target/arm: Implement MVE VCLZ

Implement the MVE VCLZ insn (and the necessary machinery
for MVE 1-input vector ops).

Note that for non-load instructions predication is always performed
at a byte level granularity regardless of element size (R_ZLSJ),
and so the masking logic here differs from that used in the VLDR
and VSTR helpers.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210617121628.20116-4-peter.maydell@linaro.org

4 files changed, 132 insertions, 0 deletions

diff --git a/target/arm/helper-mve.h b/target/arm/helper-mve.h
index e47d4164ae..c5c1315b16 100644
--- a/target/arm/helper-mve.h
+++ b/target/arm/helper-mve.h
@@ -32,3 +32,7 @@ DEF_HELPER_FLAGS_3(mve_vldrh_uw, TCG_CALL_NO_WG, void, env, ptr, i32)
 DEF_HELPER_FLAGS_3(mve_vstrb_h, TCG_CALL_NO_WG, void, env, ptr, i32)
 DEF_HELPER_FLAGS_3(mve_vstrb_w, TCG_CALL_NO_WG, void, env, ptr, i32)
 DEF_HELPER_FLAGS_3(mve_vstrh_w, TCG_CALL_NO_WG, void, env, ptr, i32)
+
+DEF_HELPER_FLAGS_3(mve_vclzb, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vclzh, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vclzw, TCG_CALL_NO_WG, void, env, ptr, ptr)
diff --git a/target/arm/mve.decode b/target/arm/mve.decode
index 3bc5f03453..24999bf703 100644
--- a/target/arm/mve.decode
+++ b/target/arm/mve.decode
@@ -20,13 +20,17 @@
 #
 
 %qd 22:1 13:3
+%qm 5:1 1:3
 
 &vldr_vstr rn qd imm p a w size l u
+&1op qd qm size
 
 @vldr_vstr ....... . . . . l:1 rn:4 ... ...... imm:7 &vldr_vstr qd=%qd u=0
 # Note that both Rn and Qd are 3 bits only (no D bit)
 @vldst_wn ... u:1 ... . . . . l:1 . rn:3 qd:3 . ... .. imm:7 &vldr_vstr
 
+@1op .... .... .... size:2 .. .... .... .... .... &1op qd=%qd qm=%qm
+
 # Vector loads and stores
 
 # Widening loads and narrowing stores:
@@ -61,3 +65,7 @@ VLDR_VSTR        1110110 1 a:1 . w:1 . .... ... 111101 .......   @vldr_vstr \
                  size=1 p=1
 VLDR_VSTR        1110110 1 a:1 . w:1 . .... ... 111110 .......   @vldr_vstr \
                  size=2 p=1
+
+# Vector miscellaneous
+
+VCLZ             1111 1111 1 . 11 .. 00 ... 0 0100 11 . 0 ... 0 @1op
diff --git a/target/arm/mve_helper.c b/target/arm/mve_helper.c
index 3c2b036c9c..f2fae523e2 100644
--- a/target/arm/mve_helper.c
+++ b/target/arm/mve_helper.c
@@ -181,3 +181,85 @@ DO_VSTR(vstrh_w, 2, stw, 4, int32_t)
 
