@@ -692,35 +692,98 @@ DEF_HELPER_FLAGS_2(rddsp, 0, tl, tl, env)
/* MIPS SIMD Architecture */
+DEF_HELPER_5(msa_add_a_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_adds_a_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_adds_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_adds_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_addv_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_addvi_df, void, env, i32, i32, i32, s64)
DEF_HELPER_4(msa_andi_b, void, env, i32, i32, i32)
+DEF_HELPER_5(msa_asub_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_asub_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ave_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ave_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_aver_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_aver_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_bclr_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_bclri_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_binsl_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_binsli_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_binsr_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_binsri_df, void, env, i32, i32, i32, i32)
DEF_HELPER_4(msa_bmnzi_b, void, env, i32, i32, i32)
DEF_HELPER_4(msa_bmzi_b, void, env, i32, i32, i32)
+DEF_HELPER_5(msa_bneg_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_bnegi_df, void, env, i32, i32, i32, i32)
DEF_HELPER_4(msa_bseli_b, void, env, i32, i32, i32)
+DEF_HELPER_5(msa_bset_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_bseti_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ceq_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_ceqi_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_cle_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_cle_u_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_clei_s_df, void, env, i32, i32, i32, s64)
DEF_HELPER_5(msa_clei_u_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_clt_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_clt_u_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_clti_s_df, void, env, i32, i32, i32, s64)
DEF_HELPER_5(msa_clti_u_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_div_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_div_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dotp_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dotp_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dpadd_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dpadd_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dpsub_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_dpsub_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_hadd_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_hadd_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_hsub_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_hsub_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ilvev_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ilvl_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ilvod_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_ilvr_df, void, env, i32, i32, i32, i32)
DEF_HELPER_4(msa_ldi_df, void, env, i32, i32, i32)
+DEF_HELPER_5(msa_maddv_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_max_a_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_max_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_max_u_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_maxi_s_df, void, env, i32, i32, i32, s64)
DEF_HELPER_5(msa_maxi_u_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_min_a_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_min_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_min_u_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_mini_s_df, void, env, i32, i32, i32, s64)
DEF_HELPER_5(msa_mini_u_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_mod_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_mod_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_msubv_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_mulv_df, void, env, i32, i32, i32, i32)
DEF_HELPER_4(msa_nori_b, void, env, i32, i32, i32)
DEF_HELPER_4(msa_ori_b, void, env, i32, i32, i32)
+DEF_HELPER_5(msa_pckev_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_pckod_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_sat_s_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_sat_u_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_shf_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_sld_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_sll_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_slli_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_splat_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_sra_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_srai_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_srar_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_srari_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_srl_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_srli_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_srlr_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_srlri_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_subs_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_subs_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_subsus_u_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_subsuu_s_df, void, env, i32, i32, i32, i32)
+DEF_HELPER_5(msa_subv_df, void, env, i32, i32, i32, i32)
DEF_HELPER_5(msa_subvi_df, void, env, i32, i32, i32, s64)
+DEF_HELPER_5(msa_vshf_df, void, env, i32, i32, i32, i32)
DEF_HELPER_4(msa_xori_b, void, env, i32, i32, i32)
@@ -195,6 +195,48 @@ static inline void msa_store_wr_elem(CPUMIPSState *env, uint64_t val,
}
}
+static inline int64_t msa_add_a_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t abs_arg1 = arg1 >= 0 ? arg1 : -arg1;
+ uint64_t abs_arg2 = arg2 >= 0 ? arg2 : -arg2;
+ return abs_arg1 + abs_arg2;
+}
+
+void helper_msa_add_a_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_add_a_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_addv_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = (int64_t) ts + tt;
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
void helper_msa_addvi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
uint32_t ws, int64_t u5)
{
@@ -211,6 +253,23 @@ void helper_msa_addvi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+void helper_msa_subv_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = (int64_t) ts - tt;
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
void helper_msa_subvi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
uint32_t ws, int64_t u5)
{
@@ -227,75 +286,84 @@ void helper_msa_subvi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-void helper_msa_andi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
- uint32_t i8)
+static inline int64_t msa_adds_a_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
- void *pwd = &(env->active_fpu.fpr[wd]);
- void *pws = &(env->active_fpu.fpr[ws]);
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwd, i) = B(pws, i) & i8;
- } DONE_ALL_ELEMENTS;
- if (env->active_msa.msair & MSAIR_WRP_BIT) {
- env->active_msa.msamodify |= (1 << wd);
+ uint64_t max_int = (uint64_t)DF_MAX_INT(df);
+ uint64_t abs_arg1 = arg1 >= 0 ? arg1 : -arg1;
+ uint64_t abs_arg2 = arg2 >= 0 ? arg2 : -arg2;
+ if (abs_arg1 > max_int || abs_arg2 > max_int) {
+ return (int64_t)max_int;
+ } else {
+ return (abs_arg1 < max_int - abs_arg2) ? abs_arg1 + abs_arg2 : max_int;
}
}
-void helper_msa_ori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
- uint32_t i8)
+void helper_msa_adds_a_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- void *pwd = &(env->active_fpu.fpr[wd]);
- void *pws = &(env->active_fpu.fpr[ws]);
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwd, i) = B(pws, i) | i8;
- } DONE_ALL_ELEMENTS;
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_adds_a_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-void helper_msa_nori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
- uint32_t i8)
+static inline int64_t msa_adds_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
- void *pwd = &(env->active_fpu.fpr[wd]);
