[Qemu-devel] [PATCH v2 05/10] target-arm: A64: Add SIMD TBL/TBLX

[Peter Maydell <peter.maydell@linaro.org>]

From: Michael Matz <matz@suse.de>

Add support for the SIMD TBL/TBLX instructions (group C3.6.2).

Signed-off-by: Michael Matz <matz@suse.de>
[PMM: rewritten to do more of the decode in translate-a64.c,
 and to do only one 64 bit pass at a time in the helper]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
---
v1->v2: change rn and numregs args to helper to i32
---
 target-arm/helper-a64.c    | 31 ++++++++++++++++++++++++++
 target-arm/helper-a64.h    |  1 +
 target-arm/translate-a64.c | 55 +++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 86 insertions(+), 1 deletion(-)

diff --git a/target-arm/helper-a64.c b/target-arm/helper-a64.c
index 4ce0d01..6ca958a 100644
--- a/target-arm/helper-a64.c
+++ b/target-arm/helper-a64.c
@@ -122,3 +122,34 @@ uint64_t HELPER(vfp_cmped_a64)(float64 x, float64 y, void *fp_status)
 {
     return float_rel_to_flags(float64_compare(x, y, fp_status));
 }
+
+uint64_t HELPER(simd_tbl)(CPUARMState *env, uint64_t result, uint64_t indices,
+                          uint32_t rn, uint32_t numregs)
+{
+    /* Helper function for SIMD TBL and TBX. We have to do the table
+     * lookup part for the 64 bits worth of indices we're passed in.
+     * result is the initial results vector (either zeroes for TBL
+     * or some guest values for TBX), rn the register number where
+     * the table starts, and numregs the number of registers in the table.
+     * We return the results of the lookups.
+     */
+    int shift;
+
+    for (shift = 0; shift < 64; shift += 8) {
+        int index = extract64(indices, shift, 8);
+        if (index < 16 * numregs) {
+            /* Convert index (a byte offset into the virtual table
+             * which is a series of 128-bit vectors concatenated)
+             * into the correct vfp.regs[] element plus a bit offset
+             * into that element, bearing in mind that the table
+             * can wrap around from V31 to V0.
+             */
+            int elt = (rn * 2 + (index >> 3)) % 64;
+            int bitidx = (index & 7) * 8;
+            uint64_t val = extract64(env->vfp.regs[elt], bitidx, 8);
+
+            result = deposit64(result, shift, 8, val);
+        }
+    }
+    return result;
+}
diff --git a/target-arm/helper-a64.h b/target-arm/helper-a64.h
index bca19f3..99832ee 100644
--- a/target-arm/helper-a64.h
+++ b/target-arm/helper-a64.h
@@ -26,3 +26,4 @@ DEF_HELPER_3(vfp_cmps_a64, i64, f32, f32, ptr)
 DEF_HELPER_3(vfp_cmpes_a64, i64, f32, f32, ptr)
 DEF_HELPER_3(vfp_cmpd_a64, i64, f64, f64, ptr)
 DEF_HELPER_3(vfp_cmped_a64, i64, f64, f64, ptr)
+DEF_HELPER_FLAGS_5(simd_tbl, TCG_CALL_NO_RWG_SE, i64, env, i64, i64, i32, i32)
diff --git a/target-arm/translate-a64.c b/target-arm/translate-a64.c
index 219af78..3eae175 100644
--- a/target-arm/translate-a64.c
+++ b/target-arm/translate-a64.c
@@ -4743,7 +4743,60 @@ static void disas_simd_ext(DisasContext *s, uint32_t insn)
  */
 static void disas_simd_tb(DisasContext *s, uint32_t insn)
 {
-    unsupported_encoding(s, insn);
+    int op2 = extract32(insn, 22, 2);
+    int is_q = extract32(insn, 30, 1);
+    int rm = extract32(insn, 16, 5);
+    int rn = extract32(insn, 5, 5);
+    int rd = extract32(insn, 0, 5);
+    int is_tblx = extract32(insn, 12, 1);
+    int len = extract32(insn, 13, 2);
+    TCGv_i64 tcg_resl, tcg_resh, tcg_idx;
+    TCGv_i32 tcg_regno, tcg_numregs;
+
+    if (op2 != 0) {
+        unallocated_encoding(s);
+        return;
+    }
+
+    /* This does a table lookup: for every byte element in the input
+     * we index into a table formed from up to four vector registers,
+     * and then the output is the result of the lookups. Our helper
+     * function does the lookup operation for a single 64 bit part of
+     * the input.
+     */
+    tcg_resl = tcg_temp_new_i64();
+    tcg_resh = tcg_temp_new_i64();
+
+    if (is_tblx) {
+        read_vec_element(s, tcg_resl, rd, 0, MO_64);
+    } else {
+        tcg_gen_movi_i64(tcg_resl, 0);
+    }
+    if (is_tblx && is_q) {
+        read_vec_element(s, tcg_resh, rd, 1, MO_64);
+    } else {
+        tcg_gen_movi_i64(tcg_resh, 0);
+    }
+
+    tcg_idx = tcg_temp_new_i64();
+    tcg_regno = tcg_const_i32(rn);
+    tcg_numregs = tcg_const_i32(len + 1);
+    read_vec_element(s, tcg_idx, rm, 0, MO_64);
+    gen_helper_simd_tbl(tcg_resl, cpu_env, tcg_resl, tcg_idx,
+                        tcg_regno, tcg_numregs);
+    if (is_q) {
+        read_vec_element(s, tcg_idx, rm, 1, MO_64);
+        gen_helper_simd_tbl(tcg_resh, cpu_env, tcg_resh, tcg_idx,
+                            tcg_regno, tcg_numregs);
+    }
+    tcg_temp_free_i64(tcg_idx);
+    tcg_temp_free_i32(tcg_regno);
+    tcg_temp_free_i32(tcg_numregs);
+
+    write_vec_element(s, tcg_resl, rd, 0, MO_64);
+    tcg_temp_free_i64(tcg_resl);
+    write_vec_element(s, tcg_resh, rd, 1, MO_64);
+    tcg_temp_free_i64(tcg_resh);
 }
 
 /* C3.6.3 ZIP/UZP/TRN
-- 
1.8.5