This patch introduces support for LUTI2/LUTI4 ACLE for SVE2.

LUTI instructions are used for efficient table lookups with 2-bit
or 4-bit indices. LUTI2 reads indexed 8-bit or 16-bit elements from
the low 128 bits of the table vector using packed 2-bit indices,
while LUTI4 can read from the low 128 or 256 bits of the table
vector or from two table vectors using packed 4-bit indices.
These instructions fill the destination vector by copying elements
indexed by segments of the source vector, selected by the vector
segment index.

The changes include the addition of a new AArch64 option
extension "lut", __ARM_FEATURE_LUT preprocessor macro, definitions
for the new LUTI instruction shapes, and implementations of the
svluti2 and svluti4 builtins.

New tests are added as well.

---

Hey,

This is a respin of
https://gcc.gnu.org/pipermail/gcc-patches/2024-July/658015.html. Rebased
with master. Regression tested on aarch64-unknown-linux-gnu and found no
regressions.

Ok for master?

Thanks,
Saurabh
---
 gcc/config/aarch64/aarch64-c.cc               |  2 +
 .../aarch64/aarch64-sve-builtins-shapes.cc    | 41 +++++++++++++++++
 .../aarch64/aarch64-sve-builtins-shapes.h     |  2 +
 .../aarch64/aarch64-sve-builtins-sve2.cc      | 17 +++++++
 .../aarch64/aarch64-sve-builtins-sve2.def     |  4 ++
 .../aarch64/aarch64-sve-builtins-sve2.h       |  2 +
 gcc/config/aarch64/aarch64-sve2.md            | 45 +++++++++++++++++++
 gcc/config/aarch64/iterators.md               | 10 +++++
 .../aarch64/sve/acle/asm/test_sve_acle.h      | 16 +++++++
 .../aarch64/sve2/acle/asm/luti2_bf16.c        | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti2_f16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti2_s16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti2_s8.c          | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti2_u16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti2_u8.c          | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti4_bf16.c        | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti4_bf16_x2.c     | 20 +++++++++
 .../aarch64/sve2/acle/asm/luti4_f16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti4_f16_x2.c      | 20 +++++++++
 .../aarch64/sve2/acle/asm/luti4_s16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti4_s16_x2.c      | 20 +++++++++
 .../aarch64/sve2/acle/asm/luti4_s8.c          | 30 +++++++++++++
 .../aarch64/sve2/acle/asm/luti4_u16.c         | 40 +++++++++++++++++
 .../aarch64/sve2/acle/asm/luti4_u16_x2.c      | 20 +++++++++
 .../aarch64/sve2/acle/asm/luti4_u8.c          | 30 +++++++++++++
 gcc/testsuite/lib/target-supports.exp         | 12 +++++
 26 files changed, 691 insertions(+)
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_bf16.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_f16.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s16.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s8.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u16.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u8.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16.c
 create mode 100644 
gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16_x2.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16.c
 create mode 100644 
gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16_x2.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16.c
 create mode 100644 
gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16_x2.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s8.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16.c
 create mode 100644 
gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16_x2.c
 create mode 100644 gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u8.c

diff --git a/gcc/config/aarch64/aarch64-c.cc b/gcc/config/aarch64/aarch64-c.cc
index 92fcf5389a3..d1e2ab9831d 100644
--- a/gcc/config/aarch64/aarch64-c.cc
+++ b/gcc/config/aarch64/aarch64-c.cc
@@ -268,6 +268,8 @@ aarch64_update_cpp_builtins (cpp_reader *pfile)
   aarch64_def_or_undef (TARGET_SVE_BF16,
 			"__ARM_FEATURE_SVE_BF16", pfile);
 
