X-Git-Url: https://repo.jachan.dev/qemu.git/blobdiff_plain/29f3ff8d6cbc28f79933aeaa25805408d0984a8f..2e4a0b176642c5ffcca063880ad2aa8e44c78c8a:/tcg/optimize.c diff --git a/tcg/optimize.c b/tcg/optimize.c index cd0e793a9d..5dbe11c3c8 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -23,169 +23,139 @@ * THE SOFTWARE. */ -#include "config.h" - -#include -#include - +#include "qemu/osdep.h" #include "qemu-common.h" +#include "exec/cpu-common.h" #include "tcg-op.h" #define CASE_OP_32_64(x) \ glue(glue(case INDEX_op_, x), _i32): \ glue(glue(case INDEX_op_, x), _i64) -typedef enum { - TCG_TEMP_UNDEF = 0, - TCG_TEMP_CONST, - TCG_TEMP_COPY, -} tcg_temp_state; +#define CASE_OP_32_64_VEC(x) \ + glue(glue(case INDEX_op_, x), _i32): \ + glue(glue(case INDEX_op_, x), _i64): \ + glue(glue(case INDEX_op_, x), _vec) struct tcg_temp_info { - tcg_temp_state state; - uint16_t prev_copy; - uint16_t next_copy; + bool is_const; + TCGTemp *prev_copy; + TCGTemp *next_copy; tcg_target_ulong val; tcg_target_ulong mask; }; -static struct tcg_temp_info temps[TCG_MAX_TEMPS]; +static inline struct tcg_temp_info *ts_info(TCGTemp *ts) +{ + return ts->state_ptr; +} -/* Reset TEMP's state to TCG_TEMP_UNDEF. If TEMP only had one copy, remove - the copy flag from the left temp. */ -static void reset_temp(TCGArg temp) +static inline struct tcg_temp_info *arg_info(TCGArg arg) { - if (temps[temp].state == TCG_TEMP_COPY) { - if (temps[temp].prev_copy == temps[temp].next_copy) { - temps[temps[temp].next_copy].state = TCG_TEMP_UNDEF; - } else { - temps[temps[temp].next_copy].prev_copy = temps[temp].prev_copy; - temps[temps[temp].prev_copy].next_copy = temps[temp].next_copy; - } - } - temps[temp].state = TCG_TEMP_UNDEF; - temps[temp].mask = -1; + return ts_info(arg_temp(arg)); } -static TCGOp *insert_op_before(TCGContext *s, TCGOp *old_op, - TCGOpcode opc, int nargs) +static inline bool ts_is_const(TCGTemp *ts) { - int oi = s->gen_next_op_idx; - int pi = s->gen_next_parm_idx; - int prev = old_op->prev; - int next = old_op - s->gen_op_buf; - TCGOp *new_op; - - tcg_debug_assert(oi < OPC_BUF_SIZE); - tcg_debug_assert(pi + nargs <= OPPARAM_BUF_SIZE); - s->gen_next_op_idx = oi + 1; - s->gen_next_parm_idx = pi + nargs; - - new_op = &s->gen_op_buf[oi]; - *new_op = (TCGOp){ - .opc = opc, - .args = pi, - .prev = prev, - .next = next - }; - if (prev >= 0) { - s->gen_op_buf[prev].next = oi; - } else { - s->gen_first_op_idx = oi; - } - old_op->prev = oi; + return ts_info(ts)->is_const; +} - return new_op; +static inline bool arg_is_const(TCGArg arg) +{ + return ts_is_const(arg_temp(arg)); } -/* Reset all temporaries, given that there are NB_TEMPS of them. */ -static void reset_all_temps(int nb_temps) +static inline bool ts_is_copy(TCGTemp *ts) { - int i; - for (i = 0; i < nb_temps; i++) { - temps[i].state = TCG_TEMP_UNDEF; - temps[i].mask = -1; - } + return ts_info(ts)->next_copy != ts; } -static int op_bits(TCGOpcode op) +/* Reset TEMP's state, possibly removing the temp for the list of copies. */ +static void reset_ts(TCGTemp *ts) { - const TCGOpDef *def = &tcg_op_defs[op]; - return def->flags & TCG_OPF_64BIT ? 