[Qemu-devel] [PULL SUBSYSTEM s390x 05/33] s390x/tcg: Implement VECTOR STRING RANGE COMPARE

From: David Hildenbrand <david@redhat.com>
To: qemu-devel@nongnu.org
Cc: Thomas Huth <thuth@redhat.com>,
	David Hildenbrand <david@redhat.com>,
	Cornelia Huck <cohuck@redhat.com>,
	Richard Henderson <richard.henderson@linaro.org>,
	qemu-s390x@nongnu.org, Richard Henderson <rth@twiddle.net>
Subject: [Qemu-devel] [PULL SUBSYSTEM s390x 05/33] s390x/tcg: Implement VECTOR STRING RANGE COMPARE
Date: Wed,  5 Jun 2019 22:33:35 +0200	[thread overview]
Message-ID: <20190605203403.29461-6-david@redhat.com> (raw)
In-Reply-To: <20190605203403.29461-1-david@redhat.com>

Unfortunately, there is no easy way to avoid looping over all elements
in v2. Provide specialized variants for !cc,!rt/!cc,rt/cc,!rt/cc,rt and
all element types. Especially for different values of rt, the compiler
might be able to optimize the code a lot.

Add s390_vec_write_element().

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 target/s390x/helper.h            |  12 +++
 target/s390x/insn-data.def       |   2 +
 target/s390x/translate_vx.inc.c  |  59 +++++++++++++
 target/s390x/vec.h               |  21 +++++
 target/s390x/vec_string_helper.c | 143 +++++++++++++++++++++++++++++++
 5 files changed, 237 insertions(+)

diff --git a/target/s390x/helper.h b/target/s390x/helper.h
index 1f9f0b463b..5db67779d3 100644
--- a/target/s390x/helper.h
+++ b/target/s390x/helper.h
@@ -236,6 +236,18 @@ DEF_HELPER_FLAGS_3(gvec_vistr32, TCG_CALL_NO_RWG, void, ptr, cptr, i32)
 DEF_HELPER_4(gvec_vistr_cc8, void, ptr, cptr, env, i32)
 DEF_HELPER_4(gvec_vistr_cc16, void, ptr, cptr, env, i32)
 DEF_HELPER_4(gvec_vistr_cc32, void, ptr, cptr, env, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc8, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc16, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc32, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc_rt8, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc_rt16, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_FLAGS_5(gvec_vstrc_rt32, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32)
+DEF_HELPER_6(gvec_vstrc_cc8, void, ptr, cptr, cptr, cptr, env, i32)
+DEF_HELPER_6(gvec_vstrc_cc16, void, ptr, cptr, cptr, cptr, env, i32)
+DEF_HELPER_6(gvec_vstrc_cc32, void, ptr, cptr, cptr, cptr, env, i32)
+DEF_HELPER_6(gvec_vstrc_cc_rt8, void, ptr, cptr, cptr, cptr, env, i32)
+DEF_HELPER_6(gvec_vstrc_cc_rt16, void, ptr, cptr, cptr, cptr, env, i32)
+DEF_HELPER_6(gvec_vstrc_cc_rt32, void, ptr, cptr, cptr, cptr, env, i32)
 
 #ifndef CONFIG_USER_ONLY
 DEF_HELPER_3(servc, i32, env, i64, i64)
diff --git a/target/s390x/insn-data.def b/target/s390x/insn-data.def
index b4a6b59608..a2969fab58 100644
--- a/target/s390x/insn-data.def
+++ b/target/s390x/insn-data.def
@@ -1201,6 +1201,8 @@
     F(0xe781, VFENE,   VRR_b, V,   0, 0, 0, 0, vfene, 0, IF_VEC)
 /* VECTOR ISOLATE STRING */
     F(0xe75c, VISTR,   VRR_a, V,   0, 0, 0, 0, vistr, 0, IF_VEC)
+/* VECTOR STRING RANGE COMPARE */
+    F(0xe78a, VSTRC,   VRR_d, V,   0, 0, 0, 0, vstrc, 0, IF_VEC)
 
