[Clang][AArch64] Add diagnostics for builtins that use ZT0. #79140
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
Similar to what we did for ZA, this patch adds diagnostics to flag when using a ZT0 builtin in a function that does not have ZT0 state.
sdesmalen-arm
requested review from
SamTebbs33,
kmclaughlin-arm and
MDevereau
llvmbot
added
clang
|
@llvm/pr-subscribers-clang Author: Sander de Smalen (sdesmalen-arm) ChangesSimilar to what we did for ZA, this patch adds diagnostics to flag when using a ZT0 builtin in a function that does not have ZT0 state. Patch is 44.57 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/79140.diff 15 Files Affected:
diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index e027e754477fcf4..a1c32abb4dcd880 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -3170,6 +3170,9 @@ def warn_attribute_arm_sm_incompat_builtin : Warning<
def warn_attribute_arm_za_builtin_no_za_state : Warning<
"builtin call is not valid when calling from a function without active ZA state">,
InGroup<DiagGroup<"undefined-arm-za">>;
+def warn_attribute_arm_zt0_builtin_no_zt0_state : Warning<
+ "builtin call is not valid when calling from a function without active ZT0 state">,
+ InGroup<DiagGroup<"undefined-arm-zt0">>;
def err_sve_vector_in_non_sve_target : Error<
"SVE vector type %0 cannot be used in a target without sve">;
def err_attribute_riscv_rvv_bits_unsupported : Error<
diff --git a/clang/include/clang/Basic/arm_sme.td b/clang/include/clang/Basic/arm_sme.td
index 4fb50b8e4e4e565..695e1bddf9ffc61 100644
--- a/clang/include/clang/Basic/arm_sme.td
+++ b/clang/include/clang/Basic/arm_sme.td
@@ -636,37 +636,37 @@ let TargetGuard = "sme2,sme-i16i64" in {
// Spill and fill of ZT0
//
let TargetGuard = "sme2" in {
- def SVLDR_ZT : Inst<"svldr_zt", "viQ", "", MergeNone, "aarch64_sme_ldr_zt", [IsOverloadNone, IsStreamingCompatible], [ImmCheck<0, ImmCheck0_0>]>;
- def SVSTR_ZT : Inst<"svstr_zt", "vi%", "", MergeNone, "aarch64_sme_str_zt", [IsOverloadNone, IsStreamingCompatible], [ImmCheck<0, ImmCheck0_0>]>;
+ def SVLDR_ZT : Inst<"svldr_zt", "viQ", "", MergeNone, "aarch64_sme_ldr_zt", [IsOverloadNone, IsStreamingCompatible, IsInOutZT0], [ImmCheck<0, ImmCheck0_0>]>;
+ def SVSTR_ZT : Inst<"svstr_zt", "vi%", "", MergeNone, "aarch64_sme_str_zt", [IsOverloadNone, IsStreamingCompatible, IsInZT0], [ImmCheck<0, ImmCheck0_0>]>;
}
//
// Zero ZT0
//
let TargetGuard = "sme2" in {
- def SVZERO_ZT : Inst<"svzero_zt", "vi", "", MergeNone, "aarch64_sme_zero_zt", [IsOverloadNone, IsStreamingCompatible], [ImmCheck<0, ImmCheck0_0>]>;
+ def SVZERO_ZT : Inst<"svzero_zt", "vi", "", MergeNone, "aarch64_sme_zero_zt", [IsOverloadNone, IsStreamingCompatible, IsOutZT0], [ImmCheck<0, ImmCheck0_0>]>;
}
//
// lookup table expand four contiguous registers
//
let TargetGuard = "sme2" in {
- def SVLUTI2_LANE_ZT_X4 : Inst<"svluti2_lane_zt_{d}_x4", "4.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x4", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
- def SVLUTI4_LANE_ZT_X4 : Inst<"svluti4_lane_zt_{d}_x4", "4.di[i", "sUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x4", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_1>]>;
+ def SVLUTI2_LANE_ZT_X4 : Inst<"svluti2_lane_zt_{d}_x4", "4.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x4", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
+ def SVLUTI4_LANE_ZT_X4 : Inst<"svluti4_lane_zt_{d}_x4", "4.di[i", "sUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x4", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_1>]>;
}
//
// lookup table expand one register
//
let TargetGuard = "sme2" in {
- def SVLUTI2_LANE_ZT : Inst<"svluti2_lane_zt_{d}", "di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_15>]>;
- def SVLUTI4_LANE_ZT : Inst<"svluti4_lane_zt_{d}", "di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_7>]>;
+ def SVLUTI2_LANE_ZT : Inst<"svluti2_lane_zt_{d}", "di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_15>]>;
+ def SVLUTI4_LANE_ZT : Inst<"svluti4_lane_zt_{d}", "di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_7>]>;
}
//
// lookup table expand two contiguous registers
//
let TargetGuard = "sme2" in {
- def SVLUTI2_LANE_ZT_X2 : Inst<"svluti2_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x2", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_7>]>;
- def SVLUTI4_LANE_ZT_X2 : Inst<"svluti4_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x2", [IsStreaming], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
+ def SVLUTI2_LANE_ZT_X2 : Inst<"svluti2_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x2", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_7>]>;
+ def SVLUTI4_LANE_ZT_X2 : Inst<"svluti4_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x2", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
}
diff --git a/clang/include/clang/Basic/arm_sve_sme_incl.td b/clang/include/clang/Basic/arm_sve_sme_incl.td
index e6ad84676053b5c..9a6ea9898ef702c 100644
--- a/clang/include/clang/Basic/arm_sve_sme_incl.td
+++ b/clang/include/clang/Basic/arm_sve_sme_incl.td
@@ -229,6 +229,9 @@ def IsStreamingOrSVE2p1 : FlagType<0x40000000000>; // Use for intrin
def IsInZA : FlagType<0x80000000000>;
def IsOutZA : FlagType<0x100000000000>;
def IsInOutZA : FlagType<0x200000000000>;
+def IsInZT0 : FlagType<0x400000000000>;
+def IsOutZT0 : FlagType<0x800000000000>;
+def IsInOutZT0 : FlagType<0x1000000000000>;
