[AArch64]Fix FEAT_SME_LUTv2 to have FEAT_SME2 implemented. (#110474)

According to [1] "If FEAT_SME_LUTv2 is implemented, then FEAT_SME2 is implemented." The LUTI4[2] not strided and MOVT[3] need FEAT_SME_LUTv2 to be defined. [1]https://developer.arm.com/documentation/109697/0100/Feature-descriptions/The-Armv9-5-architecture-extension?lang=en#md462-the-armv95-architecture-extension__FEAT_SME_LUTv2 [2]https://developer.arm.com/documentation/ddi0602/2024-06/SME-Instructions/LUTI4--four-registers--8-bit---Lookup-table-read-with-4-bit-indexes-and-8-bit-elements- [3]https://developer.arm.com/documentation/ddi0602/2024-06/SME-Instructions/MOVT--vector-to-table---Move-vector-register-to-ZT0-
llvm · Oct 4, 2024 · ef8d506 · ef8d506
1 parent 7be2ce7
commit ef8d506
Show file tree

Hide file tree

Showing 8 changed files with 26 additions and 25 deletions.
diff --git a/clang/include/clang/Basic/arm_sme.td b/clang/include/clang/Basic/arm_sme.td
@@ -818,7 +818,7 @@ multiclass ZAReadzArray<string vg_num>{
 defm SVREADZ_VG2 :  ZAReadzArray<"2">;
 defm SVREADZ_VG4 :  ZAReadzArray<"4">;
 
-let SMETargetGuard = "sme2,sme-lutv2" in {
+let SMETargetGuard = "sme-lutv2" in {
   def SVWRITE_LANE_ZT : SInst<"svwrite_lane_zt[_{d}]", "vidi", "cUcsUsiUilUlfhdb", MergeNone, "aarch64_sme_write_lane_zt", [IsStreaming, IsInOutZT0], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck1_3>]>;
   def SVWRITE_ZT : SInst<"svwrite_zt[_{d}]", "vid", "cUcsUsiUilUlfhdb", MergeNone, "aarch64_sme_write_zt", [IsStreaming, IsOutZT0], [ImmCheck<0, ImmCheck0_0>]>;
   def SVLUTI4_ZT_X4 : SInst<"svluti4_zt_{d}_x4", "4i2.u", "cUc", MergeNone, "aarch64_sme_luti4_zt_x4", [IsStreaming, IsInZT0], [ImmCheck<0, ImmCheck0_0>]>;

diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_write_lane_zt.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_write_lane_zt.c
@@ -1,10 +1,10 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
 
-// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
-// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
-// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +bf16 -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
-// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +bf16 -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
-// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
+// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme2  -target-feature +sme  -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme2  -target-feature +sme  -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +bf16 -target-feature +sme  -target-feature +sme2  -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +bf16 -target-feature +sme  -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -target-feature +sme  -target-feature +sme2 -target-feature +sme-lutv2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
 
 // REQUIRES: aarch64-registered-target
 

diff --git a/llvm/lib/Target/AArch64/AArch64Features.td b/llvm/lib/Target/AArch64/AArch64Features.td
@@ -501,7 +501,7 @@ def FeatureSSVE_FP8DOT2 : ExtensionWithMArch<"ssve-fp8dot2", "SSVE_FP8DOT2", "FE
   "Enable SVE2 FP8 2-way dot product instructions", [FeatureSSVE_FP8DOT4]>;
 
 def FeatureSME_LUTv2 : ExtensionWithMArch<"sme-lutv2", "SME_LUTv2", "FEAT_SME_LUTv2",
-  "Enable Scalable Matrix Extension (SME) LUTv2 instructions">;
+  "Enable Scalable Matrix Extension (SME) LUTv2 instructions", [FeatureSME2]>;
 
 def FeatureSMEF8F32 : ExtensionWithMArch<"sme-f8f32", "SMEF8F32", "FEAT_SME_F8F32",
   "Enable Scalable Matrix Extension (SME) F8F32 instructions", [FeatureSME2, FeatureFP8]>;

diff --git a/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td b/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td
@@ -938,10 +938,11 @@ defm FAMAX_4Z4Z : sme2_fp_sve_destructive_vector_vg4_multi<"famax", 0b0010100>;
 defm FAMIN_4Z4Z : sme2_fp_sve_destructive_vector_vg4_multi<"famin", 0b0010101>;
 } //[HasSME2, HasFAMINMAX]
 