 #ifndef CONFIG_USER_ONLY
 /* COMPARE AND SWAP AND PURGE */
diff --git a/target/s390x/translate_vx.inc.c b/target/s390x/translate_vx.inc.c
index 08a62eab52..f26ffa2895 100644
--- a/target/s390x/translate_vx.inc.c
+++ b/target/s390x/translate_vx.inc.c
@@ -217,6 +217,10 @@ static void get_vec_element_ptr_i64(TCGv_ptr ptr, uint8_t reg, TCGv_i64 enr,
     tcg_gen_gvec_4_ool(vec_full_reg_offset(v1), vec_full_reg_offset(v2), \
                        vec_full_reg_offset(v3), vec_full_reg_offset(v4), \
                        16, 16, data, fn)
+#define gen_gvec_4_ptr(v1, v2, v3, v4, ptr, data, fn) \
+    tcg_gen_gvec_4_ptr(vec_full_reg_offset(v1), vec_full_reg_offset(v2), \
+                       vec_full_reg_offset(v3), vec_full_reg_offset(v4), \
+                       ptr, 16, 16, data, fn)
 #define gen_gvec_dup_i64(es, v1, c) \
     tcg_gen_gvec_dup_i64(es, vec_full_reg_offset(v1), 16, 16, c)
 #define gen_gvec_mov(v1, v2) \
@@ -2479,3 +2483,58 @@ static DisasJumpType op_vistr(DisasContext *s, DisasOps *o)
     }
     return DISAS_NEXT;
 }
+
+static DisasJumpType op_vstrc(DisasContext *s, DisasOps *o)
+{
+    const uint8_t es = get_field(s->fields, m5);
+    const uint8_t m6 = get_field(s->fields, m6);
+    static gen_helper_gvec_4 * const g[3] = {
+        gen_helper_gvec_vstrc8,
+        gen_helper_gvec_vstrc16,
+        gen_helper_gvec_vstrc32,
+    };
+    static gen_helper_gvec_4 * const g_rt[3] = {
+        gen_helper_gvec_vstrc_rt8,
+        gen_helper_gvec_vstrc_rt16,
+        gen_helper_gvec_vstrc_rt32,
+    };
+    static gen_helper_gvec_4_ptr * const g_cc[3] = {
+        gen_helper_gvec_vstrc_cc8,
+        gen_helper_gvec_vstrc_cc16,
+        gen_helper_gvec_vstrc_cc32,
+    };
+    static gen_helper_gvec_4_ptr * const g_cc_rt[3] = {
+        gen_helper_gvec_vstrc_cc_rt8,
+        gen_helper_gvec_vstrc_cc_rt16,
+        gen_helper_gvec_vstrc_cc_rt32,
+    };
+
+    if (es > ES_32) {
+        gen_program_exception(s, PGM_SPECIFICATION);
+        return DISAS_NORETURN;
+    }
+
+    if (extract32(m6, 0, 1)) {
+        if (extract32(m6, 2, 1)) {
+            gen_gvec_4_ptr(get_field(s->fields, v1), get_field(s->fields, v2),
+                           get_field(s->fields, v3), get_field(s->fields, v4),
+                           cpu_env, m6, g_cc_rt[es]);
+        } else {
+            gen_gvec_4_ptr(get_field(s->fields, v1), get_field(s->fields, v2),
+                           get_field(s->fields, v3), get_field(s->fields, v4),
+                           cpu_env, m6, g_cc[es]);
+        }
+        set_cc_static(s);
+    } else {
+        if (extract32(m6, 2, 1)) {
+            gen_gvec_4_ool(get_field(s->fields, v1), get_field(s->fields, v2),
+                           get_field(s->fields, v3), get_field(s->fields, v4),
+                           m6, g_rt[es]);
+        } else {
+            gen_gvec_4_ool(get_field(s->fields, v1), get_field(s->fields, v2),
+                           get_field(s->fields, v3), get_field(s->fields, v4),
+                           m6, g[es]);
+        }
+    }
+    return DISAS_NEXT;
+}
diff --git a/target/s390x/vec.h b/target/s390x/vec.h
index affc62874c..a6e361869b 100644
--- a/target/s390x/vec.h
+++ b/target/s390x/vec.h
@@ -117,4 +117,25 @@ static inline void s390_vec_write_element64(S390Vector *v, uint8_t enr,
     v->doubleword[enr] = data;
 }
 
