[PATCH v3 27/44] target/arm: Implement MVE VQADD and VQSUB

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

Implement the MVE VQADD and VQSUB insns, which perform saturating
addition of a scalar to each element.  Note that individual bytes of
each result element are used or discarded according to the predicate
mask, but FPSCR.QC is only set if the predicate mask for the lowest
byte of the element is set.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/arm/helper-mve.h    | 16 ++++++++++
 target/arm/mve.decode      |  5 +++
 target/arm/mve_helper.c    | 62 ++++++++++++++++++++++++++++++++++++++
 target/arm/translate-mve.c |  4 +++
 4 files changed, 87 insertions(+)

diff --git a/target/arm/helper-mve.h b/target/arm/helper-mve.h
index 1b807e1cf5f..092efdab475 100644
--- a/target/arm/helper-mve.h
+++ b/target/arm/helper-mve.h
@@ -173,6 +173,22 @@ DEF_HELPER_FLAGS_4(mve_vhsubu_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vhsubu_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vhsubu_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 
+DEF_HELPER_FLAGS_4(mve_vqadds_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqadds_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqadds_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(mve_vqaddu_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqaddu_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqaddu_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(mve_vqsubs_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqsubs_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqsubs_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(mve_vqsubu_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqsubu_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(mve_vqsubu_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
+
 DEF_HELPER_FLAGS_4(mve_vbrsrb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vbrsrh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vbrsrw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
diff --git a/target/arm/mve.decode b/target/arm/mve.decode
index e189e2de648..c85227c675a 100644
--- a/target/arm/mve.decode
+++ b/target/arm/mve.decode
@@ -167,6 +167,11 @@ VHADD_S_scalar   1110 1110 0 . .. ... 0 ... 0 1111 . 100 .... @2scalar
 VHADD_U_scalar   1111 1110 0 . .. ... 0 ... 0 1111 . 100 .... @2scalar
 VHSUB_S_scalar   1110 1110 0 . .. ... 0 ... 1 1111 . 100 .... @2scalar
 VHSUB_U_scalar   1111 1110 0 . .. ... 0 ... 1 1111 . 100 .... @2scalar
+
+VQADD_S_scalar   1110 1110 0 . .. ... 0 ... 0 1111 . 110 .... @2scalar
+VQADD_U_scalar   1111 1110 0 . .. ... 0 ... 0 1111 . 110 .... @2scalar
+VQSUB_S_scalar   1110 1110 0 . .. ... 0 ... 1 1111 . 110 .... @2scalar
+VQSUB_U_scalar   1111 1110 0 . .. ... 0 ... 1 1111 . 110 .... @2scalar
 VBRSR            1111 1110 0 . .. ... 1 ... 1 1110 . 110 .... @2scalar
 
 # Predicate operations
diff --git a/target/arm/mve_helper.c b/target/arm/mve_helper.c
index 775c667561e..0ae2aeadbb2 100644
--- a/target/arm/mve_helper.c
+++ b/target/arm/mve_helper.c
@@ -505,6 +505,33 @@ DO_2OP_U(vhaddu, do_vhadd_u)
 DO_2OP_S(vhsubs, do_vhsub_s)
 DO_2OP_U(vhsubu, do_vhsub_u)
 
+static inline int32_t do_sat_bhw(int64_t val, int64_t min, int64_t max, bool *s)
+{
+    if (val > max) {
+        *s = true;
+        return max;
+    } else if (val < min) {
+        *s = true;
+        return min;
+    }
+    return val;
+}
+
+#define DO_SQADD_B(n, m, s) do_sat_bhw((int64_t)n + m, INT8_MIN, INT8_MAX, s)
+#define DO_SQADD_H(n, m, s) do_sat_bhw((int64_t)n + m, INT16_MIN, INT16_MAX, s)
+#define DO_SQADD_W(n, m, s) do_sat_bhw((int64_t)n + m, INT32_MIN, INT32_MAX, s)
+
+#define DO_UQADD_B(n, m, s) do_sat_bhw((int64_t)n + m, 0, UINT8_MAX, s)
+#define DO_UQADD_H(n, m, s) do_sat_bhw((int64_t)n + m, 0, UINT16_MAX, s)
+#define DO_UQADD_W(n, m, s) do_sat_bhw((int64_t)n + m, 0, UINT32_MAX, s)
+
+#define DO_SQSUB_B(n, m, s) do_sat_bhw((int64_t)n - m, INT8_MIN, INT8_MAX, s)
+#define DO_SQSUB_H(n, m, s) do_sat_bhw((int64_t)n - m, INT16_MIN, INT16_MAX, s)
+#define DO_SQSUB_W(n, m, s) do_sat_bhw((int64_t)n - m, INT32_MIN, INT32_MAX, s)
+
+#define DO_UQSUB_B(n, m, s) do_sat_bhw((int64_t)n - m, 0, UINT8_MAX, s)
+#define DO_UQSUB_H(n, m, s) do_sat_bhw((int64_t)n - m, 0, UINT16_MAX, s)
+#define DO_UQSUB_W(n, m, s) do_sat_bhw((int64_t)n - m, 0, UINT32_MAX, s)
 