- void *pws = &(env->active_fpu.fpr[ws]);
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwd, i) = ~(B(pws, i) | i8);
- } DONE_ALL_ELEMENTS;
- if (env->active_msa.msair & MSAIR_WRP_BIT) {
- env->active_msa.msamodify |= (1 << wd);
+ int64_t max_int = DF_MAX_INT(df);
+ int64_t min_int = DF_MIN_INT(df);
+ if (arg1 < 0) {
+ return (min_int - arg1 < arg2) ? arg1 + arg2 : min_int;
+ } else {
+ return (arg2 < max_int - arg1) ? arg1 + arg2 : max_int;
}
}
-void helper_msa_xori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
- uint32_t i8)
+void helper_msa_adds_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- void *pwd = &(env->active_fpu.fpr[wd]);
- void *pws = &(env->active_fpu.fpr[ws]);
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwd, i) = B(pws, i) ^ i8;
- } DONE_ALL_ELEMENTS;
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_adds_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-static inline int64_t msa_bclr_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
- int64_t arg2)
+static inline uint64_t msa_adds_u_df(CPUMIPSState *env, uint32_t df,
+ uint64_t arg1, uint64_t arg2)
{
- int32_t b_arg2 = BIT_POSITION(arg2, df);
-
- return UNSIGNED(arg1 & (~(1LL << b_arg2)), df);
+ uint64_t max_uint = DF_MAX_UINT(df);
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return (u_arg1 < max_uint - u_arg2) ? u_arg1 + u_arg2 : max_uint;
}
-void helper_msa_bclri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t m)
+void helper_msa_adds_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
- ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_bclr_df(env, df, ts, m);
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_adds_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -303,22 +371,28 @@ void helper_msa_bclri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_bneg_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
- int64_t arg2)
+static inline int64_t msa_subs_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
- int32_t b_arg2 = BIT_POSITION(arg2, df);
- return UNSIGNED(arg1 ^ (1LL << b_arg2), df);
+ int64_t max_int = DF_MAX_INT(df);
+ int64_t min_int = DF_MIN_INT(df);
+ if (arg2 > 0) {
+ return (min_int + arg2 < arg1) ? arg1 - arg2 : min_int;
+ } else {
+ return (arg1 < max_int + arg2) ? arg1 - arg2 : max_int;
+ }
}
-void helper_msa_bnegi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t m)
+void helper_msa_subs_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_bneg_df(env, df, ts, m);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_subs_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -326,22 +400,24 @@ void helper_msa_bnegi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_bset_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
- int64_t arg2)
+static inline int64_t msa_subs_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
- int32_t b_arg2 = BIT_POSITION(arg2, df);
- return UNSIGNED(arg1 | (1LL << b_arg2), df);
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return (u_arg1 > u_arg2) ? u_arg1 - u_arg2 : 0;
}
-void helper_msa_bseti_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t m)
+void helper_msa_subs_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
- ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_bset_df(env, df, ts, m);
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_subs_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -349,31 +425,34 @@ void helper_msa_bseti_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_binsl_df(CPUMIPSState *env, uint32_t df,
- int64_t dest, int64_t arg1, int64_t arg2)
+static inline int64_t msa_subsuu_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
uint64_t u_arg1 = UNSIGNED(arg1, df);
- uint64_t u_dest = UNSIGNED(dest, df);
- int32_t sh_d = BIT_POSITION(arg2, df) + 1;
- int32_t sh_a = DF_BITS(df) - sh_d;
- if (sh_d == DF_BITS(df)) {
- return u_arg1;
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ int64_t max_int = DF_MAX_INT(df);
+ int64_t min_int = DF_MIN_INT(df);
+ if (u_arg1 > u_arg2) {
+ return u_arg1 - u_arg2 < (uint64_t)max_int ?
+ (int64_t)(u_arg1 - u_arg2) :
+ max_int;
} else {
- return UNSIGNED(UNSIGNED(u_dest << sh_d, df) >> sh_d, df) |
- UNSIGNED(UNSIGNED(u_arg1 >> sh_a, df) << sh_a, df);
+ return u_arg2 - u_arg1 < (uint64_t)(-min_int) ?
+ (int64_t)(u_arg1 - u_arg2) :
+ min_int;
}
}
-void helper_msa_binsli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t m)
+void helper_msa_subsuu_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_load_wr_elem_s64(env, wd, df, i);
- td = msa_binsl_df(env, df, td, ts, m);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_subsuu_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -381,31 +460,34 @@ void helper_msa_binsli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_binsr_df(CPUMIPSState *env, uint32_t df,
- int64_t dest, int64_t arg1, int64_t arg2)
+static inline int64_t msa_subsus_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
uint64_t u_arg1 = UNSIGNED(arg1, df);
- uint64_t u_dest = UNSIGNED(dest, df);
- int32_t sh_d = BIT_POSITION(arg2, df) + 1;
- int32_t sh_a = DF_BITS(df) - sh_d;
- if (sh_d == DF_BITS(df)) {
- return u_arg1;
+ uint64_t max_uint = DF_MAX_UINT(df);
+ if (arg2 >= 0) {
+ uint64_t u_arg2 = (uint64_t)arg2;
+ return (u_arg1 > u_arg2) ?
+ (int64_t)(u_arg1 - u_arg2) :
+ 0;
} else {
- return UNSIGNED(UNSIGNED(u_dest >> sh_d, df) << sh_d, df) |
- UNSIGNED(UNSIGNED(u_arg1 << sh_a, df) >> sh_a, df);
+ uint64_t u_arg2 = (uint64_t)(-arg2);
+ return (u_arg1 < max_uint - u_arg2) ?
+ (int64_t)(u_arg1 + u_arg2) :
+ (int64_t)max_uint;
}
}
-void helper_msa_binsri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t m)
+void helper_msa_subsus_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_load_wr_elem_s64(env, wd, df, i);
- td = msa_binsr_df(env, df, td, ts, m);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_subsus_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -413,69 +495,76 @@ void helper_msa_binsri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-#define BIT_MOVE_IF_NOT_ZERO(dest, arg1, arg2, df) \
- dest = UNSIGNED(((dest & (~arg2)) | (arg1 & arg2)), df)
-
-void helper_msa_bmnzi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+void helper_msa_andi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
uint32_t i8)
{
void *pwd = &(env->active_fpu.fpr[wd]);
void *pws = &(env->active_fpu.fpr[ws]);
ALL_B_ELEMENTS(i, MSA_WRLEN) {
- BIT_MOVE_IF_NOT_ZERO(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ B(pwd, i) = B(pws, i) & i8;
} DONE_ALL_ELEMENTS;
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-#define BIT_MOVE_IF_ZERO(dest, arg1, arg2, df) \
- dest = UNSIGNED((dest & arg2) | (arg1 & (~arg2)), df)
-
-void helper_msa_bmzi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+void helper_msa_ori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
uint32_t i8)
{
void *pwd = &(env->active_fpu.fpr[wd]);
void *pws = &(env->active_fpu.fpr[ws]);
ALL_B_ELEMENTS(i, MSA_WRLEN) {
- BIT_MOVE_IF_ZERO(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ B(pwd, i) = B(pws, i) | i8;
} DONE_ALL_ELEMENTS;
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-#define BIT_SELECT(dest, arg1, arg2, df) \
- dest = UNSIGNED((arg1 & (~dest)) | (arg2 & dest), df)
+void helper_msa_nori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+ uint32_t i8)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ B(pwd, i) = ~(B(pws, i) | i8);
+ } DONE_ALL_ELEMENTS;
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
-void helper_msa_bseli_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+void helper_msa_xori_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
uint32_t i8)
{
void *pwd = &(env->active_fpu.fpr[wd]);
void *pws = &(env->active_fpu.fpr[ws]);
ALL_B_ELEMENTS(i, MSA_WRLEN) {
- BIT_SELECT(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ B(pwd, i) = B(pws, i) ^ i8;
} DONE_ALL_ELEMENTS;
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-static inline int64_t msa_ceq_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
- int64_t arg2)
+static inline int64_t msa_asub_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
{
- return arg1 == arg2 ? -1 : 0;
+ /* signed compare */
+ return (arg1 < arg2) ?