 #undef DO_VLDR
 #undef DO_VSTR
+
+/*
+ * The mergemask(D, R, M) macro performs the operation "*D = R" but
+ * storing only the bytes which correspond to 1 bits in M,
+ * leaving other bytes in *D unchanged. We use _Generic
+ * to select the correct implementation based on the type of D.
+ */
+
+static void mergemask_ub(uint8_t *d, uint8_t r, uint16_t mask)
+{
+    if (mask & 1) {
+        *d = r;
+    }
+}
+
+static void mergemask_sb(int8_t *d, int8_t r, uint16_t mask)
+{
+    mergemask_ub((uint8_t *)d, r, mask);
+}
+
+static void mergemask_uh(uint16_t *d, uint16_t r, uint16_t mask)
+{
+    uint16_t bmask = expand_pred_b_data[mask & 3];
+    *d = (*d & ~bmask) | (r & bmask);
+}
+
+static void mergemask_sh(int16_t *d, int16_t r, uint16_t mask)
+{
+    mergemask_uh((uint16_t *)d, r, mask);
+}
+
+static void mergemask_uw(uint32_t *d, uint32_t r, uint16_t mask)
+{
+    uint32_t bmask = expand_pred_b_data[mask & 0xf];
+    *d = (*d & ~bmask) | (r & bmask);
+}
+
+static void mergemask_sw(int32_t *d, int32_t r, uint16_t mask)
+{
+    mergemask_uw((uint32_t *)d, r, mask);
+}
+
+static void mergemask_uq(uint64_t *d, uint64_t r, uint16_t mask)
+{
+    uint64_t bmask = expand_pred_b_data[mask & 0xff];
+    *d = (*d & ~bmask) | (r & bmask);
+}
+
+static void mergemask_sq(int64_t *d, int64_t r, uint16_t mask)
+{
+    mergemask_uq((uint64_t *)d, r, mask);
+}
+
+#define mergemask(D, R, M)                      \
+    _Generic(D,                                 \
+             uint8_t *: mergemask_ub,           \
+             int8_t *:  mergemask_sb,           \
+             uint16_t *: mergemask_uh,          \
+             int16_t *:  mergemask_sh,          \
+             uint32_t *: mergemask_uw,          \
+             int32_t *:  mergemask_sw,          \
+             uint64_t *: mergemask_uq,          \
+             int64_t *:  mergemask_sq)(D, R, M)
+
+#define DO_1OP(OP, ESIZE, TYPE, FN)                                     \
+    void HELPER(mve_##OP)(CPUARMState *env, void *vd, void *vm)         \
+    {                                                                   \
+        TYPE *d = vd, *m = vm;                                          \
+        uint16_t mask = mve_element_mask(env);                          \
+        unsigned e;                                                     \
+        for (e = 0; e < 16 / ESIZE; e++, mask >>= ESIZE) {              \
+            mergemask(&d[H##ESIZE(e)], FN(m[H##ESIZE(e)]), mask);       \
+        }                                                               \
+        mve_advance_vpt(env);                                           \
+    }
+
+#define DO_CLZ_B(N)   (clz32(N) - 24)
+#define DO_CLZ_H(N)   (clz32(N) - 16)
+
+DO_1OP(vclzb, 1, uint8_t, DO_CLZ_B)
+DO_1OP(vclzh, 2, uint16_t, DO_CLZ_H)
+DO_1OP(vclzw, 4, uint32_t, clz32)
diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c
index 4cabdf7a69..9eb6a68c97 100644
--- a/target/arm/translate-mve.c
+++ b/target/arm/translate-mve.c
@@ -29,6 +29,7 @@
 #include "decode-mve.c.inc"
 
 typedef void MVEGenLdStFn(TCGv_ptr, TCGv_ptr, TCGv_i32);
+typedef void MVEGenOneOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr);
 
 /* Return the offset of a Qn register (same semantics as aa32_vfp_qreg()) */
 static inline long mve_qreg_offset(unsigned reg)
@@ -160,3 +161,40 @@ static bool trans_VLDR_VSTR(DisasContext *s, arg_VLDR_VSTR *a)
 DO_VLDST_WIDE_NARROW(VLDSTB_H, vldrb_sh, vldrb_uh, vstrb_h)
 DO_VLDST_WIDE_NARROW(VLDSTB_W, vldrb_sw, vldrb_uw, vstrb_w)
 DO_VLDST_WIDE_NARROW(VLDSTH_W, vldrh_sw, vldrh_uw, vstrh_w)
+
+static bool do_1op(DisasContext *s, arg_1op *a, MVEGenOneOpFn fn)
+{
+    TCGv_ptr qd, qm;
+
+    if (!dc_isar_feature(aa32_mve, s) ||
+        !mve_check_qreg_bank(s, a->qd | a->qm) ||
+        !fn) {
+        return false;
+    }
+
+    if (!mve_eci_check(s) || !vfp_access_check(s)) {
+        return true;
+    }
+
+    qd = mve_qreg_ptr(a->qd);
+    qm = mve_qreg_ptr(a->qm);
+    fn(cpu_env, qd, qm);
+    tcg_temp_free_ptr(qd);
+    tcg_temp_free_ptr(qm);
+    mve_update_eci(s);
+    return true;
+}
+
+#define DO_1OP(INSN, FN)                                        \
+    static bool trans_##INSN(DisasContext *s, arg_1op *a)       \
+    {                                                           \
+        static MVEGenOneOpFn * const fns[] = {                  \
+            gen_helper_mve_##FN##b,                             \
+            gen_helper_mve_##FN##h,                             \
+            gen_helper_mve_##FN##w,                             \
+            NULL,                                               \
+        };                                                      \
+        return do_1op(s, a, fns[a->size]);                      \
+    }
+
+DO_1OP(VCLZ, vclz)