[AArch64] Pre-commit tests for #125686 (NFC) #126643
Merged
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
|
@llvm/pr-subscribers-backend-aarch64 Author: Jonathan Thackray (jthackray) ChangesUpdate the Generate auto-tests to check both with and without Patch is 188.98 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/126643.diff 5 Files Affected:
diff --git a/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomicrmw-lsfe.ll b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomicrmw-lsfe.ll
new file mode 100644
index 000000000000000..662f4fa8c4c3a64
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/Atomics/aarch64-atomicrmw-lsfe.ll
@@ -0,0 +1,2565 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --filter-out "\b(sp)\b" --filter "^\s*(ld[^r]|st[^r]|swp|cas|bl|add|and|eor|orn|orr|sub|mvn|sxt|cmp|ccmp|csel|dmb)"
+; The base test file was generated by ./llvm/test/CodeGen/AArch64/Atomics/generate-tests.py
+; RUN: llc %s -o - -verify-machineinstrs -mtriple=aarch64 -mattr=+lsfe -O0 | FileCheck %s --check-prefixes=CHECK,-O0
+; RUN: llc %s -o - -verify-machineinstrs -mtriple=aarch64 -mattr=+lsfe -O1 | FileCheck %s --check-prefixes=CHECK,-O1
+; RUN: llc %s -o - -verify-machineinstrs -mtriple=aarch64 | FileCheck %s --check-prefixes=CHECK-no-lsfe
+
+define dso_local half @atomicrmw_fadd_half_aligned_monotonic(ptr %ptr, half %value) {
+; -O0-LABEL: atomicrmw_fadd_half_aligned_monotonic:
+; -O0: ldaxrh w0, [x9]
+; -O0: cmp w0, w10, uxth
+; -O0: stlxrh w8, w11, [x9]
+; -O0: subs w8, w8, w0, uxth
+;
+; -O1-LABEL: atomicrmw_fadd_half_aligned_monotonic:
+; -O1: ldxrh w8, [x0]
+; -O1: stxrh w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_aligned_monotonic:
+; CHECK-no-lsfe: ldxrh w8, [x0]
+; CHECK-no-lsfe: stxrh w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, half %value monotonic, align 2
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_aligned_acquire(ptr %ptr, half %value) {
+; -O0-LABEL: atomicrmw_fadd_half_aligned_acquire:
+; -O0: ldaxrh w0, [x9]
+; -O0: cmp w0, w10, uxth
+; -O0: stlxrh w8, w11, [x9]
+; -O0: subs w8, w8, w0, uxth
+;
+; -O1-LABEL: atomicrmw_fadd_half_aligned_acquire:
+; -O1: ldaxrh w8, [x0]
+; -O1: stxrh w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_aligned_acquire:
+; CHECK-no-lsfe: ldaxrh w8, [x0]
+; CHECK-no-lsfe: stxrh w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, half %value acquire, align 2
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_aligned_release(ptr %ptr, half %value) {
+; -O0-LABEL: atomicrmw_fadd_half_aligned_release:
+; -O0: ldaxrh w0, [x9]
+; -O0: cmp w0, w10, uxth
+; -O0: stlxrh w8, w11, [x9]
+; -O0: subs w8, w8, w0, uxth
+;
+; -O1-LABEL: atomicrmw_fadd_half_aligned_release:
+; -O1: ldxrh w8, [x0]
+; -O1: stlxrh w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_aligned_release:
+; CHECK-no-lsfe: ldxrh w8, [x0]
+; CHECK-no-lsfe: stlxrh w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, half %value release, align 2
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_aligned_acq_rel(ptr %ptr, half %value) {
+; -O0-LABEL: atomicrmw_fadd_half_aligned_acq_rel:
+; -O0: ldaxrh w0, [x9]
+; -O0: cmp w0, w10, uxth
+; -O0: stlxrh w8, w11, [x9]
+; -O0: subs w8, w8, w0, uxth
+;
+; -O1-LABEL: atomicrmw_fadd_half_aligned_acq_rel:
+; -O1: ldaxrh w8, [x0]
+; -O1: stlxrh w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_aligned_acq_rel:
+; CHECK-no-lsfe: ldaxrh w8, [x0]
+; CHECK-no-lsfe: stlxrh w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, half %value acq_rel, align 2