+ (uint64_t)(arg2 - arg1) : (uint64_t)(arg1 - arg2);
}
-void helper_msa_ceqi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t i5)
+void helper_msa_asub_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_ceq_df(env, df, ts, i5);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_asub_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -483,21 +572,50 @@ void helper_msa_ceqi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_cle_s_df(CPUMIPSState *env, uint32_t df,
+static inline uint64_t msa_asub_u_df(CPUMIPSState *env, uint32_t df,
+ uint64_t arg1, uint64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ /* unsigned compare */
+ return (u_arg1 < u_arg2) ?
+ (uint64_t)(u_arg2 - u_arg1) : (uint64_t)(u_arg1 - u_arg2);
+}
+
+void helper_msa_asub_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_asub_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_ave_s_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
- return arg1 <= arg2 ? -1 : 0;
+ /* signed shift */
+ return (arg1 >> 1) + (arg2 >> 1) + (arg1 & arg2 & 1);
}
-void helper_msa_clei_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t s5)
+void helper_msa_ave_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_cle_s_df(env, df, ts, s5);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_ave_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -505,23 +623,25 @@ void helper_msa_clei_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_cle_u_df(CPUMIPSState *env, uint32_t df,
- int64_t arg1, int64_t arg2)
+static inline uint64_t msa_ave_u_df(CPUMIPSState *env, uint32_t df,
+ uint64_t arg1, uint64_t arg2)
{
uint64_t u_arg1 = UNSIGNED(arg1, df);
uint64_t u_arg2 = UNSIGNED(arg2, df);
- return u_arg1 <= u_arg2 ? -1 : 0;
+ /* unsigned shift */
+ return (u_arg1 >> 1) + (u_arg2 >> 1) + (u_arg1 & u_arg2 & 1);
}
-void helper_msa_clei_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t u5)
+void helper_msa_ave_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- uint64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_i64(env, ws, df, i);
- td = msa_cle_u_df(env, df, ts, u5);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_ave_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -529,21 +649,49 @@ void helper_msa_clei_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_clt_s_df(CPUMIPSState *env, uint32_t df,
+static inline int64_t msa_aver_s_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
- return arg1 < arg2 ? -1 : 0;
+ /* signed shift */
+ return (arg1 >> 1) + (arg2 >> 1) + ((arg1 | arg2) & 1);
}
-void helper_msa_clti_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t s5)
+void helper_msa_aver_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_clt_s_df(env, df, ts, s5);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_aver_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline uint64_t msa_aver_u_df(CPUMIPSState *env, uint32_t df,
+ uint64_t arg1, uint64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ /* unsigned shift */
+ return (u_arg1 >> 1) + (u_arg2 >> 1) + ((u_arg1 | u_arg2) & 1);
+}
+
+void helper_msa_aver_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_aver_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -551,85 +699,1520 @@ void helper_msa_clti_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_clt_u_df(CPUMIPSState *env, uint32_t df,
- int64_t arg1, int64_t arg2)
-{
- uint64_t u_arg1 = UNSIGNED(arg1, df);
- uint64_t u_arg2 = UNSIGNED(arg2, df);
- return u_arg1 < u_arg2 ? -1 : 0;
-}
-
-void helper_msa_clti_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t u5)
+static inline int64_t msa_bclr_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return UNSIGNED(arg1 & (~(1LL << b_arg2)), df);
+}
+
+void helper_msa_bclr_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_bclr_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_bclri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t m)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_bclr_df(env, df, ts, m);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_bneg_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return UNSIGNED(arg1 ^ (1LL << b_arg2), df);
+}
+
+void helper_msa_bneg_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_bneg_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_bnegi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t m)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_bneg_df(env, df, ts, m);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_bset_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return UNSIGNED(arg1 | (1LL << b_arg2), df);
+}
+
+void helper_msa_bset_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_bset_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_bseti_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t m)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_bset_df(env, df, ts, m);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_binsl_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_dest = UNSIGNED(dest, df);
+ int32_t sh_d = BIT_POSITION(arg2, df) + 1;
+ int32_t sh_a = DF_BITS(df) - sh_d;
+ if (sh_d == DF_BITS(df)) {
+ return u_arg1;
+ } else {
+ return UNSIGNED(UNSIGNED(u_dest << sh_d, df) >> sh_d, df) |
+ UNSIGNED(UNSIGNED(u_arg1 >> sh_a, df) << sh_a, df);
+ }
+}
+
+void helper_msa_binsl_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_binsl_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_binsli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t m)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_binsl_df(env, df, td, ts, m);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_binsr_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_dest = UNSIGNED(dest, df);
+ int32_t sh_d = BIT_POSITION(arg2, df) + 1;
+ int32_t sh_a = DF_BITS(df) - sh_d;
+ if (sh_d == DF_BITS(df)) {
+ return u_arg1;
+ } else {
+ return UNSIGNED(UNSIGNED(u_dest >> sh_d, df) << sh_d, df) |
+ UNSIGNED(UNSIGNED(u_arg1 << sh_a, df) >> sh_a, df);
+ }
+}
+
+void helper_msa_binsr_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_binsr_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_binsri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t m)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_binsr_df(env, df, td, ts, m);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+#define BIT_MOVE_IF_NOT_ZERO(dest, arg1, arg2, df) \
+ dest = UNSIGNED(((dest & (~arg2)) | (arg1 & arg2)), df)
+
+void helper_msa_bmnzi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+ uint32_t i8)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ BIT_MOVE_IF_NOT_ZERO(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ } DONE_ALL_ELEMENTS;
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+#define BIT_MOVE_IF_ZERO(dest, arg1, arg2, df) \
+ dest = UNSIGNED((dest & arg2) | (arg1 & (~arg2)), df)
+
+void helper_msa_bmzi_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+ uint32_t i8)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ BIT_MOVE_IF_ZERO(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ } DONE_ALL_ELEMENTS;
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+#define BIT_SELECT(dest, arg1, arg2, df) \