 #define DO_2OP_SCALAR(OP, ESIZE, TYPE, FN)                              \
     void HELPER(glue(mve_, OP))(CPUARMState *env, void *vd, void *vn,   \
@@ -520,6 +547,27 @@ DO_2OP_U(vhsubu, do_vhsub_u)
         mve_advance_vpt(env);                                           \
     }
 
+#define DO_2OP_SAT_SCALAR(OP, ESIZE, TYPE, FN)                          \
+    void HELPER(glue(mve_, OP))(CPUARMState *env, void *vd, void *vn,   \
+                                uint32_t rm)                            \
+    {                                                                   \
+        TYPE *d = vd, *n = vn;                                          \
+        TYPE m = rm;                                                    \
+        uint16_t mask = mve_element_mask(env);                          \
+        unsigned e;                                                     \
+        bool qc = false;                                                \
+        for (e = 0; e < 16 / ESIZE; e++, mask >>= ESIZE) {              \
+            bool sat = false;                                           \
+            mergemask(&d[H##ESIZE(e)], FN(n[H##ESIZE(e)], m, &sat),     \
+                      mask);                                            \
+            qc |= sat & mask & 1;                                       \
+        }                                                               \
+        if (qc) {                                                       \
+            env->vfp.qc[0] = qc;                                        \
+        }                                                               \
+        mve_advance_vpt(env);                                           \
+    }
+
 /* provide unsigned 2-op scalar helpers for all sizes */
 #define DO_2OP_SCALAR_U(OP, FN)                 \
     DO_2OP_SCALAR(OP##b, 1, uint8_t, FN)        \
@@ -538,6 +586,20 @@ DO_2OP_SCALAR_U(vhaddu_scalar, do_vhadd_u)
 DO_2OP_SCALAR_S(vhsubs_scalar, do_vhsub_s)
 DO_2OP_SCALAR_U(vhsubu_scalar, do_vhsub_u)
 
+DO_2OP_SAT_SCALAR(vqaddu_scalarb, 1, uint8_t, DO_UQADD_B)
+DO_2OP_SAT_SCALAR(vqaddu_scalarh, 2, uint16_t, DO_UQADD_H)
+DO_2OP_SAT_SCALAR(vqaddu_scalarw, 4, uint32_t, DO_UQADD_W)
+DO_2OP_SAT_SCALAR(vqadds_scalarb, 1, int8_t, DO_SQADD_B)
+DO_2OP_SAT_SCALAR(vqadds_scalarh, 2, int16_t, DO_SQADD_H)
+DO_2OP_SAT_SCALAR(vqadds_scalarw, 4, int32_t, DO_SQADD_W)
+
+DO_2OP_SAT_SCALAR(vqsubu_scalarb, 1, uint8_t, DO_UQSUB_B)
+DO_2OP_SAT_SCALAR(vqsubu_scalarh, 2, uint16_t, DO_UQSUB_H)
+DO_2OP_SAT_SCALAR(vqsubu_scalarw, 4, uint32_t, DO_UQSUB_W)
+DO_2OP_SAT_SCALAR(vqsubs_scalarb, 1, int8_t, DO_SQSUB_B)
+DO_2OP_SAT_SCALAR(vqsubs_scalarh, 2, int16_t, DO_SQSUB_H)
+DO_2OP_SAT_SCALAR(vqsubs_scalarw, 4, int32_t, DO_SQSUB_W)
+
 static inline uint32_t do_vbrsrb(uint32_t n, uint32_t m)
 {
     m &= 0xff;
diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c
index 7c4c06e434c..27c69d9c7dd 100644
--- a/target/arm/translate-mve.c
+++ b/target/arm/translate-mve.c
@@ -446,6 +446,10 @@ DO_2OP_SCALAR(VHADD_S_scalar, vhadds_scalar)
 DO_2OP_SCALAR(VHADD_U_scalar, vhaddu_scalar)
 DO_2OP_SCALAR(VHSUB_S_scalar, vhsubs_scalar)
 DO_2OP_SCALAR(VHSUB_U_scalar, vhsubu_scalar)
+DO_2OP_SCALAR(VQADD_S_scalar, vqadds_scalar)
+DO_2OP_SCALAR(VQADD_U_scalar, vqaddu_scalar)
+DO_2OP_SCALAR(VQSUB_S_scalar, vqsubs_scalar)
+DO_2OP_SCALAR(VQSUB_U_scalar, vqsubu_scalar)
 DO_2OP_SCALAR(VBRSR, vbrsr)
 
 static bool do_long_dual_acc(DisasContext *s, arg_vmlaldav *a,
-- 
2.20.1