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_aligned_seq_cst(ptr %ptr, half %value) {
+; -O0-LABEL: atomicrmw_fadd_half_aligned_seq_cst:
+; -O0: ldaxrh w0, [x9]
+; -O0: cmp w0, w10, uxth
+; -O0: stlxrh w8, w11, [x9]
+; -O0: subs w8, w8, w0, uxth
+;
+; -O1-LABEL: atomicrmw_fadd_half_aligned_seq_cst:
+; -O1: ldaxrh w8, [x0]
+; -O1: stlxrh w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_aligned_seq_cst:
+; CHECK-no-lsfe: ldaxrh w8, [x0]
+; CHECK-no-lsfe: stlxrh w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, half %value seq_cst, align 2
+ ret half %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_aligned_monotonic(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_aligned_monotonic:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: ldaxrh w9, [x11]
+; -O0: cmp w9, w8, uxth
+; -O0: stlxrh w10, w12, [x11]
+; -O0: subs w8, w9, w8, uxth
+; -O0: subs w8, w8, #1
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_aligned_monotonic:
+; -O1: ldxrh w9, [x0]
+; -O1: add w9, w9, w8
+; -O1: add w9, w10, w9
+; -O1: stxrh w10, w9, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_aligned_monotonic:
+; CHECK-no-lsfe: ldxrh w9, [x0]
+; CHECK-no-lsfe: add w9, w9, w8
+; CHECK-no-lsfe: add w9, w10, w9
+; CHECK-no-lsfe: stxrh w10, w9, [x0]
+ %r = atomicrmw fadd ptr %ptr, bfloat %value monotonic, align 2
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_aligned_acquire(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_aligned_acquire:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: ldaxrh w9, [x11]
+; -O0: cmp w9, w8, uxth
+; -O0: stlxrh w10, w12, [x11]
+; -O0: subs w8, w9, w8, uxth
+; -O0: subs w8, w8, #1
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_aligned_acquire:
+; -O1: ldaxrh w9, [x0]
+; -O1: add w9, w9, w8
+; -O1: add w9, w10, w9
+; -O1: stxrh w10, w9, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_aligned_acquire:
+; CHECK-no-lsfe: ldaxrh w9, [x0]
+; CHECK-no-lsfe: add w9, w9, w8
+; CHECK-no-lsfe: add w9, w10, w9
+; CHECK-no-lsfe: stxrh w10, w9, [x0]
+ %r = atomicrmw fadd ptr %ptr, bfloat %value acquire, align 2
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_aligned_release(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_aligned_release:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: ldaxrh w9, [x11]
+; -O0: cmp w9, w8, uxth
+; -O0: stlxrh w10, w12, [x11]
+; -O0: subs w8, w9, w8, uxth
+; -O0: subs w8, w8, #1
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_aligned_release:
+; -O1: ldxrh w9, [x0]
+; -O1: add w9, w9, w8
+; -O1: add w9, w10, w9
+; -O1: stlxrh w10, w9, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_aligned_release:
+; CHECK-no-lsfe: ldxrh w9, [x0]
+; CHECK-no-lsfe: add w9, w9, w8
+; CHECK-no-lsfe: add w9, w10, w9
+; CHECK-no-lsfe: stlxrh w10, w9, [x0]
+ %r = atomicrmw fadd ptr %ptr, bfloat %value release, align 2
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_aligned_acq_rel(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_aligned_acq_rel:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: ldaxrh w9, [x11]
+; -O0: cmp w9, w8, uxth
+; -O0: stlxrh w10, w12, [x11]
+; -O0: subs w8, w9, w8, uxth
+; -O0: subs w8, w8, #1
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_aligned_acq_rel:
+; -O1: ldaxrh w9, [x0]
+; -O1: add w9, w9, w8
+; -O1: add w9, w10, w9
+; -O1: stlxrh w10, w9, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_aligned_acq_rel:
+; CHECK-no-lsfe: ldaxrh w9, [x0]
+; CHECK-no-lsfe: add w9, w9, w8
+; CHECK-no-lsfe: add w9, w10, w9
+; CHECK-no-lsfe: stlxrh w10, w9, [x0]
+ %r = atomicrmw fadd ptr %ptr, bfloat %value acq_rel, align 2