+  aarch64_def_or_undef (TARGET_LUT, "__ARM_FEATURE_LUT", pfile);
+
   aarch64_def_or_undef (TARGET_FP8, "__ARM_FEATURE_FP8", pfile);
 
   aarch64_def_or_undef (TARGET_FP8DOT2, "__ARM_FEATURE_FP8DOT2", pfile);
diff --git a/gcc/config/aarch64/aarch64-sve-builtins-shapes.cc b/gcc/config/aarch64/aarch64-sve-builtins-shapes.cc
index ca721dd2c09..0f6d366b2d6 100644
--- a/gcc/config/aarch64/aarch64-sve-builtins-shapes.cc
+++ b/gcc/config/aarch64/aarch64-sve-builtins-shapes.cc
@@ -903,6 +903,47 @@ struct load_ext_gather_base : public overloaded_base<1>
   }
 };
 
+
+/* sv<v0>_t svlut_<t0>(sv<t0>_t, svuint8_t, uint64_t)
+   where the final argument is a constant index, the instruction divides
+   the vector argument in BITS-bit quantities.  */
+template<unsigned int BITS>
+struct luti_base : public nonoverloaded_base
+{
+  void
+  build (function_builder &b, const function_group_info &group) const override
+  {
+    /* Format: return type, table vector, indices vector, immediate value.  */
+    build_all (b, "v0,t0,vu8,su64", group, MODE_none);
+  }
+
+  bool
+  check (function_checker &c) const override
+  {
+    int max_range;
+    bool byte_mode = c.type_suffix (0).element_bits == 8;
+
+    if (BITS == 2)
+      max_range = byte_mode ? 3 : 7;
+    else if (BITS == 4)
+      max_range = byte_mode ? 1 : 7;
+    else
+      /* Unsupported number of indices bits for LUTI.  */
+      gcc_unreachable ();
+
+    return c.require_immediate_range (2, 0, max_range);
+  }
+
+};
+
+/* Specializations for 2-bit and 4-bit indices.  */
+using luti2_def = luti_base<2>;
+SHAPE (luti2)
+
+using luti4_def = luti_base<4>;
+SHAPE (luti4)
+
+
 /* sv<t0>x<g>_t svfoo_t0_g(uint64_t, svuint8_t, uint64_t)
 
    where the first argument is the ZT register number (currently always 0)
diff --git a/gcc/config/aarch64/aarch64-sve-builtins-shapes.h b/gcc/config/aarch64/aarch64-sve-builtins-shapes.h
index 56e2f57b036..349eae6e34a 100644
--- a/gcc/config/aarch64/aarch64-sve-builtins-shapes.h
+++ b/gcc/config/aarch64/aarch64-sve-builtins-shapes.h
@@ -164,6 +164,8 @@ namespace aarch64_sve
     extern const function_shape *const load_gather64_vs_offset;
     extern const function_shape *const load_replicate;
     extern const function_shape *const load_za;
+    extern const function_shape *const luti2;
+    extern const function_shape *const luti4;
     extern const function_shape *const luti2_lane_zt;
     extern const function_shape *const luti4_lane_zt;
     extern const function_shape *const mmla;
diff --git a/gcc/config/aarch64/aarch64-sve-builtins-sve2.cc b/gcc/config/aarch64/aarch64-sve-builtins-sve2.cc
index 0c5290411bb..d9922de7ca5 100644
--- a/gcc/config/aarch64/aarch64-sve-builtins-sve2.cc
+++ b/gcc/config/aarch64/aarch64-sve-builtins-sve2.cc
@@ -914,6 +914,21 @@ public:
   unsigned int m_base;
 };
 
+class svluti_lane_impl : public function_base
+{
+public:
+  CONSTEXPR svluti_lane_impl (unsigned int bits) : m_bits (bits)
+  {}
+
+  rtx expand (function_expander &e) const override
+  {
+    auto mode = e.tuple_mode (0);
+    return e.use_exact_insn (code_for_aarch64_sve_luti (m_bits, mode));
+  }
+
+  unsigned int m_bits;
+};
+
 } /* end anonymous namespace */
 
 namespace aarch64_sve {
@@ -1205,5 +1220,7 @@ FUNCTION (svzip, multireg_permute, (UNSPEC_ZIP))
 FUNCTION (svzipq, multireg_permute, (UNSPEC_ZIPQ))
 FUNCTION (svzipq1, svzipq_impl, (0))
 FUNCTION (svzipq2, svzipq_impl, (1))
+FUNCTION (svluti2_lane, svluti_lane_impl, (2))
+FUNCTION (svluti4_lane, svluti_lane_impl, (4))
 