+ dest = UNSIGNED((arg1 & (~dest)) | (arg2 & dest), df)
+
+void helper_msa_bseli_b(CPUMIPSState *env, uint32_t wd, uint32_t ws,
+ uint32_t i8)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ BIT_SELECT(B(pwd, i), B(pws, i), i8, DF_BYTE);
+ } DONE_ALL_ELEMENTS;
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_ceq_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ return arg1 == arg2 ? -1 : 0;
+}
+
+void helper_msa_ceq_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_ceq_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_ceqi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t i5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_ceq_df(env, df, ts, i5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_cle_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return arg1 <= arg2 ? -1 : 0;
+}
+
+void helper_msa_cle_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_cle_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_clei_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t s5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_cle_s_df(env, df, ts, s5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_cle_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return u_arg1 <= u_arg2 ? -1 : 0;
+}
+
+void helper_msa_cle_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_cle_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_clei_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t u5)
+{
+ uint64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ td = msa_cle_u_df(env, df, ts, u5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_clt_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return arg1 < arg2 ? -1 : 0;
+}
+
+void helper_msa_clt_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_clt_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_clti_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t s5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_clt_s_df(env, df, ts, s5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_clt_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return u_arg1 < u_arg2 ? -1 : 0;
+}
+
+void helper_msa_clt_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_clt_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_clti_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t u5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_clt_u_df(env, df, ts, u5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+#define SIGNED_EVEN(a, df) \
+ ((((int64_t)(a)) << (64 - DF_BITS(df)/2)) >> (64 - DF_BITS(df)/2))
+#define UNSIGNED_EVEN(a, df) \
+ ((((uint64_t)(a)) << (64 - DF_BITS(df)/2)) >> (64 - DF_BITS(df)/2))
+
+#define SIGNED_ODD(a, df) \
+ ((((int64_t)(a)) << (64 - DF_BITS(df))) >> (64 - DF_BITS(df)/2))
+#define UNSIGNED_ODD(a, df) \
+ ((((uint64_t)(a)) << (64 - DF_BITS(df))) >> (64 - DF_BITS(df)/2))
+
+#define SIGNED_EXTRACT(e, o, a, df) \
+ int64_t e = SIGNED_EVEN(a, df); \
+ int64_t o = SIGNED_ODD(a, df);
+
+#define UNSIGNED_EXTRACT(e, o, a, df) \
+ int64_t e = UNSIGNED_EVEN(a, df); \
+ int64_t o = UNSIGNED_ODD(a, df);
+
+static inline int64_t msa_hadd_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return SIGNED_ODD(arg1, df) + SIGNED_EVEN(arg2, df);
+}
+
+void helper_msa_hadd_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_hadd_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_hadd_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return UNSIGNED_ODD(arg1, df) + UNSIGNED_EVEN(arg2, df);
+}
+
+void helper_msa_hadd_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_hadd_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_hsub_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return SIGNED_ODD(arg1, df) - SIGNED_EVEN(arg2, df);
+}
+
+void helper_msa_hsub_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_hsub_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_hsub_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return UNSIGNED_ODD(arg1, df) - UNSIGNED_EVEN(arg2, df);
+}
+
+void helper_msa_hsub_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_hsub_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dotp_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ SIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ SIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return (even_arg1 * even_arg2) + (odd_arg1 * odd_arg2);
+}
+
+void helper_msa_dotp_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_dotp_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dotp_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ UNSIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ UNSIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return (even_arg1 * even_arg2) + (odd_arg1 * odd_arg2);
+}
+
+void helper_msa_dotp_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_dotp_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dpadd_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ SIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ SIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return dest + (even_arg1 * even_arg2) + (odd_arg1 * odd_arg2);
+}
+
+void helper_msa_dpadd_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_dpadd_s_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dpadd_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ UNSIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ UNSIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return dest + (even_arg1 * even_arg2) + (odd_arg1 * odd_arg2);
+}
+
+void helper_msa_dpadd_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_dpadd_u_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dpsub_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ SIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ SIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return dest - ((even_arg1 * even_arg2) + (odd_arg1 * odd_arg2));
+}
+
+void helper_msa_dpsub_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_dpsub_s_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_dpsub_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ UNSIGNED_EXTRACT(even_arg1, odd_arg1, arg1, df);
+ UNSIGNED_EXTRACT(even_arg2, odd_arg2, arg2, df);
+ return dest - ((even_arg1 * even_arg2) + (odd_arg1 * odd_arg2));
+}
+
+void helper_msa_dpsub_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_dpsub_u_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_ilvev_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ B(pwx, 2*i) = B(pwt, 2*i);
+ B(pwx, 2*i+1) = B(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ H(pwx, 2*i) = H(pwt, 2*i);
+ H(pwx, 2*i+1) = H(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ W(pwx, 2*i) = W(pwt, 2*i);
+ W(pwx, 2*i+1) = W(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ D(pwx, 2*i) = D(pwt, 2*i);
+ D(pwx, 2*i+1) = D(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_ilvev_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_ilvev_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_ilvod_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ B(pwx, 2*i) = B(pwt, 2*i+1);
+ B(pwx, 2*i+1) = B(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ H(pwx, 2*i) = H(pwt, 2*i+1);