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_aligned_seq_cst(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_aligned_seq_cst:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: ldaxrh w9, [x11]
+; -O0: cmp w9, w8, uxth
+; -O0: stlxrh w10, w12, [x11]
+; -O0: subs w8, w9, w8, uxth
+; -O0: subs w8, w8, #1
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_aligned_seq_cst:
+; -O1: ldaxrh w9, [x0]
+; -O1: add w9, w9, w8
+; -O1: add w9, w10, w9
+; -O1: stlxrh w10, w9, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_aligned_seq_cst:
+; CHECK-no-lsfe: ldaxrh w9, [x0]
+; CHECK-no-lsfe: add w9, w9, w8
+; CHECK-no-lsfe: add w9, w10, w9
+; CHECK-no-lsfe: stlxrh w10, w9, [x0]
+ %r = atomicrmw fadd ptr %ptr, bfloat %value seq_cst, align 2
+ ret bfloat %r
+}
+
+define dso_local float @atomicrmw_fadd_float_aligned_monotonic(ptr %ptr, float %value) {
+; -O0-LABEL: atomicrmw_fadd_float_aligned_monotonic:
+; -O0: ldaxr w0, [x9]
+; -O0: cmp w0, w10
+; -O0: stlxr w8, w11, [x9]
+; -O0: subs w8, w0, w8
+;
+; -O1-LABEL: atomicrmw_fadd_float_aligned_monotonic:
+; -O1: ldxr w8, [x0]
+; -O1: stxr w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_aligned_monotonic:
+; CHECK-no-lsfe: ldxr w8, [x0]
+; CHECK-no-lsfe: stxr w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, float %value monotonic, align 4
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_aligned_acquire(ptr %ptr, float %value) {
+; -O0-LABEL: atomicrmw_fadd_float_aligned_acquire:
+; -O0: ldaxr w0, [x9]
+; -O0: cmp w0, w10
+; -O0: stlxr w8, w11, [x9]
+; -O0: subs w8, w0, w8
+;
+; -O1-LABEL: atomicrmw_fadd_float_aligned_acquire:
+; -O1: ldaxr w8, [x0]
+; -O1: stxr w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_aligned_acquire:
+; CHECK-no-lsfe: ldaxr w8, [x0]
+; CHECK-no-lsfe: stxr w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, float %value acquire, align 4
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_aligned_release(ptr %ptr, float %value) {
+; -O0-LABEL: atomicrmw_fadd_float_aligned_release:
+; -O0: ldaxr w0, [x9]
+; -O0: cmp w0, w10
+; -O0: stlxr w8, w11, [x9]
+; -O0: subs w8, w0, w8
+;
+; -O1-LABEL: atomicrmw_fadd_float_aligned_release:
+; -O1: ldxr w8, [x0]
+; -O1: stlxr w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_aligned_release:
+; CHECK-no-lsfe: ldxr w8, [x0]
+; CHECK-no-lsfe: stlxr w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, float %value release, align 4
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_aligned_acq_rel(ptr %ptr, float %value) {
+; -O0-LABEL: atomicrmw_fadd_float_aligned_acq_rel:
+; -O0: ldaxr w0, [x9]
+; -O0: cmp w0, w10
+; -O0: stlxr w8, w11, [x9]
+; -O0: subs w8, w0, w8
+;
+; -O1-LABEL: atomicrmw_fadd_float_aligned_acq_rel:
+; -O1: ldaxr w8, [x0]
+; -O1: stlxr w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_aligned_acq_rel:
+; CHECK-no-lsfe: ldaxr w8, [x0]
+; CHECK-no-lsfe: stlxr w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, float %value acq_rel, align 4
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_aligned_seq_cst(ptr %ptr, float %value) {
+; -O0-LABEL: atomicrmw_fadd_float_aligned_seq_cst:
+; -O0: ldaxr w0, [x9]
+; -O0: cmp w0, w10
+; -O0: stlxr w8, w11, [x9]
+; -O0: subs w8, w0, w8
+;
+; -O1-LABEL: atomicrmw_fadd_float_aligned_seq_cst:
+; -O1: ldaxr w8, [x0]
+; -O1: stlxr w9, w8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_aligned_seq_cst:
+; CHECK-no-lsfe: ldaxr w8, [x0]
+; CHECK-no-lsfe: stlxr w9, w8, [x0]
+ %r = atomicrmw fadd ptr %ptr, float %value seq_cst, align 4
+ ret float %r
+}
+