// These must be kept in sync with the flags in include/clang/Basic/TargetBuiltins.h
class ImmCheckType<int val> {
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 1f83dcf07b6f9e5..7833d5a2ea20eec 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3013,6 +3013,11 @@ enum ArmSMEState : unsigned {
ArmOutZA = 0b10,
ArmInOutZA = 0b11,
ArmZAMask = 0b11,
+
+ ArmInZT0 = 0b01 << 2,
+ ArmOutZT0 = 0b10 << 2,
+ ArmInOutZT0 = 0b11 << 2,
+ ArmZT0Mask = 0b11 << 2
};
bool Sema::ParseSVEImmChecks(
@@ -3206,6 +3211,13 @@ static bool hasArmZAState(const FunctionDecl *FD) {
(FD->hasAttr<ArmNewAttr>() && FD->getAttr<ArmNewAttr>()->isNewZA());
}
+static bool hasArmZT0State(const FunctionDecl *FD) {
+ const auto *T = FD->getType()->getAs<FunctionProtoType>();
+ return (T && FunctionType::getArmZT0State(T->getAArch64SMEAttributes()) !=
+ FunctionType::ARM_None) ||
+ (FD->hasAttr<ArmNewAttr>() && FD->getAttr<ArmNewAttr>()->isNewZT0());
+}
+
static ArmSMEState getSMEState(unsigned BuiltinID) {
switch (BuiltinID) {
default:
@@ -3233,6 +3245,11 @@ bool Sema::CheckSMEBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
Diag(TheCall->getBeginLoc(),
diag::warn_attribute_arm_za_builtin_no_za_state)
<< TheCall->getSourceRange();
+
+ if ((getSMEState(BuiltinID) & ArmZT0Mask) && !hasArmZT0State(FD))
+ Diag(TheCall->getBeginLoc(),
+ diag::warn_attribute_arm_zt0_builtin_no_zt0_state)
+ << TheCall->getSourceRange();
}
// Range check SME intrinsics that take immediate values.
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_ldr_str_zt.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_ldr_str_zt.c
index 1a495d3b117ecc1..3e4454d943358c8 100644
--- a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_ldr_str_zt.c
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_ldr_str_zt.c
@@ -20,7 +20,7 @@
// CPP-CHECK-NEXT: tail call void @llvm.aarch64.sme.ldr.zt(i32 0, ptr [[BASE:%.*]])
// CPP-CHECK-NEXT: ret void
//
-void test_svldr_zt(const void *base) __arm_streaming_compatible __arm_out("za") {
+void test_svldr_zt(const void *base) __arm_streaming_compatible __arm_out("zt0") {
svldr_zt(0, base);
}
@@ -36,6 +36,6 @@ void test_svldr_zt(const void *base) __arm_streaming_compatible __arm_out("za")
// CPP-CHECK-NEXT: tail call void @llvm.aarch64.sme.str.zt(i32 0, ptr [[BASE:%.*]])
// CPP-CHECK-NEXT: ret void
//
-void test_svstr_zt(void *base) __arm_streaming_compatible __arm_in("za") {
+void test_svstr_zt(void *base) __arm_streaming_compatible __arm_in("zt0") {
svstr_zt(0, base);
}
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt.c
index d656178fdf7a127..7c210fbe6923e92 100644
--- a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt.c
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt.c
@@ -19,7 +19,7 @@
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sme.luti2.lane.zt.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
-svuint8_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint8_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u8(0, zn, 15);
}
@@ -34,7 +34,7 @@ svuint8_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za") {
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sme.luti2.lane.zt.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
-svint8_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint8_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s8(0, zn, 15);
}
@@ -48,7 +48,7 @@ svint8_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za") {
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sme.luti2.lane.zt.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP0]]
//
-svuint16_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint16_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u16(0, zn, 15);
}
@@ -63,7 +63,7 @@ svuint16_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sme.luti2.lane.zt.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP0]]
//
-svint16_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint16_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s16(0, zn, 15);
}
@@ -77,7 +77,7 @@ svint16_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x half> @llvm.aarch64.sme.luti2.lane.zt.nxv8f16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 8 x half> [[TMP0]]
//
-svfloat16_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svfloat16_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_f16(0, zn, 15);
}
@@ -91,7 +91,7 @@ svfloat16_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("za"
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x bfloat> @llvm.aarch64.sme.luti2.lane.zt.nxv8bf16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 8 x bfloat> [[TMP0]]
//
-svbfloat16_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svbfloat16_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_bf16(0, zn, 15);
}
@@ -105,7 +105,7 @@ svbfloat16_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in("z
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sme.luti2.lane.zt.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 4 x i32> [[TMP0]]
//
-svuint32_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint32_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u32(0, zn, 15);
}
@@ -119,7 +119,7 @@ svuint32_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sme.luti2.lane.zt.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 4 x i32> [[TMP0]]
//
