[clang] [llvm] [AArch64] Make apple-m4 armv8.7-a again (from armv9.2-a). (PR #106312)
Ahmed Bougacha via cfe-commits
cfe-commits at lists.llvm.org
Tue Aug 27 16:55:55 PDT 2024
https://github.com/ahmedbougacha created https://github.com/llvm/llvm-project/pull/106312
This is a partial revert of c66e1d6f3429. Even though that
allowed us to declare v9.2-a support without picking up SVE2
in both the backend and the driver, the frontend itself still
enabled SVE via the arch version's default extensions.
Avoid that by reverting back to v8.7-a while we look into
longer-term solutions.
>From 148a1199e6b6750cdbc467c064d6a0de348b0cf2 Mon Sep 17 00:00:00 2001
From: Ahmed Bougacha <ahmed at bougacha.org>
Date: Tue, 27 Aug 2024 16:48:35 -0700
Subject: [PATCH 1/2] [AArch64] Add incorrect target(cpu=apple-m4) fn attr
test. NFC.
This has SVE but isn't supposed to.
---
clang/test/CodeGen/aarch64-targetattr.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/clang/test/CodeGen/aarch64-targetattr.c b/clang/test/CodeGen/aarch64-targetattr.c
index d6227be2ebef83..79d49e522d3e1d 100644
--- a/clang/test/CodeGen/aarch64-targetattr.c
+++ b/clang/test/CodeGen/aarch64-targetattr.c
@@ -191,6 +191,14 @@ __attribute__((target("no-v9.3a")))
//
void minusarch() {}
+__attribute__((target("cpu=apple-m4")))
+// CHECK-LABEL: define {{[^@]+}}@applem4
+// CHECK-SAME: () #[[ATTR18:[0-9]+]] {
+// CHECK-NEXT: entry:
+// CHECK-NEXT: ret void
+//
+void applem4() {}
+
//.
// CHECK: attributes #[[ATTR0]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-features"="+crc,+fp-armv8,+lse,+neon,+ras,+rdm,+v8.1a,+v8.2a,+v8a" }
// CHECK: attributes #[[ATTR1]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-features"="+crc,+fp-armv8,+fullfp16,+lse,+neon,+ras,+rdm,+sve,+v8.1a,+v8.2a,+v8a" }
@@ -210,6 +218,7 @@ void minusarch() {}
// CHECK: attributes #[[ATTR15]] = { noinline nounwind optnone "branch-target-enforcement" "guarded-control-stack" "no-trapping-math"="true" "sign-return-address"="non-leaf" "sign-return-address-key"="a_key" "stack-protector-buffer-size"="8" "target-cpu"="neoverse-n1" "target-features"="+aes,+bf16,+bti,+ccidx,+complxnum,+crc,+dit,+dotprod,+flagm,+fp-armv8,+fullfp16,+i8mm,+jsconv,+lse,+neon,+pauth,+perfmon,+predres,+ras,+rcpc,+rdm,+sb,+sha2,+spe,+ssbs,+sve,+sve2,+v8.1a,+v8.2a,+v8.3a,+v8.4a,+v8.5a,+v8.6a,+v8a" "tune-cpu"="cortex-a710" }
// CHECK: attributes #[[ATTR16]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" }
// CHECK: attributes #[[ATTR17]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-features"="-v9.3a" }
+// CHECK: attributes #[[ATTR18]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="apple-m4" "target-features"="+aes,+bf16,+bti,+ccidx,+complxnum,+crc,+dit,+dotprod,+flagm,+fp-armv8,+fp16fml,+fpac,+fullfp16,+i8mm,+jsconv,+lse,+mec,+neon,+pauth,+perfmon,+predres,+ras,+rcpc,+rdm,+rme,+sb,+sha2,+sha3,+sme,+sme-f64f64,+sme-i16i64,+sme2,+ssbs,+sve,+sve2,+v8.1a,+v8.2a,+v8.3a,+v8.4a,+v8.5a,+v8.6a,+v8.7a,+v8a,+v9.1a,+v9.2a,+v9a,+wfxt" }
//.
// CHECK: [[META0:![0-9]+]] = !{i32 1, !"wchar_size", i32 4}
// CHECK: [[META1:![0-9]+]] = !{!"{{.*}}clang version {{.*}}"}
>From 0bec083fa59f52ccd7ba3bfe1cab637a892550d3 Mon Sep 17 00:00:00 2001
From: Ahmed Bougacha <ahmed at bougacha.org>
Date: Tue, 27 Aug 2024 16:51:16 -0700
Subject: [PATCH 2/2] [AArch64] Make apple-m4 armv8.7-a again (from armv9.2-a).
This is a partial revert of c66e1d6f3429. Even though that
allowed us to declare v9.2-a support without picking up SVE2
in both the backend and the driver, the frontend itself still
enabled SVE via the arch version's default extensions.
Avoid that by reverting back to v8.7-a while we look into
longer-term solutions.