+define dso_local double @atomicrmw_fadd_double_aligned_monotonic(ptr %ptr, double %value) {
+; -O0-LABEL: atomicrmw_fadd_double_aligned_monotonic:
+; -O0: ldaxr x0, [x9]
+; -O0: cmp x0, x10
+; -O0: stlxr w8, x11, [x9]
+; -O0: subs x8, x0, x8
+;
+; -O1-LABEL: atomicrmw_fadd_double_aligned_monotonic:
+; -O1: ldxr x8, [x0]
+; -O1: stxr w9, x8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_double_aligned_monotonic:
+; CHECK-no-lsfe: ldxr x8, [x0]
+; CHECK-no-lsfe: stxr w9, x8, [x0]
+ %r = atomicrmw fadd ptr %ptr, double %value monotonic, align 8
+ ret double %r
+}
+
+define dso_local double @atomicrmw_fadd_double_aligned_acquire(ptr %ptr, double %value) {
+; -O0-LABEL: atomicrmw_fadd_double_aligned_acquire:
+; -O0: ldaxr x0, [x9]
+; -O0: cmp x0, x10
+; -O0: stlxr w8, x11, [x9]
+; -O0: subs x8, x0, x8
+;
+; -O1-LABEL: atomicrmw_fadd_double_aligned_acquire:
+; -O1: ldaxr x8, [x0]
+; -O1: stxr w9, x8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_double_aligned_acquire:
+; CHECK-no-lsfe: ldaxr x8, [x0]
+; CHECK-no-lsfe: stxr w9, x8, [x0]
+ %r = atomicrmw fadd ptr %ptr, double %value acquire, align 8
+ ret double %r
+}
+
+define dso_local double @atomicrmw_fadd_double_aligned_release(ptr %ptr, double %value) {
+; -O0-LABEL: atomicrmw_fadd_double_aligned_release:
+; -O0: ldaxr x0, [x9]
+; -O0: cmp x0, x10
+; -O0: stlxr w8, x11, [x9]
+; -O0: subs x8, x0, x8
+;
+; -O1-LABEL: atomicrmw_fadd_double_aligned_release:
+; -O1: ldxr x8, [x0]
+; -O1: stlxr w9, x8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_double_aligned_release:
+; CHECK-no-lsfe: ldxr x8, [x0]
+; CHECK-no-lsfe: stlxr w9, x8, [x0]
+ %r = atomicrmw fadd ptr %ptr, double %value release, align 8
+ ret double %r
+}
+
+define dso_local double @atomicrmw_fadd_double_aligned_acq_rel(ptr %ptr, double %value) {
+; -O0-LABEL: atomicrmw_fadd_double_aligned_acq_rel:
+; -O0: ldaxr x0, [x9]
+; -O0: cmp x0, x10
+; -O0: stlxr w8, x11, [x9]
+; -O0: subs x8, x0, x8
+;
+; -O1-LABEL: atomicrmw_fadd_double_aligned_acq_rel:
+; -O1: ldaxr x8, [x0]
+; -O1: stlxr w9, x8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_double_aligned_acq_rel:
+; CHECK-no-lsfe: ldaxr x8, [x0]
+; CHECK-no-lsfe: stlxr w9, x8, [x0]
+ %r = atomicrmw fadd ptr %ptr, double %value acq_rel, align 8
+ ret double %r
+}
+
+define dso_local double @atomicrmw_fadd_double_aligned_seq_cst(ptr %ptr, double %value) {
+; -O0-LABEL: atomicrmw_fadd_double_aligned_seq_cst:
+; -O0: ldaxr x0, [x9]
+; -O0: cmp x0, x10
+; -O0: stlxr w8, x11, [x9]
+; -O0: subs x8, x0, x8
+;
+; -O1-LABEL: atomicrmw_fadd_double_aligned_seq_cst:
+; -O1: ldaxr x8, [x0]
+; -O1: stlxr w9, x8, [x0]
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_double_aligned_seq_cst:
+; CHECK-no-lsfe: ldaxr x8, [x0]
+; CHECK-no-lsfe: stlxr w9, x8, [x0]
+ %r = atomicrmw fadd ptr %ptr, double %value seq_cst, align 8
+ ret double %r
+}
+
+define dso_local half @atomicrmw_fadd_half_unaligned_monotonic(ptr %ptr, half %value) {
+; CHECK-LABEL: atomicrmw_fadd_half_unaligned_monotonic:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_unaligned_monotonic:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, half %value monotonic, align 1
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_unaligned_acquire(ptr %ptr, half %value) {
+; CHECK-LABEL: atomicrmw_fadd_half_unaligned_acquire:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_unaligned_acquire:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, half %value acquire, align 1
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_unaligned_release(ptr %ptr, half %value) {