 } /* end namespace aarch64_sve */
diff --git a/gcc/config/aarch64/aarch64-sve-builtins-sve2.def b/gcc/config/aarch64/aarch64-sve-builtins-sve2.def
index e726fa1fb68..0c4f8251ac0 100644
--- a/gcc/config/aarch64/aarch64-sve-builtins-sve2.def
+++ b/gcc/config/aarch64/aarch64-sve-builtins-sve2.def
@@ -164,6 +164,10 @@ DEF_SVE_FUNCTION (svwhilegt, compare_scalar, while, none)
 DEF_SVE_FUNCTION (svwhilerw, compare_ptr, all_data, none)
 DEF_SVE_FUNCTION (svwhilewr, compare_ptr, all_data, none)
 DEF_SVE_FUNCTION (svxar, ternary_shift_right_imm, all_integer, none)
+DEF_SVE_FUNCTION (svluti2_lane, luti2, bhs_data, none)
+DEF_SVE_FUNCTION (svluti4_lane, luti4, bhs_data, none)
+DEF_SVE_FUNCTION_GS (svluti4_lane, luti4, bhs_data, x2, none)
+
 #undef REQUIRED_EXTENSIONS
 
 #define REQUIRED_EXTENSIONS nonstreaming_sve (AARCH64_FL_SVE2)
diff --git a/gcc/config/aarch64/aarch64-sve-builtins-sve2.h b/gcc/config/aarch64/aarch64-sve-builtins-sve2.h
index 14d668d2649..6d7d0af2641 100644
--- a/gcc/config/aarch64/aarch64-sve-builtins-sve2.h
+++ b/gcc/config/aarch64/aarch64-sve-builtins-sve2.h
@@ -250,6 +250,8 @@ namespace aarch64_sve
     extern const function_base *const svzipq;
     extern const function_base *const svzipq1;
     extern const function_base *const svzipq2;
+    extern const function_base *const svluti2_lane;
+    extern const function_base *const svluti4_lane;
   }
 }
 
diff --git a/gcc/config/aarch64/aarch64-sve2.md b/gcc/config/aarch64/aarch64-sve2.md
index f8cfe08f4c0..7dcbc0700da 100644
--- a/gcc/config/aarch64/aarch64-sve2.md
+++ b/gcc/config/aarch64/aarch64-sve2.md
@@ -133,6 +133,7 @@
 ;; ---- Optional AES extensions
 ;; ---- Optional SHA-3 extensions
 ;; ---- Optional SM4 extensions
+;; ---- Table lookup
 
 ;; =========================================================================
 ;; == Moves
@@ -4211,3 +4212,47 @@
   "sm4ekey\t%0.s, %1.s, %2.s"
   [(set_attr "type" "crypto_sm4")]
 )
+
+;; -------------------------------------------------------------------------
+;; ---- Table lookup
+;; -------------------------------------------------------------------------
+;; Includes:
+;; - LUTI2
+;; - LUTI4
+;; -------------------------------------------------------------------------
+
+(define_insn "@aarch64_sve_luti<LUTI_BITS><mode>"
+  [(set (match_operand:SVE_FULL_BS 0 "register_operand" "=w")
+	(unspec:SVE_FULL_BS
+	  [(match_operand:SVE_FULL_BS 1 "register_operand" "w")
+	   (match_operand:VNx16QI 2 "register_operand" "w")
+	   (match_operand:DI 3 "const_int_operand")
+	   (const_int LUTI_BITS)]
+	  UNSPEC_SVE_LUTI))]
+  "TARGET_SVE2"
+  "luti<LUTI_BITS>\t%0.<Vetype>, { %1.<Vetype> }, %2[%3]"
+)
+
+(define_insn "@aarch64_sve_luti<LUTI_BITS><mode>"
+  [(set (match_operand:<VSINGLE> 0 "register_operand" "=w")
+	(unspec:<VSINGLE>
+		[(match_operand:SVE_FULL_H 1 "aligned_register_operand" "w")
+		(match_operand:VNx16QI 2 "register_operand" "w")
+		(match_operand:DI 3 "const_int_operand")
+		(const_int LUTI_BITS)]
+		UNSPEC_SVE_LUTI))]
+  "TARGET_SVE2"
+  "luti<LUTI_BITS>\t%0.<Vetype>, { %1.<Vetype> }, %2[%3]"
+)
+
+(define_insn "@aarch64_sve_luti<LUTI_BITS><mode>"
+  [(set (match_operand:<VSINGLE> 0 "register_operand" "=w")
+	(unspec:<VSINGLE>
+		[(match_operand:SVE_FULL_Hx2 1 "aligned_register_operand" "Uw2")
+		(match_operand:VNx16QI 2 "register_operand" "w")
+		(match_operand:DI 3 "const_int_operand")
+		(const_int LUTI_BITS)]
+		UNSPEC_SVE_LUTI))]
+  "TARGET_SVE2"
+  "luti<LUTI_BITS>\t%0.<Vetype>, %1, %2[%3]"
+)
diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md
index ff0f34dd043..74dc1a59dc7 100644
--- a/gcc/config/aarch64/iterators.md
+++ b/gcc/config/aarch64/iterators.md
@@ -553,6 +553,15 @@
 (define_mode_iterator SVE_FULL_BHS [VNx16QI VNx8HI VNx4SI
 				    VNx8BF VNx8HF VNx4SF])
 