-let Predicates = [HasSME2, HasSME_LUTv2] in {
+
+let Predicates = [HasSME_LUTv2] in {
 defm MOVT_TIZ : sme2_movt_zt_to_zt<"movt",  0b0011111, int_aarch64_sme_write_lane_zt, int_aarch64_sme_write_zt>;
 def LUTI4_4ZZT2Z    : sme2_luti4_vector_vg4<0b00, 0b00,"luti4">;
-} //[HasSME2, HasSME_LUTv2]
+} //[HasSME_LUTv2]
 
 let Predicates = [HasSME2p1, HasSME_LUTv2] in {
 def LUTI4_S_4ZZT2Z  : sme2_luti4_vector_vg4_strided<0b00, 0b00, "luti4">;

diff --git a/llvm/test/MC/AArch64/FP8/directive-arch-negative.s b/llvm/test/MC/AArch64/FP8/directive-arch-negative.s
@@ -51,5 +51,5 @@ luti2  z0.h, { z0.h }, z0[0]
 .arch armv9-a+sme-lutv2
 .arch armv9-a+nosme-lutv2
 luti4  { z0.b - z3.b }, zt0, { z0, z1 }
-// CHECK: error: instruction requires: sme2 sme-lutv2
+// CHECK: error: instruction requires: sme-lutv2
 // CHECK: luti4  { z0.b - z3.b }, zt0, { z0, z1 }
diff --git a/llvm/test/MC/AArch64/FP8_SME2/lut.s b/llvm/test/MC/AArch64/FP8_SME2/lut.s
@@ -1,33 +1,33 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2,+sme2p1,+sme-lutv2 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-lutv2 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
 
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2,+sme2p1,+sme-lutv2 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2,+sme2p1,+sme-lutv2 --no-print-imm-hex - \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-lutv2 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-lutv2 --no-print-imm-hex - \
 // RUN:        | FileCheck %s --check-prefix=CHECK-INST
 
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2,+sme2p1,+sme-lutv2 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-lutv2 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme-lutv2 --no-print-imm-hex - \
 // RUN:        | FileCheck %s --check-prefix=CHECK-UNKNOWN
 
 // Disassemble encoding and check the re-encoding (-show-encoding) matches.
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2,+sme2p1,+sme-lutv2 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-lutv2 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2,+sme2p1,+sme-lutv2 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-lutv2 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 luti4   {z0.b-z3.b}, zt0, {z0-z1}  // 11000000-10001011-00000000-00000000
 // CHECK-INST: luti4   { z0.b - z3.b }, zt0, { z0, z1 }
 // CHECK-ENCODING: [0x00,0x00,0x8b,0xc0]
-// CHECK-ERROR: instruction requires: sme2 sme-lutv2
+// CHECK-ERROR: instruction requires: sme-lutv2
 // CHECK-UNKNOWN: c08b0000 <unknown>
 
 luti4   {z28.b-z31.b}, zt0, {z30-z31}  // 11000000-10001011-00000011-11011100
 // CHECK-INST: luti4   { z28.b - z31.b }, zt0, { z30, z31 }
 // CHECK-ENCODING: [0xdc,0x03,0x8b,0xc0]
-// CHECK-ERROR: instruction requires: sme2 sme-lutv2
+// CHECK-ERROR: instruction requires: sme-lutv2
 // CHECK-UNKNOWN: c08b03dc <unknown>
 
 // Strided

diff --git a/llvm/test/MC/AArch64/FP8_SME2/movt-diagnostics.s b/llvm/test/MC/AArch64/FP8_SME2/movt-diagnostics.s
@@ -1,5 +1,5 @@
 
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2,+sme-lutv2  2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-lutv2  2>&1 < %s | FileCheck %s
 // --------------------------------------------------------------------------//
 // Invalid vector select register
 movt   z0, z31

diff --git a/llvm/test/MC/AArch64/FP8_SME2/movt.s b/llvm/test/MC/AArch64/FP8_SME2/movt.s
@@ -1,14 +1,14 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2,+sme-lutv2 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-lutv2 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
 
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2,+sme-lutv2 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2,+sme-lutv2 --no-print-imm-hex - \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-lutv2 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-lutv2 --no-print-imm-hex - \
 // RUN:        | FileCheck %s --check-prefix=CHECK-INST
 
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2,+sme-lutv2 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-lutv2 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme-lutv2 --no-print-imm-hex - \
 // RUN:        | FileCheck %s --check-prefix=CHECK-UNKNOWN
 
@@ -21,11 +21,11 @@
 movt    zt0, z0  // 11000000-01001111-00000011-11100000
 // CHECK-INST: movt    zt0, z0
 // CHECK-ENCODING: [0xe0,0x03,0x4f,0xc0]
-// CHECK-ERROR: instruction requires: sme2 sme-lutv2
+// CHECK-ERROR: instruction requires: sme-lutv2
 // CHECK-UNKNOWN: c04f03e0 <unknown>
 
 movt    zt0[3, mul vl], z31  // 11000000-01001111-00110011-11111111
 // CHECK-INST: movt    zt0[3, mul vl], z31
 // CHECK-ENCODING: [0xff,0x33,0x4f,0xc0]
-// CHECK-ERROR: instruction requires: sme2 sme-lutv2
+// CHECK-ERROR: instruction requires: sme-lutv2
 // CHECK-UNKNOWN: c04f33ff <unknown>