+ H(pwx, 2*i+1) = H(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ W(pwx, 2*i) = W(pwt, 2*i+1);
+ W(pwx, 2*i+1) = W(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ D(pwx, 2*i) = D(pwt, 2*i+1);
+ D(pwx, 2*i+1) = D(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_ilvod_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_ilvod_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_ilvl_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ B(pwx, 2*i) = BL(pwt, i);
+ B(pwx, 2*i+1) = BL(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ H(pwx, 2*i) = HL(pwt, i);
+ H(pwx, 2*i+1) = HL(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ W(pwx, 2*i) = WL(pwt, i);
+ W(pwx, 2*i+1) = WL(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ D(pwx, 2*i) = DL(pwt, i);
+ D(pwx, 2*i+1) = DL(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_ilvl_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_ilvl_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_ilvr_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ B(pwx, 2*i) = BR(pwt, i);
+ B(pwx, 2*i+1) = BR(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ H(pwx, 2*i) = HR(pwt, i);
+ H(pwx, 2*i+1) = HR(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ W(pwx, 2*i) = WR(pwt, i);
+ W(pwx, 2*i+1) = WR(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ D(pwx, 2*i) = DR(pwt, i);
+ D(pwx, 2*i+1) = DR(pws, i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_ilvr_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_ilvr_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_pckev_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ BR(pwx, i) = B(pwt, 2*i);
+ BL(pwx, i) = B(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ HR(pwx, i) = H(pwt, 2*i);
+ HL(pwx, i) = H(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ WR(pwx, i) = W(pwt, 2*i);
+ WL(pwx, i) = W(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ DR(pwx, i) = D(pwt, 2*i);
+ DL(pwx, i) = D(pws, 2*i);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_pckev_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_pckev_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_pckod_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ BR(pwx, i) = B(pwt, 2*i+1);
+ BL(pwx, i) = B(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ HR(pwx, i) = H(pwt, 2*i+1);
+ HL(pwx, i) = H(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ WR(pwx, i) = W(pwt, 2*i+1);
+ WL(pwx, i) = W(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_Q_ELEMENTS(i, MSA_WRLEN) {
+ DR(pwx, i) = D(pwt, 2*i+1);
+ DL(pwx, i) = D(pws, 2*i+1);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_pckod_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_pckod_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_vshf_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, void *pwt)
+{
+ uint32_t n = MSA_WRLEN / DF_BITS(df);
+ uint32_t k;
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ /* byte data format */
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ k = (B(pwd, i) & 0x3f) % (2 * n);
+ B(pwx, i) =
+ (B(pwd, i) & 0xc0) ? 0 : k < n ? B(pwt, k) : B(pws, k - n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ /* half data format */
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ k = (H(pwd, i) & 0x3f) % (2 * n);
+ H(pwx, i) =
+ (H(pwd, i) & 0xc0) ? 0 : k < n ? H(pwt, k) : H(pws, k - n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ /* word data format */
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ k = (W(pwd, i) & 0x3f) % (2 * n);
+ W(pwx, i) =
+ (W(pwd, i) & 0xc0) ? 0 : k < n ? W(pwt, k) : W(pws, k - n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ /* double data format */
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ k = (D(pwd, i) & 0x3f) % (2 * n);
+ D(pwx, i) =
+ (D(pwd, i) & 0xc0) ? 0 : k < n ? D(pwt, k) : D(pws, k - n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_vshf_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ void *pwt = &(env->active_fpu.fpr[wt]);
+ msa_vshf_df(env, df, pwd, pws, pwt);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+
+#define SHF_POS(i, imm) ((i & 0xfc) + ((imm >> (2 * (i & 0x03))) & 0x03))
+
+static inline void msa_shf_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, uint32_t imm)
+{
+ wr_t wx, *pwx = &wx;
+ switch (df) {
+ case DF_BYTE:
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ B(pwx, i) = B(pws, SHF_POS(i, imm));
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ H(pwx, i) = H(pws, SHF_POS(i, imm));
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ W(pwx, i) = W(pws, SHF_POS(i, imm));
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+ msa_move_v(pwd, pwx);
+}
+
+void helper_msa_shf_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t imm)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ msa_shf_df(env, df, pwd, pws, imm);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_maddv_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ return dest + arg1 * arg2;
+}
+
+void helper_msa_maddv_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_maddv_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_msubv_df(CPUMIPSState *env, uint32_t df,
+ int64_t dest, int64_t arg1, int64_t arg2)
+{
+ return dest - arg1 * arg2;
+}
+
+void helper_msa_msubv_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_load_wr_elem_s64(env, wd, df, i);
+ td = msa_msubv_df(env, df, td, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_max_a_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t abs_arg1 = arg1 >= 0 ? arg1 : -arg1;
+ uint64_t abs_arg2 = arg2 >= 0 ? arg2 : -arg2;
+ return abs_arg1 > abs_arg2 ? arg1 : arg2;
+}
+
+void helper_msa_max_a_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_max_a_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_max_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return arg1 > arg2 ? arg1 : arg2;
+}
+
+void helper_msa_max_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_max_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_maxi_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t s5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_max_s_df(env, df, ts, s5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_max_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return u_arg1 > u_arg2 ? arg1 : arg2;
+}
+
+void helper_msa_max_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_max_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_maxi_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t u5)
+{
+ uint64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ td = msa_max_u_df(env, df, ts, u5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_min_a_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t abs_arg1 = arg1 >= 0 ? arg1 : -arg1;
+ uint64_t abs_arg2 = arg2 >= 0 ? arg2 : -arg2;
+ return abs_arg1 < abs_arg2 ? arg1 : arg2;
+}
+
+void helper_msa_min_a_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_min_a_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_min_s_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ return arg1 < arg2 ? arg1 : arg2;
+}
+