+;; Fully-packed SVE vector byte modes that have 32-bit or smaller elements.
+(define_mode_iterator SVE_FULL_BS [VNx16QI VNx4SI VNx4SF])
+
+;; Fully-packed half word SVE vector modes
+(define_mode_iterator SVE_FULL_H [VNx8HI VNx8HF VNx8BF])
+
+;; Pairs of fully-packed SVE vector modes (half word only)
+(define_mode_iterator SVE_FULL_Hx2 [VNx16HI VNx16HF VNx16BF])
+
 ;; Fully-packed SVE vector modes that have 32-bit elements.
 (define_mode_iterator SVE_FULL_S [VNx4SI VNx4SF])
 
@@ -1186,6 +1195,7 @@
     UNSPEC_UZPQ2
     UNSPEC_ZIPQ1
     UNSPEC_ZIPQ2
+    UNSPEC_SVE_LUTI
 
     ;; All used in aarch64-sme.md
     UNSPEC_SME_ADD
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/acle/asm/test_sve_acle.h b/gcc/testsuite/gcc.target/aarch64/sve/acle/asm/test_sve_acle.h
index d3ae707ac49..c0dd89fa924 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/acle/asm/test_sve_acle.h
+++ b/gcc/testsuite/gcc.target/aarch64/sve/acle/asm/test_sve_acle.h
@@ -780,4 +780,20 @@
 		    "w" (z16), "w" (z22), "w" (z29));		\
   }
 