+static inline void s390_vec_write_element(S390Vector *v, uint8_t enr,
+                                          uint8_t es, uint64_t data)
+{
+    switch (es) {
+    case MO_8:
+        s390_vec_write_element8(v, enr, data);
+        break;
+    case MO_16:
+        s390_vec_write_element16(v, enr, data);
+        break;
+    case MO_32:
+        s390_vec_write_element32(v, enr, data);
+        break;
+    case MO_64:
+        s390_vec_write_element64(v, enr, data);
+        break;
+    default:
+        g_assert_not_reached();
+    }
+}
+
 #endif /* S390X_VEC_H */
diff --git a/target/s390x/vec_string_helper.c b/target/s390x/vec_string_helper.c
index 6bafa23bd7..c516c0ceeb 100644
--- a/target/s390x/vec_string_helper.c
+++ b/target/s390x/vec_string_helper.c
@@ -328,3 +328,146 @@ void HELPER(gvec_vistr_cc##BITS)(void *v1, const void *v2, CPUS390XState *env, \
 DEF_VISTR_CC_HELPER(8)
 DEF_VISTR_CC_HELPER(16)
 DEF_VISTR_CC_HELPER(32)
+
+static bool element_compare(uint32_t data, uint32_t l, uint8_t c)
+{
+    const bool equal = extract32(c, 7, 1);
+    const bool lower = extract32(c, 6, 1);
+    const bool higher = extract32(c, 5, 1);
+
+    if (data < l) {
+        return lower;
+    } else if (data > l) {
+        return higher;
+    }
+    return equal;
+}
+
+static int vstrc(void *v1, const void *v2, const void *v3, const void *v4,
+                 bool in, bool rt, bool zs, uint8_t es)
+{
+    const uint64_t mask = get_element_lsbs_mask(es);
+    uint64_t a0 = s390_vec_read_element64(v2, 0);
+    uint64_t a1 = s390_vec_read_element64(v2, 1);
+    int first_zero = 16, first_match = 16;
+    S390Vector rt_result = {};
+    uint64_t z0, z1;
+    int i, j;
+
+    if (zs) {
+        z0 = zero_search(a0, mask);
+        z1 = zero_search(a1, mask);
+        first_zero = match_index(z0, z1);
+    }
+
+    for (i = 0; i < 16 / (1 << es); i++) {
+        const uint32_t data = s390_vec_read_element(v2, i, es);
+        const int cur_byte = i * (1 << es);
+        bool any_match = false;
+
+        /* if we don't need a bit vector, we can stop early */
+        if (cur_byte == first_zero && !rt) {
+            break;
+        }
+
+        for (j = 0; j < 16 / (1 << es); j += 2) {
+            const uint32_t l1 = s390_vec_read_element(v3, j, es);
+            const uint32_t l2 = s390_vec_read_element(v3, j + 1, es);
+            /* we are only interested in the highest byte of each element */
+            const uint8_t c1 = s390_vec_read_element8(v4, j * (1 << es));
+            const uint8_t c2 = s390_vec_read_element8(v4, (j + 1) * (1 << es));
+
+            if (element_compare(data, l1, c1) &&
+                element_compare(data, l2, c2)) {
+                any_match = true;
+                break;
+            }
+        }
+        /* invert the result if requested */
+        any_match = in ^ any_match;
+
+        if (any_match) {
+            /* indicate bit vector if requested */
+            if (rt) {
+                const uint64_t val = -1ull;
+
+                first_match = MIN(cur_byte, first_match);
+                s390_vec_write_element(&rt_result, i, es, val);
+            } else {
+                /* stop on the first match */
+                first_match = cur_byte;
+                break;
+            }
+        }
+    }
+
+    if (rt) {
+        *(S390Vector *)v1 = rt_result;
+    } else {
+        s390_vec_write_element64(v1, 0, MIN(first_match, first_zero));
+        s390_vec_write_element64(v1, 1, 0);