+; CHECK-LABEL: atomicrmw_fadd_half_unaligned_release:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_unaligned_release:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, half %value release, align 1
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_unaligned_acq_rel(ptr %ptr, half %value) {
+; CHECK-LABEL: atomicrmw_fadd_half_unaligned_acq_rel:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_unaligned_acq_rel:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, half %value acq_rel, align 1
+ ret half %r
+}
+
+define dso_local half @atomicrmw_fadd_half_unaligned_seq_cst(ptr %ptr, half %value) {
+; CHECK-LABEL: atomicrmw_fadd_half_unaligned_seq_cst:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_half_unaligned_seq_cst:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, half %value seq_cst, align 1
+ ret half %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_unaligned_monotonic(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_unaligned_monotonic:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: bl __atomic_compare_exchange
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_unaligned_monotonic:
+; -O1: add w8, w8, w20
+; -O1: add w8, w9, w8
+; -O1: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_unaligned_monotonic:
+; CHECK-no-lsfe: add w8, w8, w20
+; CHECK-no-lsfe: add w8, w9, w8
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, bfloat %value monotonic, align 1
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_unaligned_acquire(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_unaligned_acquire:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: bl __atomic_compare_exchange
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_unaligned_acquire:
+; -O1: add w8, w8, w20
+; -O1: add w8, w9, w8
+; -O1: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_unaligned_acquire:
+; CHECK-no-lsfe: add w8, w8, w20
+; CHECK-no-lsfe: add w8, w9, w8
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, bfloat %value acquire, align 1
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_unaligned_release(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_unaligned_release:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: bl __atomic_compare_exchange
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_unaligned_release:
+; -O1: add w8, w8, w20
+; -O1: add w8, w9, w8
+; -O1: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_unaligned_release:
+; CHECK-no-lsfe: add w8, w8, w20
+; CHECK-no-lsfe: add w8, w9, w8
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, bfloat %value release, align 1
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_unaligned_acq_rel(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_unaligned_acq_rel:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: bl __atomic_compare_exchange
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_unaligned_acq_rel:
+; -O1: add w8, w8, w20
+; -O1: add w8, w9, w8
+; -O1: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_unaligned_acq_rel:
+; CHECK-no-lsfe: add w8, w8, w20
+; CHECK-no-lsfe: add w8, w9, w8
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, bfloat %value acq_rel, align 1
+ ret bfloat %r
+}
+
+define dso_local bfloat @atomicrmw_fadd_bfloat_unaligned_seq_cst(ptr %ptr, bfloat %value) {
+; -O0-LABEL: atomicrmw_fadd_bfloat_unaligned_seq_cst:
+; -O0: add w8, w8, w9
+; -O0: add w8, w8, w9
+; -O0: bl __atomic_compare_exchange
+;
+; -O1-LABEL: atomicrmw_fadd_bfloat_unaligned_seq_cst:
+; -O1: add w8, w8, w20
+; -O1: add w8, w9, w8