+#define TEST_1X2_NARROW(NAME, RTYPE, TTYPE, ZTYPE, CODE1, CODE2)	\
+  PROTO(NAME, void, ())							\
+  {									\
+    register RTYPE z0 __asm ("z0");					\
+    register ZTYPE z5 __asm ("z5");					\
+    register TTYPE z6 __asm ("z6");					\
+    register RTYPE z16 __asm ("z16");					\
+    register ZTYPE z22 __asm ("z22");					\
+    register TTYPE z29 __asm ("z29");					\
+    register RTYPE z0_res __asm ("z0");					\
+    __asm volatile ("" : "=w" (z0), "=w" (z5), "=w" (z6),		\
+		    "=w" (z16), "=w" (z22), "=w" (z29));		\
+    INVOKE (CODE1, CODE2);						\
+    __asm volatile ("" :: "w" (z0_res), "w" (z5), "w" (z22));		\
+  }
+
 #endif
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_bf16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_bf16.c
new file mode 100644
index 00000000000..f423bfae2c6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_bf16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	 luti2	z1\.h, \{ z28\.h \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svbfloat16_t, svuint8_t, z1,
+                svluti2_lane_bf16 (z28, z0, 0),
+                svluti2_lane_bf16 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	 luti2	z1\.h, \{ z28\.h \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svbfloat16_t, svuint8_t, z1,
+                svluti2_lane_bf16 (z28, z0, 1),
+                svluti2_lane_bf16 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	 luti2	z28\.h, \{ z28\.h \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svbfloat16_t, svuint8_t, z28,
+                svluti2_lane_bf16 (z28, z0, 2),
+                svluti2_lane_bf16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_f16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_f16.c
new file mode 100644
index 00000000000..ca6cd4e7173
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_f16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	 luti2	z1\.h, \{ z28\.h \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svfloat16_t, svuint8_t, z1,
+                svluti2_lane_f16 (z28, z0, 0),
+                svluti2_lane_f16 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	 luti2	z1\.h, \{ z28\.h \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svfloat16_t, svuint8_t, z1,
+                svluti2_lane_f16 (z28, z0, 1),
+                svluti2_lane_f16 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	 luti2	z28\.h, \{ z28\.h \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svfloat16_t, svuint8_t, z28,
+                svluti2_lane_f16 (z28, z0, 2),
+                svluti2_lane_f16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s16.c
new file mode 100644
index 00000000000..a199027fda8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	luti2	z1\.h, \{ z28\.h \}, z0\[0\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svint16_t, svuint8_t, z1,
+                svluti2_lane_s16 (z28, z0, 0),
+                svluti2_lane_s16 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	luti2	z1\.h, \{ z28\.h \}, z0\[1\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svint16_t, svuint8_t, z1,
+                svluti2_lane_s16 (z28, z0, 1),
+                svluti2_lane_s16 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	luti2	z28\.h, \{ z28\.h \}, z0\[2\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svint16_t, svuint8_t, z28,
+                svluti2_lane_s16 (z28, z0, 2),
+                svluti2_lane_s16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s8.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s8.c
new file mode 100644
index 00000000000..68516bdb6c2
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_s8.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	 luti2	z1\.b, \{ z28\.b \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svint8_t, svuint8_t, z1,
+                svluti2_lane_s8 (z28, z0, 0),
+                svluti2_lane_s8 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	 luti2	z1\.b, \{ z28\.b \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svint8_t, svuint8_t, z1,
+                svluti2_lane_s8 (z28, z0, 1),
+                svluti2_lane_s8 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	 luti2	z28\.b, \{ z28\.b \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svint8_t, svuint8_t, z28,
+                svluti2_lane_s8 (z28, z0, 2),
+                svluti2_lane_s8 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u16.c
new file mode 100644
index 00000000000..d055f7a28a0
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	luti2	z1\.h, \{ z28\.h \}, z0\[0\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svuint16_t, svuint8_t, z1,
+                svluti2_lane_u16 (z28, z0, 0),
+                svluti2_lane_u16 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	luti2	z1\.h, \{ z28\.h \}, z0\[1\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svuint16_t, svuint8_t, z1,