64 : 32; + struct tcg_temp_info *ti = ts_info(ts); + struct tcg_temp_info *pi = ts_info(ti->prev_copy); + struct tcg_temp_info *ni = ts_info(ti->next_copy); + + ni->prev_copy = ti->prev_copy; + pi->next_copy = ti->next_copy; + ti->next_copy = ts; + ti->prev_copy = ts; + ti->is_const = false; + ti->mask = -1; } -static TCGOpcode op_to_mov(TCGOpcode op) +static void reset_temp(TCGArg arg) { - switch (op_bits(op)) { - case 32: - return INDEX_op_mov_i32; - case 64: - return INDEX_op_mov_i64; - default: - fprintf(stderr, "op_to_mov: unexpected return value of " - "function op_bits.\n"); - tcg_abort(); - } + reset_ts(arg_temp(arg)); } -static TCGOpcode op_to_movi(TCGOpcode op) +/* Initialize and activate a temporary. */ +static void init_ts_info(struct tcg_temp_info *infos, + TCGTempSet *temps_used, TCGTemp *ts) { - switch (op_bits(op)) { - case 32: - return INDEX_op_movi_i32; - case 64: - return INDEX_op_movi_i64; - default: - fprintf(stderr, "op_to_movi: unexpected return value of " - "function op_bits.\n"); - tcg_abort(); + size_t idx = temp_idx(ts); + if (!test_bit(idx, temps_used->l)) { + struct tcg_temp_info *ti = &infos[idx]; + + ts->state_ptr = ti; + ti->next_copy = ts; + ti->prev_copy = ts; + ti->is_const = false; + ti->mask = -1; + set_bit(idx, temps_used->l); } } -static TCGArg find_better_copy(TCGContext *s, TCGArg temp) +static void init_arg_info(struct tcg_temp_info *infos, + TCGTempSet *temps_used, TCGArg arg) +{ + init_ts_info(infos, temps_used, arg_temp(arg)); +} + +static TCGTemp *find_better_copy(TCGContext *s, TCGTemp *ts) { - TCGArg i; + TCGTemp *i; /* If this is already a global, we can't do better. */ - if (temp < s->nb_globals) { - return temp; + if (ts->temp_global) { + return ts; } /* Search for a global first. */ - for (i = temps[temp].next_copy ; i != temp ; i = temps[i].next_copy) { - if (i < s->nb_globals) { + for (i = ts_info(ts)->next_copy; i != ts; i = ts_info(i)->next_copy) { + if (i->temp_global) { return i; } } /* If it is a temp, search for a temp local. */ - if (!s->temps[temp].temp_local) { - for (i = temps[temp].next_copy ; i != temp ; i = temps[i].next_copy) { - if (s->temps[i].temp_local) { + if (!ts->temp_local) { + for (i = ts_info(ts)->next_copy; i != ts; i = ts_info(i)->next_copy) { + if (ts->temp_local) { return i; } } } /* Failure to find a better representation, return the same temp. */ - return temp; + return ts; } -static bool temps_are_copies(TCGArg arg1, TCGArg arg2) +static bool ts_are_copies(TCGTemp *ts1, TCGTemp *ts2) { - TCGArg i; + TCGTemp *i; - if (arg1 == arg2) { + if (ts1 == ts2) { return true; } - if (temps[arg1].state != TCG_TEMP_COPY - || temps[arg2].state != TCG_TEMP_COPY) { + if (!ts_is_copy(ts1) || !ts_is_copy(ts2)) { return false; } - for (i = temps[arg1].next_copy ; i != arg1 ; i = temps[i].next_copy) { - if (i == arg2) { + for (i = ts_info(ts1)->next_copy; i != ts1; i = ts_info(i)->next_copy) { + if (i == ts2) { return true; } } @@ -193,71 +163,90 @@ static bool temps_are_copies(TCGArg arg1, TCGArg