---
clang/test/CodeGen/aarch64-targetattr.c | 2 +-
llvm/lib/Target/AArch64/AArch64Processors.td | 5 ++++-
llvm/unittests/TargetParser/TargetParserTest.cpp | 2 +-
3 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/clang/test/CodeGen/aarch64-targetattr.c b/clang/test/CodeGen/aarch64-targetattr.c
index 79d49e522d3e1d..1bc78a6e1f8c0f 100644
--- a/clang/test/CodeGen/aarch64-targetattr.c
+++ b/clang/test/CodeGen/aarch64-targetattr.c
@@ -218,7 +218,7 @@ void applem4() {}
// CHECK: attributes #[[ATTR15]] = { noinline nounwind optnone "branch-target-enforcement" "guarded-control-stack" "no-trapping-math"="true" "sign-return-address"="non-leaf" "sign-return-address-key"="a_key" "stack-protector-buffer-size"="8" "target-cpu"="neoverse-n1" "target-features"="+aes,+bf16,+bti,+ccidx,+complxnum,+crc,+dit,+dotprod,+flagm,+fp-armv8,+fullfp16,+i8mm,+jsconv,+lse,+neon,+pauth,+perfmon,+predres,+ras,+rcpc,+rdm,+sb,+sha2,+spe,+ssbs,+sve,+sve2,+v8.1a,+v8.2a,+v8.3a,+v8.4a,+v8.5a,+v8.6a,+v8a" "tune-cpu"="cortex-a710" }
// CHECK: attributes #[[ATTR16]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" }
// CHECK: attributes #[[ATTR17]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-features"="-v9.3a" }
-// CHECK: attributes #[[ATTR18]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="apple-m4" "target-features"="+aes,+bf16,+bti,+ccidx,+complxnum,+crc,+dit,+dotprod,+flagm,+fp-armv8,+fp16fml,+fpac,+fullfp16,+i8mm,+jsconv,+lse,+mec,+neon,+pauth,+perfmon,+predres,+ras,+rcpc,+rdm,+rme,+sb,+sha2,+sha3,+sme,+sme-f64f64,+sme-i16i64,+sme2,+ssbs,+sve,+sve2,+v8.1a,+v8.2a,+v8.3a,+v8.4a,+v8.5a,+v8.6a,+v8.7a,+v8a,+v9.1a,+v9.2a,+v9a,+wfxt" }
+// CHECK: attributes #[[ATTR18]] = { noinline nounwind optnone "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="apple-m4" "target-features"="+aes,+bf16,+bti,+ccidx,+complxnum,+crc,+dit,+dotprod,+flagm,+fp-armv8,+fp16fml,+fpac,+fullfp16,+i8mm,+jsconv,+lse,+neon,+pauth,+perfmon,+predres,+ras,+rcpc,+rdm,+sb,+sha2,+sha3,+sme,+sme-f64f64,+sme-i16i64,+sme2,+ssbs,+v8.1a,+v8.2a,+v8.3a,+v8.4a,+v8.5a,+v8.6a,+v8.7a,+v8a,+wfxt" }
//.
// CHECK: [[META0:![0-9]+]] = !{i32 1, !"wchar_size", i32 4}
// CHECK: [[META1:![0-9]+]] = !{!"{{.*}}clang version {{.*}}"}
diff --git a/llvm/lib/Target/AArch64/AArch64Processors.td b/llvm/lib/Target/AArch64/AArch64Processors.td
index 84d8cae3a0a5d1..31e3db19e3462d 100644
--- a/llvm/lib/Target/AArch64/AArch64Processors.td
+++ b/llvm/lib/Target/AArch64/AArch64Processors.td
@@ -895,7 +895,10 @@ def ProcessorFeatures {
FeatureLSE, FeaturePAuth, FeatureFPAC,
FeatureRAS, FeatureRCPC, FeatureRDM,
FeatureBF16, FeatureDotProd, FeatureMatMulInt8, FeatureSSBS];
- list<SubtargetFeature> AppleM4 = [HasV9_2aOps, FeatureSHA2, FeatureFPARMv8,
+ // Technically apple-m4 is ARMv9.2a, but a quirk of LLVM defines v9.0 as
+ // requiring SVE, which is optional according to the Arm ARM and not
+ // supported by the core. ARMv8.7a is the next closest choice.
+ list<SubtargetFeature> AppleM4 = [HasV8_7aOps, FeatureSHA2, FeatureFPARMv8,
FeatureNEON, FeaturePerfMon, FeatureSHA3,
FeatureFullFP16, FeatureFP16FML,
FeatureAES, FeatureBF16,
diff --git a/llvm/unittests/TargetParser/TargetParserTest.cpp b/llvm/unittests/TargetParser/TargetParserTest.cpp
index 7d999b826252a2..13db80ab5c68ea 100644
--- a/llvm/unittests/TargetParser/TargetParserTest.cpp
+++ b/llvm/unittests/TargetParser/TargetParserTest.cpp
@@ -1122,7 +1122,7 @@ INSTANTIATE_TEST_SUITE_P(
AArch64CPUTestParams("apple-a16", "armv8.6-a"),
AArch64CPUTestParams("apple-m3", "armv8.6-a"),
AArch64CPUTestParams("apple-a17", "armv8.6-a"),
- AArch64CPUTestParams("apple-m4", "armv9.2-a"),
+ AArch64CPUTestParams("apple-m4", "armv8.7-a"),
AArch64CPUTestParams("exynos-m3", "armv8-a"),
AArch64CPUTestParams("exynos-m4", "armv8.2-a"),
AArch64CPUTestParams("exynos-m5", "armv8.2-a"),
More information about the cfe-commits
mailing list