge. No need for a reply here.
| program OmpAtomicCapture | ||
| use omp_lib | ||
|
|
||
| !CHECK: %[[val_X_alloca:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"} |
There was a problem hiding this comment.
Nit: Rename all captured variables to use CAPS for better readability.
Porting relevant tests to HLFIR.
NimishMishra
force-pushed
the
atomic-update-capture-hlfir-lowering
branch
from
5c89b29 to
bf0cfdf
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Port atomic update and capture tests to HLFIR.