+; -O1: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_bfloat_unaligned_seq_cst:
+; CHECK-no-lsfe: add w8, w8, w20
+; CHECK-no-lsfe: add w8, w9, w8
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, bfloat %value seq_cst, align 1
+ ret bfloat %r
+}
+
+define dso_local float @atomicrmw_fadd_float_unaligned_monotonic(ptr %ptr, float %value) {
+; CHECK-LABEL: atomicrmw_fadd_float_unaligned_monotonic:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_unaligned_monotonic:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, float %value monotonic, align 1
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_unaligned_acquire(ptr %ptr, float %value) {
+; CHECK-LABEL: atomicrmw_fadd_float_unaligned_acquire:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_unaligned_acquire:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, float %value acquire, align 1
+ ret float %r
+}
+
+define dso_local float @atomicrmw_fadd_float_unaligned_release(ptr %ptr, float %value) {
+; CHECK-LABEL: atomicrmw_fadd_float_unaligned_release:
+; CHECK: bl __atomic_compare_exchange
+;
+; CHECK-no-lsfe-LABEL: atomicrmw_fadd_float_unaligned_release:
+; CHECK-no-lsfe: bl __atomic_compare_exchange
+ %r = atomicrmw fadd ptr %ptr, float %value release, align 1
+ ret float %r
+}
+
+define dso_local float @atomicrmw_...
[truncated]
|
|
✅ With the latest revision this PR passed the Python code formatter. |
Update the `generate-tests.py` script to create new tests for
`atomicrmw {fadd,fmin,fmax}` and test these with `half`, `float`,
`bfloat` and `double`.
Generate auto-tests to check both with and without `+lsfe`, so
that when llvm#125868 is merged, `+lsfe` will use a single atomic
floating-point instruction.
jthackray
force-pushed
the
pre-commit-tests
branch
from
e4399f2 to
3c897b1
Compare
jthackray
requested review from
tmatheson-arm,
CarolineConcatto,
kmclaughlin-arm,
Stylie777,
Lukacma and
rgwott
jthackray
force-pushed
the
pre-commit-tests
branch
from
ec88dda to
5bf38e7
Compare
tmatheson-arm
approved these changes
Feb 12, 2025
flovent
pushed a commit
to flovent/llvm-project
that referenced
this pull request
Feb 13, 2025
Update the `generate-tests.py` script to create new tests for `atomicrmw
{fadd,fmin,fmax}` and test these with `half`, `float`, `bfloat` and
`double`.
Generate fp auto-tests to check both with and without `+lsfe`, so that when
llvm#125686 is merged, `+lsfe` will use a single atomic floating-point
instruction.
joaosaffran
pushed a commit
to joaosaffran/llvm-project
that referenced
this pull request
Feb 14, 2025
Update the `generate-tests.py` script to create new tests for `atomicrmw
{fadd,fmin,fmax}` and test these with `half`, `float`, `bfloat` and
`double`.
Generate fp auto-tests to check both with and without `+lsfe`, so that when
llvm#125686 is merged, `+lsfe` will use a single atomic floating-point
instruction.
sivan-shani
pushed a commit
to sivan-shani/llvm-project
that referenced
this pull request
Feb 24, 2025
Update the `generate-tests.py` script to create new tests for `atomicrmw
{fadd,fmin,fmax}` and test these with `half`, `float`, `bfloat` and
`double`.
Generate fp auto-tests to check both with and without `+lsfe`, so that when
llvm#125686 is merged, `+lsfe` will use a single atomic floating-point
instruction.
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.
Update the
generate-tests.pyscript to create new tests foratomicrmw {fadd,fmin,fmax}and test these withhalf,float,bfloatanddouble.Generate auto-tests to check both with and without
+lsfe, so that when #125686 is merged,+lsfewill use a single atomic floating-point instruction.