+    }
+
+    if (first_zero == 16 && first_match == 16) {
+        return 3; /* no match */
+    } else if (first_zero == 16) {
+        return 1; /* matching elements, no match for zero */
+    } else if (first_match < first_zero) {
+        return 2; /* matching elements before match for zero */
+    }
+    return 0; /* match for zero */
+}
+
+#define DEF_VSTRC_HELPER(BITS)                                                 \
+void HELPER(gvec_vstrc##BITS)(void *v1, const void *v2, const void *v3,        \
+                              const void *v4, uint32_t desc)                   \
+{                                                                              \
+    const bool in = extract32(simd_data(desc), 3, 1);                          \
+    const bool zs = extract32(simd_data(desc), 1, 1);                          \
+                                                                               \
+    vstrc(v1, v2, v3, v4, in, 0, zs, MO_##BITS);                               \
+}
+DEF_VSTRC_HELPER(8)
+DEF_VSTRC_HELPER(16)
+DEF_VSTRC_HELPER(32)
+
+#define DEF_VSTRC_RT_HELPER(BITS)                                              \
+void HELPER(gvec_vstrc_rt##BITS)(void *v1, const void *v2, const void *v3,     \
+                                 const void *v4, uint32_t desc)                \
+{                                                                              \
+    const bool in = extract32(simd_data(desc), 3, 1);                          \
+    const bool zs = extract32(simd_data(desc), 1, 1);                          \
+                                                                               \
+    vstrc(v1, v2, v3, v4, in, 1, zs, MO_##BITS);                               \
+}
+DEF_VSTRC_RT_HELPER(8)
+DEF_VSTRC_RT_HELPER(16)
+DEF_VSTRC_RT_HELPER(32)
+
+#define DEF_VSTRC_CC_HELPER(BITS)                                              \
+void HELPER(gvec_vstrc_cc##BITS)(void *v1, const void *v2, const void *v3,     \
+                                 const void *v4, CPUS390XState *env,           \
+                                 uint32_t desc)                                \
+{                                                                              \
+    const bool in = extract32(simd_data(desc), 3, 1);                          \
+    const bool zs = extract32(simd_data(desc), 1, 1);                          \
+                                                                               \
+    env->cc_op = vstrc(v1, v2, v3, v4, in, 0, zs, MO_##BITS);                  \
+}
+DEF_VSTRC_CC_HELPER(8)
+DEF_VSTRC_CC_HELPER(16)
+DEF_VSTRC_CC_HELPER(32)
+
+#define DEF_VSTRC_CC_RT_HELPER(BITS)                                           \
+void HELPER(gvec_vstrc_cc_rt##BITS)(void *v1, const void *v2, const void *v3,  \
+                                    const void *v4, CPUS390XState *env,        \
+                                    uint32_t desc)                             \
+{                                                                              \
+    const bool in = extract32(simd_data(desc), 3, 1);                          \
+    const bool zs = extract32(simd_data(desc), 1, 1);                          \
+                                                                               \
+    env->cc_op = vstrc(v1, v2, v3, v4, in, 1, zs, MO_##BITS);                  \
+}
+DEF_VSTRC_CC_RT_HELPER(8)
+DEF_VSTRC_CC_RT_HELPER(16)
+DEF_VSTRC_CC_RT_HELPER(32)
-- 
2.21.0