-svint32_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint32_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s32(0, zn, 15);
}
@@ -133,6 +133,6 @@ svint32_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sme.luti2.lane.zt.nxv4f32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 15)
// CPP-CHECK-NEXT: ret <vscale x 4 x float> [[TMP0]]
//
-svfloat32_t test_svluti2_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svfloat32_t test_svluti2_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_f32(0, zn, 15);
}
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
index 60cd24fdf4630f6..d7ef75ce01dd70d 100644
--- a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
@@ -26,7 +26,7 @@
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
//
-svuint8x2_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint8x2_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u8_x2(0, zn, 7);
}
@@ -49,7 +49,7 @@ svuint8x2_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
//
-svint8x2_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint8x2_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s8_x2(0, zn, 7);
}
@@ -71,7 +71,7 @@ svint8x2_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
//
-svuint16x2_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint16x2_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u16_x2(0, zn, 7);
}
@@ -94,7 +94,7 @@ svuint16x2_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("za
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
//
-svint16x2_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint16x2_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s16_x2(0, zn, 7);
}
@@ -116,7 +116,7 @@ svint16x2_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_in("za"
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> [[TMP2]], <vscale x 8 x half> [[TMP3]], i64 8)
// CPP-CHECK-NEXT: ret <vscale x 16 x half> [[TMP4]]
//
-svfloat16x2_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svfloat16x2_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_f16_x2(0, zn, 7);
}
@@ -138,7 +138,7 @@ svfloat16x2_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_in("z
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], i64 8)
// CPP-CHECK-NEXT: ret <vscale x 16 x bfloat> [[TMP4]]
//
-svbfloat16x2_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svbfloat16x2_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_bf16_x2(0, zn, 7);
}
@@ -160,7 +160,7 @@ svbfloat16x2_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_in(
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
//
-svuint32x2_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint32x2_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u32_x2(0, zn, 7);
}
@@ -182,7 +182,7 @@ svuint32x2_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_in("za
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
//
-svint32x2_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint32x2_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s32_x2(0, zn, 7);
}
@@ -204,6 +204,6 @@ svint32x2_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_in("za"
// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> [[TMP2]], <vscale x 4 x float> [[TMP3]], i64 4)
// CPP-CHECK-NEXT: ret <vscale x 8 x float> [[TMP4]]
//
-svfloat32x2_t test_svluti2_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_in("za") {
+svfloat32x2_t test_svluti2_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_f32_x2(0, zn, 7);
}
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x4.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x4.c
index e05748c8a64244f..f65c0ef61f81871 100644
--- a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x4.c
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x4.c
@@ -34,7 +34,7 @@
// CPP-CHECK-NEXT: [[TMP8:%.*]] = tail call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> [[TMP6]], <vscale x 16 x i8> [[TMP7]], i64 48)
// CPP-CHECK-NEXT: ret <vscale x 64 x i8> [[TMP8]]
//
-svuint8x4_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint8x4_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_u8_x4(0, zn, 3);
}
@@ -65,7 +65,7 @@ svuint8x4_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP8:%.*]] = tail call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> [[TMP6]], <vscale x 16 x i8> [[TMP7]], i64 48)
// CPP-CHECK-NEXT: ret <vscale x 64 x i8> [[TMP8]]
//
-svint8x4_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za") {
+svint8x4_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("zt0") {
return svluti2_lane_zt_s8_x4(0, zn, 3);
}
@@ -95,7 +95,7 @@ svint8x4_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_in("za")
// CPP-CHECK-NEXT: [[TMP8:%.*]] = tail call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> [[TMP6]], <vscale x 8 x i16> [[TMP7]], i64 24)
// CPP-CHECK-NEXT: ret <vscale x 32 x i16> [[TMP8]]
//
-svuint16x4_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_in("za") {
+svuint16x4_t test_svluti2_lane_zt_u16(svuint8_t...
[truncated]
|
| } | ||
|
|
||
| __arm_new("zt0") | ||
| void new_zt0(void) __arm_streaming { svzero_zt(0); } // no warning |
There was a problem hiding this comment.
Can this be changed to:
// expected-no-warning
to match similar tests above?
SamTebbs33
approved these changes
Jan 23, 2024
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.
Similar to what we did for ZA, this patch adds diagnostics to flag when using a ZT0 builtin in a function that does not have ZT0 state.