+void helper_msa_min_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_min_s_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_mini_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t s5)
+{
+ int64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ td = msa_min_s_df(env, df, ts, s5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline int64_t msa_min_u_df(CPUMIPSState *env, uint32_t df,
+ int64_t arg1, int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ uint64_t u_arg2 = UNSIGNED(arg2, df);
+ return u_arg1 < u_arg2 ? arg1 : arg2;
+}
+
+void helper_msa_min_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ uint64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_min_u_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+void helper_msa_mini_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, int64_t u5)
+{
+ uint64_t td, ts;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_i64(env, ws, df, i);
+ td = msa_min_u_df(env, df, ts, u5);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
+static inline void msa_splat_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, target_ulong rt)
+{
+ uint32_t n = rt % DF_ELEMENTS(df, MSA_WRLEN);
+ msa_check_index(env, df, n);
+ switch (df) {
+ case DF_BYTE:
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ B(pwd, i) = B(pws, n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_HALF:
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ H(pwd, i) = H(pws, n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_WORD:
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ W(pwd, i) = W(pws, n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ D(pwd, i) = D(pws, n);
+ } DONE_ALL_ELEMENTS;
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+}
+
+void helper_msa_splat_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t rt)
{
- int64_t td, ts;
- int i;
- int df_bits = 8 * (1 << df);
- for (i = 0; i < MSA_WRLEN / df_bits; i++) {
- ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_clt_u_df(env, df, ts, u5);
- msa_store_wr_elem(env, td, wd, df, i);
- }
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ msa_splat_df(env, df, pwd, pws, env->active_tc.gpr[rt]);
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-#define SHF_POS(i, imm) ((i & 0xfc) + ((imm >> (2 * (i & 0x03))) & 0x03))
-
-static inline void msa_shf_df(CPUMIPSState *env, uint32_t df, void *pwd,
- void *pws, uint32_t imm)
+void helper_msa_ldi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t s10)
{
- wr_t wx, *pwx = &wx;
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ int64_t s64 = ((int64_t)s10 << 54) >> 54;
switch (df) {
case DF_BYTE:
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwx, i) = B(pws, SHF_POS(i, imm));
- } DONE_ALL_ELEMENTS;
- break;
+ ALL_B_ELEMENTS(i, MSA_WRLEN) {
+ B(pwd, i) = (int8_t)s10;
+ } DONE_ALL_ELEMENTS;
+ break;
case DF_HALF:
- ALL_H_ELEMENTS(i, MSA_WRLEN) {
- H(pwx, i) = H(pws, SHF_POS(i, imm));
- } DONE_ALL_ELEMENTS;
- break;
+ ALL_H_ELEMENTS(i, MSA_WRLEN) {
+ H(pwd, i) = (int16_t)s64;
+ } DONE_ALL_ELEMENTS;
+ break;
case DF_WORD:
- ALL_W_ELEMENTS(i, MSA_WRLEN) {
- W(pwx, i) = W(pws, SHF_POS(i, imm));
- } DONE_ALL_ELEMENTS;
- break;
+ ALL_W_ELEMENTS(i, MSA_WRLEN) {
+ W(pwd, i) = (int32_t)s64;
+ } DONE_ALL_ELEMENTS;
+ break;
+ case DF_DOUBLE:
+ ALL_D_ELEMENTS(i, MSA_WRLEN) {
+ D(pwd, i) = s64;
+ } DONE_ALL_ELEMENTS;
+ break;
default:
/* shouldn't get here */
assert(0);
}
- msa_move_v(pwd, pwx);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
}
-void helper_msa_shf_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, uint32_t imm)
+void helper_msa_mulv_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- void *pwd = &(env->active_fpu.fpr[wd]);
- void *pws = &(env->active_fpu.fpr[ws]);
- msa_shf_df(env, df, pwd, pws, imm);
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = ts * tt;
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
if (env->active_msa.msair & MSAIR_WRP_BIT) {
env->active_msa.msamodify |= (1 << wd);
}
}
-static inline int64_t msa_max_s_df(CPUMIPSState *env, uint32_t df,
+static inline int64_t msa_div_s_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
- return arg1 > arg2 ? arg1 : arg2;
+ if (arg1 == DF_MIN_INT(df) && arg2 == -1) {
+ return DF_MIN_INT(df);
+ }
+ return arg2 ? arg1 / arg2 : 0;
}
-void helper_msa_maxi_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t s5)
+void helper_msa_div_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_max_s_df(env, df, ts, s5);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_div_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -637,23 +2220,24 @@ void helper_msa_maxi_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_max_u_df(CPUMIPSState *env, uint32_t df,
+static inline int64_t msa_div_u_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
uint64_t u_arg1 = UNSIGNED(arg1, df);
uint64_t u_arg2 = UNSIGNED(arg2, df);
- return u_arg1 > u_arg2 ? arg1 : arg2;
+ return u_arg2 ? u_arg1 / u_arg2 : 0;
}
-void helper_msa_maxi_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t u5)
+void helper_msa_div_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- uint64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_i64(env, ws, df, i);
- td = msa_max_u_df(env, df, ts, u5);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_div_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -661,21 +2245,25 @@ void helper_msa_maxi_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_min_s_df(CPUMIPSState *env, uint32_t df,
+static inline int64_t msa_mod_s_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
- return arg1 < arg2 ? arg1 : arg2;
+ if (arg1 == DF_MIN_INT(df) && arg2 == -1) {
+ return 0;
+ }
+ return arg2 ? arg1 % arg2 : 0;
}
-void helper_msa_mini_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t s5)
+void helper_msa_mod_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- int64_t td, ts;
+ int64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_s64(env, ws, df, i);
- td = msa_min_s_df(env, df, ts, s5);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_mod_s_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -683,23 +2271,24 @@ void helper_msa_mini_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-static inline int64_t msa_min_u_df(CPUMIPSState *env, uint32_t df,
+static inline int64_t msa_mod_u_df(CPUMIPSState *env, uint32_t df,
int64_t arg1, int64_t arg2)
{
uint64_t u_arg1 = UNSIGNED(arg1, df);
uint64_t u_arg2 = UNSIGNED(arg2, df);
- return u_arg1 < u_arg2 ? arg1 : arg2;
+ return u_arg2 ? u_arg1 % u_arg2 : 0;
}
-void helper_msa_mini_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t ws, int64_t u5)
+void helper_msa_mod_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
{
- uint64_t td, ts;
+ uint64_t td, ts, tt;
int i;
int df_bits = 8 * (1 << df);
for (i = 0; i < MSA_WRLEN / df_bits; i++) {
ts = msa_load_wr_elem_i64(env, ws, df, i);
- td = msa_min_u_df(env, df, ts, u5);
+ tt = msa_load_wr_elem_i64(env, wt, df, i);
+ td = msa_mod_u_df(env, df, ts, tt);
msa_store_wr_elem(env, td, wd, df, i);
}
if (env->active_msa.msair & MSAIR_WRP_BIT) {
@@ -707,41 +2296,6 @@ void helper_msa_mini_u_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
-void helper_msa_ldi_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