arg2) return false; } -static void tcg_opt_gen_movi(TCGContext *s, TCGOp *op, TCGArg *args, - TCGArg dst, TCGArg val) +static bool args_are_copies(TCGArg arg1, TCGArg arg2) { - TCGOpcode new_op = op_to_movi(op->opc); + return ts_are_copies(arg_temp(arg1), arg_temp(arg2)); +} + +static void tcg_opt_gen_movi(TCGContext *s, TCGOp *op, TCGArg dst, TCGArg val) +{ + const TCGOpDef *def; + TCGOpcode new_op; tcg_target_ulong mask; + struct tcg_temp_info *di = arg_info(dst); + def = &tcg_op_defs[op->opc]; + if (def->flags & TCG_OPF_VECTOR) { + new_op = INDEX_op_dupi_vec; + } else if (def->flags & TCG_OPF_64BIT) { + new_op = INDEX_op_movi_i64; + } else { + new_op = INDEX_op_movi_i32; + } op->opc = new_op; + /* TCGOP_VECL and TCGOP_VECE remain unchanged. */ + op->args[0] = dst; + op->args[1] = val; reset_temp(dst); - temps[dst].state = TCG_TEMP_CONST; - temps[dst].val = val; + di->is_const = true; + di->val = val; mask = val; if (TCG_TARGET_REG_BITS > 32 && new_op == INDEX_op_movi_i32) { /* High bits of the destination are now garbage. */ mask |= ~0xffffffffull; } - temps[dst].mask = mask; - - args[0] = dst; - args[1] = val; + di->mask = mask; } -static void tcg_opt_gen_mov(TCGContext *s, TCGOp *op, TCGArg *args, - TCGArg dst, TCGArg src) +static void tcg_opt_gen_mov(TCGContext *s, TCGOp *op, TCGArg dst, TCGArg src) { - if (temps_are_copies(dst, src)) { + TCGTemp *dst_ts = arg_temp(dst); + TCGTemp *src_ts = arg_temp(src); + const TCGOpDef *def; + struct tcg_temp_info *di; + struct tcg_temp_info *si; + tcg_target_ulong mask; + TCGOpcode new_op; + + if (ts_are_copies(dst_ts, src_ts)) { tcg_op_remove(s, op); return; } - if (temps[src].state == TCG_TEMP_CONST) { - tcg_opt_gen_movi(s, op, args, dst, temps[src].val); - return; + reset_ts(dst_ts); + di = ts_info(dst_ts); + si = ts_info(src_ts); + def = &tcg_op_defs[op->opc]; + if (def->flags & TCG_OPF_VECTOR) { + new_op = INDEX_op_mov_vec; + } else if (def->flags & TCG_OPF_64BIT) { + new_op = INDEX_op_mov_i64; + } else { + new_op = INDEX_op_mov_i32; } - - TCGOpcode new_op = op_to_mov(op->opc); - tcg_target_ulong mask; - op->opc = new_op; + /* TCGOP_VECL and TCGOP_VECE remain unchanged. */ + op->args[0] = dst; + op->args[1] = src; - reset_temp(dst); - mask = temps[src].mask; + mask = si->mask; if (TCG_TARGET_REG_BITS > 32 && new_op == INDEX_op_mov_i32) { /* High bits of the destination are now garbage. */ mask |= ~0xffffffffull; } - temps[dst].mask = mask; + di->mask = mask; - assert(temps[src].state != TCG_TEMP_CONST); + if (src_ts->type == dst_ts->type) { + struct tcg_temp_info *ni = ts_info(si->next_copy); - if (s->temps[src].type == s->temps[dst].type) { - if (temps[src].state != TCG_TEMP_COPY) { - temps[src].state = TCG_TEMP_COPY; - temps[src].next_copy = src; - temps[src].prev_copy = src; - } - temps[dst].state = TCG_TEMP_COPY; - temps[dst].next_copy = temps[src].next_copy; - temps[dst].prev_copy = src; - temps[temps[dst].next_copy].prev_copy = dst; - temps[src].next_copy = dst; + di->next_copy = si->next_copy; + di->prev_copy = src_ts; + ni->prev_copy = dst_ts; + si->next_copy = dst_ts; + di->is_const = si->is_const; + di->val = si->val; } - - args[0] = dst; - args[1] = src; } static TCGArg do_constant_folding_2(TCGOpcode op, TCGArg x, TCGArg y) @@ -292,7 +281,6 @@ static TCGArg do_constant_folding_2(TCGOpcode op, TCGArg x, TCGArg y) case INDEX_op_shr_i32: return (uint32_t)x >> (y & 31); - case INDEX_op_trunc_shr_i32: case INDEX_op_shr_i64: return (uint64_t)x >> (y & 63); @@ -335,6 +323,24 @@ static TCGArg do_constant_folding_2(TCGOpcode op, TCGArg x, TCGArg y) CASE_OP_32_64(nor): return ~(x | y); + case INDEX_op_clz_i32: + return (uint32_t)x ? clz32(x) : y; + + case INDEX_op_clz_i64: + return x ? clz64(x) : y; + + case INDEX_op_ctz_i32: + return (uint32_t)x ? ctz32(x) : y; + + case INDEX_op_ctz_i64: + return x ? ctz64(x) : y; + + case INDEX_op_ctpop_i32: + return ctpop32(x); + + case INDEX_op_ctpop_i64: + return ctpop64(x); + CASE_OP_32_64(ext8s): return (int8_t)x; @@ -347,12 +353,18 @@ static TCGArg do_constant_folding_2(TCGOpcode op, TCGArg x, TCGArg y) CASE_OP_32_64(ext16u): return (uint16_t)x; + case INDEX_op_ext_i32_i64: case INDEX_op_ext32s_i64: return (int32_t)x; + case INDEX_op_extu_i32_i64: + case INDEX_op_extrl_i64_i32: case INDEX_op_ext32u_i64: return (uint32_t)x; + case INDEX_op_extrh_i64_i32: + return (uint64_t)x >> 32; + case INDEX_op_muluh_i32: return ((uint64_t)(uint32_t)x * (uint32_t)y) >> 32; case INDEX_op_mulsh_i32: @@ -393,8 +405,9 @@ static TCGArg do_constant_folding_2(TCGOpcode op, TCGArg x, TCGArg y) static TCGArg do_constant_folding(TCGOpcode op, TCGArg x, TCGArg y) { + const TCGOpDef *def = &tcg_op_defs[op]; TCGArg res = do_constant_folding_2(op, x, y); - if (op_bits(op) == 32) { + if (!(def->flags & TCG_OPF_64BIT)) { res = (int32_t)res; } return res; @@ -481,18 +494,19 @@ static bool do_constant_folding_cond_eq(TCGCond c) static TCGArg do_constant_folding_cond(TCGOpcode op, TCGArg x, TCGArg y, TCGCond c) { - if (temps[x].state == TCG_TEMP_CONST && temps[y].state == TCG_TEMP_CONST) { - switch (op_bits(op)) { - case 32: - return do_constant_folding_cond_32(temps[x].val, temps[y].val, c); - case 64: - return do_constant_folding_cond_64(temps[x].val, temps[y].val, c); - default: - tcg_abort(); + tcg_target_ulong xv = arg_info(x)->val; + tcg_target_ulong yv = arg_info(y)->val; + if (arg_is_const(x) && arg_is_const(y)) { + const TCGOpDef *def = &tcg_op_defs[op]; + tcg_debug_assert(!(def->flags & TCG_OPF_VECTOR)); + if (def->flags & TCG_OPF_64BIT) { + return do_constant_folding_cond_64(xv, yv, c); + } else { + return do_constant_folding_cond_32(xv, yv, c); } - } else if (temps_are_copies(x, y)) { + } else if (args_are_copies(x, y)) { return do_constant_folding_cond_eq(c); - } else if (temps[y].state == TCG_TEMP_CONST && temps[y].val == 0) { + } else if (arg_is_const(y) && yv == 0) { switch (c) { case TCG_COND_LTU: return 0; @@ -501,9 +515,8 @@ static TCGArg do_constant_folding_cond(TCGOpcode op, TCGArg x, default: return 2; } - } else { - return 2; } + return 2; } /* Return 2 if the condition can't be simplified, and the result @@ -513,14 +526,15 @@ static TCGArg