+                svluti2_lane_u16 (z28, z0, 1),
+                svluti2_lane_u16 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	luti2	z28\.h, \{ z28\.h \}, z0\[2\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svuint16_t, svuint8_t, z28,
+                svluti2_lane_u16 (z28, z0, 2),
+                svluti2_lane_u16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u8.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u8.c
new file mode 100644
index 00000000000..2a9b7465ad7
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti2_u8.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti2_test_imm0:
+**	 luti2	z1\.b, \{ z28\.b \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm0, svuint8_t, svuint8_t, z1,
+                svluti2_lane_u8 (z28, z0, 0),
+                svluti2_lane_u8 (z28, z0, 0))
+
+/*
+** luti2_test_imm1:
+**	 luti2	z1\.b, \{ z28\.b \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_imm1, svuint8_t, svuint8_t, z1,
+                svluti2_lane_u8 (z28, z0, 1),
+                svluti2_lane_u8 (z28, z0, 1))
+
+/*
+** luti2_test_tied:
+**	 luti2	z28\.b, \{ z28\.b \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti2_test_tied, svuint8_t, svuint8_t, z28,
+                svluti2_lane_u8 (z28, z0, 2),
+                svluti2_lane_u8 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16.c
new file mode 100644
index 00000000000..7dd29db3cd8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	 luti4	z1\.h, \{ z28\.h \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svbfloat16_t, svuint8_t, z1,
+                svluti4_lane_bf16 (z28, z0, 0),
+                svluti4_lane_bf16 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	 luti4	z1\.h, \{ z28\.h \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svbfloat16_t, svuint8_t, z1,
+                svluti4_lane_bf16 (z28, z0, 1),
+                svluti4_lane_bf16 (z28, z0, 1))
+
+/*
+** luti4_test_tied:
+**	 luti4	z28\.h, \{ z28\.h \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_tied, svbfloat16_t, svuint8_t, z28,
+                svluti4_lane_bf16 (z28, z0, 2),
+                svluti4_lane_bf16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16_x2.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16_x2.c
new file mode 100644
index 00000000000..afb1bfa55c4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_bf16_x2.c
@@ -0,0 +1,20 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_bf16_x2:
+**	luti4	z0\.h, \{z6\.h \- z7\.h\}, z5\[1\]
+**	ret
+*/
+
+TEST_1X2_NARROW(luti4_test_bf16_x2, svbfloat16_t, svbfloat16x2_t, svuint8_t,
+                 z0_res = svluti4_lane_bf16_x2(z6, z5, 1),
+                 z0_res = svluti4_lane_bf16_x2(z6, z5, 1))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16.c
new file mode 100644
index 00000000000..b04c5b6329c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	 luti4	z1\.h, \{ z28\.h \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svfloat16_t, svuint8_t, z1,
+                svluti4_lane_f16 (z28, z0, 0),
+                svluti4_lane_f16 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	 luti4	z1\.h, \{ z28\.h \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svfloat16_t, svuint8_t, z1,
+                svluti4_lane_f16 (z28, z0, 1),
+                svluti4_lane_f16 (z28, z0, 1))
+
+/*
+** luti4_test_tied:
+**	 luti4	z28\.h, \{ z28\.h \}, z0\[2\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_tied, svfloat16_t, svuint8_t, z28,
+                svluti4_lane_f16 (z28, z0, 2),
+                svluti4_lane_f16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16_x2.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16_x2.c
new file mode 100644
index 00000000000..ff382f790a9
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_f16_x2.c
@@ -0,0 +1,20 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_f16_x2:
+**	luti4	z0\.h, \{z6\.h \- z7\.h\}, z5\[1\]
+**	ret
+*/
+
+TEST_1X2_NARROW(luti4_test_f16_x2, svfloat16_t, svfloat16x2_t, svuint8_t,
+                 z0_res = svluti4_lane_f16_x2(z6, z5, 1),
+                 z0_res = svluti4_lane_f16_x2(z6, z5, 1))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16.c
new file mode 100644
index 00000000000..2d9bf3109d8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	luti4	z1\.h, \{ z28\.h \}, z0\[0\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svint16_t, svuint8_t, z1,
+                svluti4_lane_s16 (z28, z0, 0),
+                svluti4_lane_s16 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	luti4	z1\.h, \{ z28\.h \}, z0\[1\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svint16_t, svuint8_t, z1,
+                svluti4_lane_s16 (z28, z0, 1),