- uint32_t s10)
-{
- void *pwd = &(env->active_fpu.fpr[wd]);
- int64_t s64 = ((int64_t)s10 << 54) >> 54;
- switch (df) {
- case DF_BYTE:
- ALL_B_ELEMENTS(i, MSA_WRLEN) {
- B(pwd, i) = (int8_t)s10;
- } DONE_ALL_ELEMENTS;
- break;
- case DF_HALF:
- ALL_H_ELEMENTS(i, MSA_WRLEN) {
- H(pwd, i) = (int16_t)s64;
- } DONE_ALL_ELEMENTS;
- break;
- case DF_WORD:
- ALL_W_ELEMENTS(i, MSA_WRLEN) {
- W(pwd, i) = (int32_t)s64;
- } DONE_ALL_ELEMENTS;
- break;
- case DF_DOUBLE:
- ALL_D_ELEMENTS(i, MSA_WRLEN) {
- D(pwd, i) = s64;
- } DONE_ALL_ELEMENTS;
- break;
- default:
- /* shouldn't get here */
- assert(0);
- }
- if (env->active_msa.msair & MSAIR_WRP_BIT) {
- env->active_msa.msamodify |= (1 << wd);
- }
-}
-
static inline int64_t msa_sat_u_df(CPUMIPSState *env, uint32_t df, int64_t arg,
uint32_t m)
{
@@ -790,6 +2344,30 @@ void helper_msa_sat_s_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+static inline int64_t msa_sll_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return arg1 << b_arg2;
+}
+
+void helper_msa_sll_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_sll_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
void helper_msa_slli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
uint32_t ws, uint32_t m)
{
@@ -806,6 +2384,30 @@ void helper_msa_slli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+static inline int64_t msa_sra_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return arg1 >> b_arg2;
+}
+
+void helper_msa_sra_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_sra_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
void helper_msa_srai_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
uint32_t ws, uint32_t m)
{
@@ -822,6 +2424,31 @@ void helper_msa_srai_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+static inline int64_t msa_srl_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ return u_arg1 >> b_arg2;
+}
+
+void helper_msa_srl_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_srl_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
static inline int64_t msa_srli_df(CPUMIPSState *env, uint32_t df, int64_t arg,
uint32_t m)
{
@@ -845,6 +2472,35 @@ void helper_msa_srli_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+static inline int64_t msa_srar_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ if (b_arg2 == 0) {
+ return arg1;
+ } else {
+ int64_t r_bit = (arg1 >> (b_arg2 - 1)) & 1;
+ return (arg1 >> b_arg2) + r_bit;
+ }
+}
+
+void helper_msa_srar_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_srar_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
static inline int64_t msa_srari_df(CPUMIPSState *env, uint32_t df, int64_t arg,
uint32_t m)
{
@@ -872,6 +2528,36 @@ void helper_msa_srari_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
}
}
+static inline int64_t msa_srlr_df(CPUMIPSState *env, uint32_t df, int64_t arg1,
+ int64_t arg2)
+{
+ uint64_t u_arg1 = UNSIGNED(arg1, df);
+ int32_t b_arg2 = BIT_POSITION(arg2, df);
+ if (b_arg2 == 0) {
+ return u_arg1;
+ } else {
+ uint64_t r_bit = (u_arg1 >> (b_arg2 - 1)) & 1;
+ return (u_arg1 >> b_arg2) + r_bit;
+ }
+}
+
+void helper_msa_srlr_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t wt)
+{
+ int64_t td, ts, tt;
+ int i;
+ int df_bits = 8 * (1 << df);
+ for (i = 0; i < MSA_WRLEN / df_bits; i++) {
+ ts = msa_load_wr_elem_s64(env, ws, df, i);
+ tt = msa_load_wr_elem_s64(env, wt, df, i);
+ td = msa_srlr_df(env, df, ts, tt);
+ msa_store_wr_elem(env, td, wd, df, i);
+ }
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
+
static inline int64_t msa_srlri_df(CPUMIPSState *env, uint32_t df, int64_t arg,
uint32_t m)
{
@@ -899,3 +2585,56 @@ void helper_msa_srlri_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
env->active_msa.msamodify |= (1 << wd);
}
}
+static inline void msa_sld_df(CPUMIPSState *env, uint32_t df, void *pwd,
+ void *pws, target_ulong rt)
+{
+ uint32_t n = rt % DF_ELEMENTS(df, MSA_WRLEN);
+ uint8_t v[64];
+ uint32_t i, k;
+#define CONCATENATE_AND_SLIDE(s, k) \
+ do { \
+ for (i = 0; i < s; i++) { \
+ v[i] = B(pws, s * k + i); \
+ v[i + s] = B(pwd, s * k + i); \
+ } \
+ for (i = 0; i < s; i++) { \
+ B(pwd, s * k + i) = v[i + n]; \
+ } \
+ } while (0)
+
+ msa_check_index(env, df, n);
+ switch (df) {
+ case DF_BYTE:
+ CONCATENATE_AND_SLIDE(MSA_WRLEN/8, 0);
+ break;
+ case DF_HALF:
+ for (k = 0; k < 2; k++) {
+ CONCATENATE_AND_SLIDE(MSA_WRLEN/16, k);
+ }
+ break;
+ case DF_WORD:
+ for (k = 0; k < 4; k++) {
+ CONCATENATE_AND_SLIDE(MSA_WRLEN/32, k);
+ }
+ break;
+ case DF_DOUBLE:
+ for (k = 0; k < 8; k++) {
+ CONCATENATE_AND_SLIDE(MSA_WRLEN/64, k);
+ }
+ break;
+ default:
+ /* shouldn't get here */
+ assert(0);
+ }
+}
+
+void helper_msa_sld_df(CPUMIPSState *env, uint32_t df, uint32_t wd,
+ uint32_t ws, uint32_t rt)
+{
+ void *pwd = &(env->active_fpu.fpr[wd]);
+ void *pws = &(env->active_fpu.fpr[ws]);
+ msa_sld_df(env, df, pwd, pws, env->active_tc.gpr[rt]);
+ if (env->active_msa.msair & MSAIR_WRP_BIT) {
+ env->active_msa.msamodify |= (1 << wd);
+ }
+}
@@ -15030,6 +15030,295 @@ static void gen_msa_bit(CPUMIPSState *env, DisasContext *ctx)
tcg_temp_free_i32(tws);
}
+static void gen_msa_3r(CPUMIPSState *env, DisasContext *ctx)
+{
+#define MASK_MSA_3R(op) (MASK_MSA_MINOR(op) | (op & (0x7 << 23)))
+ uint32_t opcode = ctx->opcode;
+
+ uint8_t df = (ctx->opcode >> 21) & 0x3 /* df [22:21] */;
+ uint8_t wt = (ctx->opcode >> 16) & 0x1f /* wt [20:16] */;
+ uint8_t ws = (ctx->opcode >> 11) & 0x1f /* ws [15:11] */;
+ uint8_t wd = (ctx->opcode >> 6) & 0x1f /* wd [10:6] */;
+
+ TCGv_i32 tdf = tcg_const_i32(df);
+ TCGv_i32 twd = tcg_const_i32(wd);
+ TCGv_i32 tws = tcg_const_i32(ws);
+ TCGv_i32 twt = tcg_const_i32(wt);
+
+ switch (MASK_MSA_3R(opcode)) {
+ case OPC_MSA_SLL_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_sll_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ADDV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_addv_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_CEQ_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ceq_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ADD_A_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_add_a_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SUBS_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_subs_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MULV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_mulv_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SLD_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_sld_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_VSHF_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_vshf_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SRA_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_sra_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SUBV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_subv_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ADDS_A_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_adds_a_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SUBS_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_subs_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MADDV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