[Qemu-devel] [PATCH v2 17/67] target/arm: Implement SVE Index Generation Group

[Richard Henderson <richard.henderson@linaro.org>]

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/arm/helper-sve.h    |  5 ++++
 target/arm/sve_helper.c    | 40 +++++++++++++++++++++++++++
 target/arm/translate-sve.c | 67 ++++++++++++++++++++++++++++++++++++++++++++++
 target/arm/sve.decode      | 14 ++++++++++
 4 files changed, 126 insertions(+)

diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index b31d497f31..2a2dbe98dd 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -363,6 +363,11 @@ DEF_HELPER_FLAGS_6(sve_mls_s, TCG_CALL_NO_RWG,
 DEF_HELPER_FLAGS_6(sve_mls_d, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, ptr, i32)
 
+DEF_HELPER_FLAGS_4(sve_index_b, TCG_CALL_NO_RWG, void, ptr, i32, i32, i32)
+DEF_HELPER_FLAGS_4(sve_index_h, TCG_CALL_NO_RWG, void, ptr, i32, i32, i32)
+DEF_HELPER_FLAGS_4(sve_index_s, TCG_CALL_NO_RWG, void, ptr, i32, i32, i32)
+DEF_HELPER_FLAGS_4(sve_index_d, TCG_CALL_NO_RWG, void, ptr, i64, i64, i32)
+
 DEF_HELPER_FLAGS_5(sve_and_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_5(sve_bic_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_5(sve_eor_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, i32)
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c
index 4b08a38ce8..950012e70a 100644
--- a/target/arm/sve_helper.c
+++ b/target/arm/sve_helper.c
@@ -991,3 +991,43 @@ DO_ZPZZZ_D(sve_mls_d, uint64_t, DO_MLS)
 #undef DO_MLS
 #undef DO_ZPZZZ
 #undef DO_ZPZZZ_D
+
+void HELPER(sve_index_b)(void *vd, uint32_t start,
+                         uint32_t incr, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc);
+    uint8_t *d = vd;
+    for (i = 0; i < opr_sz; i += 1) {
+        d[H1(i)] = start + i * incr;
+    }
+}
+
+void HELPER(sve_index_h)(void *vd, uint32_t start,
+                         uint32_t incr, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc) / 2;
+    uint16_t *d = vd;
+    for (i = 0; i < opr_sz; i += 1) {
+        d[H2(i)] = start + i * incr;
+    }
+}
+
+void HELPER(sve_index_s)(void *vd, uint32_t start,
+                         uint32_t incr, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc) / 4;
+    uint32_t *d = vd;
+    for (i = 0; i < opr_sz; i += 1) {
+        d[H4(i)] = start + i * incr;
+    }
+}
+
+void HELPER(sve_index_d)(void *vd, uint64_t start,
+                         uint64_t incr, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc) / 8;
+    uint64_t *d = vd;
+    for (i = 0; i < opr_sz; i += 1) {
+        d[i] = start + i * incr;
+    }
+}
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index 8baec6c674..773f0bfded 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -675,6 +675,73 @@ DO_ZPZZZ(MLS, mls)
 
 #undef DO_ZPZZZ
 
+/*
+ *** SVE Index Generation Group
+ */
+
+static void do_index(DisasContext *s, int esz, int rd,
+                     TCGv_i64 start, TCGv_i64 incr)
+{
+    unsigned vsz = vec_full_reg_size(s);
+    TCGv_i32 desc = tcg_const_i32(simd_desc(vsz, vsz, 0));
+    TCGv_ptr t_zd = tcg_temp_new_ptr();
+
+    tcg_gen_addi_ptr(t_zd, cpu_env, vec_full_reg_offset(s, rd));
+    if (esz == 3) {
+        gen_helper_sve_index_d(t_zd, start, incr, desc);
+    } else {
+        typedef void index_fn(TCGv_ptr, TCGv_i32, TCGv_i32, TCGv_i32);
+        static index_fn * const fns[3] = {
+            gen_helper_sve_index_b,
+            gen_helper_sve_index_h,
+            gen_helper_sve_index_s,
+        };
+        TCGv_i32 s32 = tcg_temp_new_i32();
+        TCGv_i32 i32 = tcg_temp_new_i32();
+
+        tcg_gen_extrl_i64_i32(s32, start);
+        tcg_gen_extrl_i64_i32(i32, incr);
+        fns[esz](t_zd, s32, i32, desc);
+
+        tcg_temp_free_i32(s32);
+        tcg_temp_free_i32(i32);
+    }
+    tcg_temp_free_ptr(t_zd);
+    tcg_temp_free_i32(desc);
+}
+
+static void trans_INDEX_ii(DisasContext *s, arg_INDEX_ii *a, uint32_t insn)
+{
+    TCGv_i64 start = tcg_const_i64(a->imm1);
+    TCGv_i64 incr = tcg_const_i64(a->imm2);
+    do_index(s, a->esz, a->rd, start, incr);
+    tcg_temp_free_i64(start);
+    tcg_temp_free_i64(incr);
+}
+
+static void trans_INDEX_ir(DisasContext *s, arg_INDEX_ir *a, uint32_t insn)
+{
+    TCGv_i64 start = tcg_const_i64(a->imm);
+    TCGv_i64 incr = cpu_reg(s, a->rm);
+    do_index(s, a->esz, a->rd, start, incr);
+    tcg_temp_free_i64(start);
+}
+
+static void trans_INDEX_ri(DisasContext *s, arg_INDEX_ri *a, uint32_t insn)
+{
+    TCGv_i64 start = cpu_reg(s, a->rn);
+    TCGv_i64 incr = tcg_const_i64(a->imm);
+    do_index(s, a->esz, a->rd, start, incr);
+    tcg_temp_free_i64(incr);
+}
+
+static void trans_INDEX_rr(DisasContext *s, arg_INDEX_rr *a, uint32_t insn)
+{
+    TCGv_i64 start = cpu_reg(s, a->rn);
+    TCGv_i64 incr = cpu_reg(s, a->rm);
+    do_index(s, a->esz, a->rd, start, incr);
+}
+
 /*
  *** SVE Predicate Logical Operations Group
  */
diff --git a/target/arm/sve.decode b/target/arm/sve.decode
index b40d7dc9a2..d7b078e92f 100644
--- a/target/arm/sve.decode
+++ b/target/arm/sve.decode
@@ -226,6 +226,20 @@ ORR_zzz		00000100 01 1 ..... 001 100 ..... .....		@rd_rn_rm_e0
 EOR_zzz		00000100 10 1 ..... 001 100 ..... .....		@rd_rn_rm_e0
 BIC_zzz		00000100 11 1 ..... 001 100 ..... .....		@rd_rn_rm_e0
 
+### SVE Index Generation Group
+
+# SVE index generation (immediate start, immediate increment)
+INDEX_ii	00000100 esz:2 1 imm2:s5 010000 imm1:s5 rd:5
+
+# SVE index generation (immediate start, register increment)
+INDEX_ir	00000100 esz:2 1 rm:5 010010 imm:s5 rd:5
+
+# SVE index generation (register start, immediate increment)
+INDEX_ri	00000100 esz:2 1 imm:s5 010001 rn:5 rd:5
+
+# SVE index generation (register start, register increment)
+INDEX_rr	00000100 .. 1 ..... 010011 ..... .....		@rd_rn_rm
+
 ### SVE Predicate Logical Operations Group
 
 # SVE predicate logical operations
-- 
2.14.3