+                svluti4_lane_s16 (z28, z0, 1))
+
+/*
+** luti4_test_tied:
+**	luti4	z28\.h, \{ z28\.h \}, z0\[2\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_tied, svint16_t, svuint8_t, z28,
+                svluti4_lane_s16 (z28, z0, 2),
+                svluti4_lane_s16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16_x2.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16_x2.c
new file mode 100644
index 00000000000..0de54702944
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s16_x2.c
@@ -0,0 +1,20 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_s16_x2:
+**	luti4	z0\.h, \{z6\.h \- z7\.h\}, z5\[1\]
+**	ret
+*/
+
+TEST_1X2_NARROW(luti4_test_s16_x2, svint16_t, svint16x2_t, svuint8_t,
+                 z0_res = svluti4_lane_s16_x2(z6, z5, 1),
+                 z0_res = svluti4_lane_s16_x2(z6, z5, 1))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s8.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s8.c
new file mode 100644
index 00000000000..c9de18ebc8f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_s8.c
@@ -0,0 +1,30 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	 luti4	z1\.b, \{ z28\.b \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svint8_t, svuint8_t, z1,
+                svluti4_lane_s8 (z28, z0, 0),
+                svluti4_lane_s8 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	 luti4	z1\.b, \{ z28\.b \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svint8_t, svuint8_t, z1,
+                svluti4_lane_s8 (z28, z0, 1),
+                svluti4_lane_s8 (z28, z0, 1))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16.c
new file mode 100644
index 00000000000..31ac56ea18c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16.c
@@ -0,0 +1,40 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	luti4	z1\.h, \{ z28\.h \}, z0\[0\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svuint16_t, svuint8_t, z1,
+                svluti4_lane_u16 (z28, z0, 0),
+                svluti4_lane_u16 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	luti4	z1\.h, \{ z28\.h \}, z0\[1\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svuint16_t, svuint8_t, z1,
+                svluti4_lane_u16 (z28, z0, 1),
+                svluti4_lane_u16 (z28, z0, 1))
+
+/*
+** luti4_test_tied:
+**	luti4	z28\.h, \{ z28\.h \}, z0\[2\]
+**	ret
+*/
+
+TEST_XN_SINGLE (luti4_test_tied, svuint16_t, svuint8_t, z28,
+                svluti4_lane_u16 (z28, z0, 2),
+                svluti4_lane_u16 (z28, z0, 2))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16_x2.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16_x2.c
new file mode 100644
index 00000000000..5289ffda0af
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u16_x2.c
@@ -0,0 +1,20 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_u16_x2:
+**	luti4	z0\.h, \{z6\.h \- z7\.h\}, z5\[1\]
+**	ret
+*/
+
+TEST_1X2_NARROW(luti4_test_u16_x2, svuint16_t, svuint16x2_t, svuint8_t,
+                 z0_res = svluti4_lane_u16_x2(z6, z5, 1),
+                 z0_res = svluti4_lane_u16_x2(z6, z5, 1))
+
diff --git a/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u8.c b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u8.c
new file mode 100644
index 00000000000..9768f24bd46
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve2/acle/asm/luti4_u8.c
@@ -0,0 +1,30 @@
+/* { dg-do compile } */
+/* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
+
+#include "test_sve_acle.h"
+
+#pragma GCC target "+sve2+lut"
+#if STREAMING_COMPATIBLE
+#pragma GCC target "+sme2"
+#endif
+
+/*
+** luti4_test_imm0:
+**	 luti4	z1\.b, \{ z28\.b \}, z0\[0\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm0, svuint8_t, svuint8_t, z1,
+                svluti4_lane_u8 (z28, z0, 0),
+                svluti4_lane_u8 (z28, z0, 0))
+
+/*
+** luti4_test_imm1:
+**	 luti4	z1\.b, \{ z28\.b \}, z0\[1\]
+**	 ret
+*/
+
+TEST_XN_SINGLE (luti4_test_imm1, svuint8_t, svuint8_t, z1,
+                svluti4_lane_u8 (z28, z0, 1),
+                svluti4_lane_u8 (z28, z0, 1))
+
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 45ba2f47a9d..e0d9867801c 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -4800,6 +4800,18 @@ proc check_effective_target_aarch64_sve2 { } {
     }]
 }
 
+# Return 1 if this is an AArch64 target supporting LUT (Lookup table)
+proc check_effective_target_aarch64_lut { } {
+    if { ![istarget aarch64*-*-*] || ![check_effective_target_aarch64_sve2] } {
+        return 0
+    }
+    return [check_no_compiler_messages aarch64_lut assembly {
+        #if !defined (__ARM_FEATURE_LUT)
+        #error FOO
+        #endif
+    }]
+}
+
 # Return 1 if this is an AArch64 target only supporting SVE (not SVE2).
 proc check_effective_target_aarch64_sve1_only { } {
     return [expr { [check_effective_target_aarch64_sve]

Reply via email to