_maddv_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SPLAT_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_splat_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SRAR_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_srar_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SRL_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_srl_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MAX_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_max_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_CLT_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_clt_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ADDS_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_adds_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SUBSUS_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_subsus_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MSUBV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_msubv_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_PCKEV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_pckev_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SRLR_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_srlr_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_BCLR_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_bclr_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MAX_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_max_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_CLT_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_clt_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ADDS_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_adds_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_SUBSUU_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_subsuu_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_PCKOD_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_pckod_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_BSET_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_bset_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MIN_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_min_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_CLE_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_cle_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_AVE_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ave_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ASUB_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_asub_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DIV_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_div_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ILVL_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ilvl_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_BNEG_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_bneg_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MIN_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_min_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_CLE_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_cle_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_AVE_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ave_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ASUB_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_asub_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DIV_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_div_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ILVR_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ilvr_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_BINSL_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_binsl_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MAX_A_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_max_a_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_AVER_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_aver_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MOD_S_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_mod_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ILVEV_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ilvev_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_BINSR_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_binsr_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MIN_A_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_min_a_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_AVER_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_aver_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_MOD_U_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_mod_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_ILVOD_df:
+ check_msa_access(env, ctx, wt, ws, wd);
+ gen_helper_msa_ilvod_df(cpu_env, tdf, twd, tws, twt);
+ break;
+
+ case OPC_MSA_DOTP_S_df:
+ case OPC_MSA_DOTP_U_df:
+ case OPC_MSA_DPADD_S_df:
+ case OPC_MSA_DPADD_U_df:
+ case OPC_MSA_DPSUB_S_df:
+ case OPC_MSA_HADD_S_df:
+ case OPC_MSA_DPSUB_U_df:
+ case OPC_MSA_HADD_U_df:
+ case OPC_MSA_HSUB_S_df:
+ case OPC_MSA_HSUB_U_df:
+ if (df == 0) {
+ if (check_msa_access(env, ctx, -1, -1, -1)) {
+ generate_exception(ctx, EXCP_RI);
+ }
+ }
+ check_msa_access(env, ctx, wt, ws, wd);
+ switch (MASK_MSA_3R(opcode)) {
+ case OPC_MSA_DOTP_S_df:
+ gen_helper_msa_dotp_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DOTP_U_df:
+ gen_helper_msa_dotp_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DPADD_S_df:
+ gen_helper_msa_dpadd_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DPADD_U_df:
+ gen_helper_msa_dpadd_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DPSUB_S_df:
+ gen_helper_msa_dpsub_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_HADD_S_df:
+ gen_helper_msa_hadd_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_DPSUB_U_df:
+ gen_helper_msa_dpsub_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_HADD_U_df:
+ gen_helper_msa_hadd_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_HSUB_S_df:
+ gen_helper_msa_hsub_s_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ case OPC_MSA_HSUB_U_df:
+ gen_helper_msa_hsub_u_df(cpu_env, tdf, twd, tws, twt);
+ break;
+ }
+ break;
+ default:
+ MIPS_INVAL("MSA instruction");
+ generate_exception(ctx, EXCP_RI);
+ break;
+ }
+ tcg_temp_free_i32(twd);
+ tcg_temp_free_i32(tws);
+ tcg_temp_free_i32(twt);
+ tcg_temp_free_i32(tdf);
+}
+
static void gen_msa(CPUMIPSState *env, DisasContext *ctx)
{
uint32_t opcode = ctx->opcode;
@@ -15049,6 +15338,17 @@ static void gen_msa(CPUMIPSState *env, DisasContext *ctx)
case OPC_MSA_BIT_0A:
gen_msa_bit(env, ctx);
break;
+ case OPC_MSA_3R_0D:
+ case OPC_MSA_3R_0E:
+ case OPC_MSA_3R_0F:
+ case OPC_MSA_3R_10:
+ case OPC_MSA_3R_11:
+ case OPC_MSA_3R_12:
+ case OPC_MSA_3R_13:
+ case OPC_MSA_3R_14:
+ case OPC_MSA_3R_15:
+ gen_msa_3r(env, ctx);
+ break;
default:
MIPS_INVAL("MSA instruction");
generate_exception(ctx, EXCP_RI);
add MSA 3R format instructions Signed-off-by: Yongbok Kim <yongbok.kim@imgtec.com> --- target-mips/helper.h | 63 ++ target-mips/msa_helper.c | 2193 +++++++++++++++++++++++++++++++++++++++++----- target-mips/translate.c | 300 +++++++ 3 files changed, 2329 insertions(+), 227 deletions(-)