do_constant_folding_cond2(TCGArg *p1, TCGArg *p2, TCGCond c) TCGArg al = p1[0], ah = p1[1]; TCGArg bl = p2[0], bh = p2[1]; - if (temps[bl].state == TCG_TEMP_CONST - && temps[bh].state == TCG_TEMP_CONST) { - uint64_t b = ((uint64_t)temps[bh].val << 32) | (uint32_t)temps[bl].val; + if (arg_is_const(bl) && arg_is_const(bh)) { + tcg_target_ulong blv = arg_info(bl)->val; + tcg_target_ulong bhv = arg_info(bh)->val; + uint64_t b = deposit64(blv, 32, 32, bhv); - if (temps[al].state == TCG_TEMP_CONST - && temps[ah].state == TCG_TEMP_CONST) { - uint64_t a; - a = ((uint64_t)temps[ah].val << 32) | (uint32_t)temps[al].val; + if (arg_is_const(al) && arg_is_const(ah)) { + tcg_target_ulong alv = arg_info(al)->val; + tcg_target_ulong ahv = arg_info(ah)->val; + uint64_t a = deposit64(alv, 32, 32, ahv); return do_constant_folding_cond_64(a, b, c); } if (b == 0) { @@ -534,7 +548,7 @@ static TCGArg do_constant_folding_cond2(TCGArg *p1, TCGArg *p2, TCGCond c) } } } - if (temps_are_copies(al, bl) && temps_are_copies(ah, bh)) { + if (args_are_copies(al, bl) && args_are_copies(ah, bh)) { return do_constant_folding_cond_eq(c); } return 2; @@ -544,8 +558,8 @@ static bool swap_commutative(TCGArg dest, TCGArg *p1, TCGArg *p2) { TCGArg a1 = *p1, a2 = *p2; int sum = 0; - sum += temps[a1].state == TCG_TEMP_CONST; - sum -= temps[a2].state == TCG_TEMP_CONST; + sum += arg_is_const(a1); + sum -= arg_is_const(a2); /* Prefer the constant in second argument, and then the form op a, a, b, which is better handled on non-RISC hosts. */ @@ -560,10 +574,10 @@ static bool swap_commutative(TCGArg dest, TCGArg *p1, TCGArg *p2) static bool swap_commutative2(TCGArg *p1, TCGArg *p2) { int sum = 0; - sum += temps[p1[0]].state == TCG_TEMP_CONST; - sum += temps[p1[1]].state == TCG_TEMP_CONST; - sum -= temps[p2[0]].state == TCG_TEMP_CONST; - sum -= temps[p2[1]].state == TCG_TEMP_CONST; + sum += arg_is_const(p1[0]); + sum += arg_is_const(p1[1]); + sum -= arg_is_const(p2[0]); + sum -= arg_is_const(p2[1]); if (sum > 0) { TCGArg t; t = p1[0], p1[0] = p2[0], p2[0] = t; @@ -576,7 +590,10 @@ static bool swap_commutative2(TCGArg *p1, TCGArg *p2) /* Propagate constants and copies, fold constant expressions. */ void tcg_optimize(TCGContext *s) { - int oi, oi_next, nb_temps, nb_globals; + int nb_temps, nb_globals; + TCGOp *op, *op_next, *prev_mb = NULL; + struct tcg_temp_info *infos; + TCGTempSet temps_used; /* Array VALS has an element for each temp. If this temp holds a constant then its value is kept in VALS' element. @@ -585,85 +602,94 @@ void tcg_optimize(TCGContext *s) nb_temps = s->nb_temps; nb_globals = s->nb_globals; - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); + infos = tcg_malloc(sizeof(struct tcg_temp_info) * nb_temps); - for (oi = s->gen_first_op_idx; oi >= 0; oi = oi_next) { + QTAILQ_FOREACH_SAFE(op, &s->ops, link, op_next) { tcg_target_ulong mask, partmask, affected; int nb_oargs, nb_iargs, i; TCGArg tmp; - - TCGOp * const op = &s->gen_op_buf[oi]; - TCGArg * const args = &s->gen_opparam_buf[op->args]; TCGOpcode opc = op->opc; const TCGOpDef *def = &tcg_op_defs[opc]; - oi_next = op->next; + /* Count the arguments, and initialize the temps that are + going to be used */ if (opc == INDEX_op_call) { - nb_oargs = op->callo; - nb_iargs = op->calli; + nb_oargs = TCGOP_CALLO(op); + nb_iargs = TCGOP_CALLI(op); + for (i = 0; i < nb_oargs + nb_iargs; i++) { + TCGTemp *ts = arg_temp(op->args[i]); + if (ts) { + init_ts_info(infos, &temps_used, ts); + } + } } else { nb_oargs = def->nb_oargs; nb_iargs = def->nb_iargs; + for (i = 0; i < nb_oargs + nb_iargs; i++) { + init_arg_info(infos, &temps_used, op->args[i]); + } } /* Do copy propagation */ for (i = nb_oargs; i < nb_oargs + nb_iargs; i++) { - if (temps[args[i]].state == TCG_TEMP_COPY) { - args[i] = find_better_copy(s, args[i]); + TCGTemp *ts = arg_temp(op->args[i]); + if (ts && ts_is_copy(ts)) { + op->args[i] = temp_arg(find_better_copy(s, ts)); } } /* For commutative operations make constant second argument */ switch (opc) { - CASE_OP_32_64(add): - CASE_OP_32_64(mul): - CASE_OP_32_64(and): - CASE_OP_32_64(or): - CASE_OP_32_64(xor): + CASE_OP_32_64_VEC(add): + CASE_OP_32_64_VEC(mul): + CASE_OP_32_64_VEC(and): + CASE_OP_32_64_VEC(or): + CASE_OP_32_64_VEC(xor): CASE_OP_32_64(eqv): CASE_OP_32_64(nand): CASE_OP_32_64(nor): CASE_OP_32_64(muluh): CASE_OP_32_64(mulsh): - swap_commutative(args[0], &args[1], &args[2]); + swap_commutative(op->args[0], &op->args[1], &op->args[2]); break; CASE_OP_32_64(brcond): - if (swap_commutative(-1, &args[0], &args[1])) { - args[2] = tcg_swap_cond(args[2]); + if (swap_commutative(-1, &op->args[0], &op->args[1])) { + op->args[2] = tcg_swap_cond(op->args[2]); } break; CASE_OP_32_64(setcond): - if (swap_commutative(args[0], &args[1], &args[2])) { - args[3] = tcg_swap_cond(args[3]); + if (swap_commutative(op->args[0], &op->args[1], &op->args[2])) { + op->args[3] = tcg_swap_cond(op->args[3]); } break; CASE_OP_32_64(movcond): - if (swap_commutative(-1, &args[1], &args[2])) { - args[5] = tcg_swap_cond(args[5]); + if (swap_commutative(-1, &op->args[1], &op->args[2])) { + op->args[5] = tcg_swap_cond(op->args[5]); } /* For movcond, we canonicalize the "false" input reg to match the destination reg so that the tcg backend can implement a "move if true" operation. */ - if (swap_commutative(args[0], &args[4], &args[3])) { - args[5] = tcg_invert_cond(args[5]); + if (swap_commutative(op->args[0], &op->args[4], &op->args[3])) { + op->args[5] = tcg_invert_cond(op->args[5]); } break; CASE_OP_32_64(add2): - swap_commutative(args[0], &args[2], &args[4]); - swap_commutative(args[1], &args[3], &args[5]); + swap_commutative(op->args[0], &op->args[2], &op->args[4]); + swap_commutative(op->args[1], &op->args[3], &op->args[5]); break; CASE_OP_32_64(mulu2): CASE_OP_32_64(muls2): - swap_commutative(args[0], &args[2], &args[3]); + swap_commutative(op->args[0], &op->args[2], &op->args[3]); break; case INDEX_op_brcond2_i32: - if (swap_commutative2(&args[0], &args[2])) { - args[4] = tcg_swap_cond(args[4]); + if (swap_commutative2(&op->args[0], &op->args[2])) { + op->args[4] = tcg_swap_cond(op->args[4]); } break; case INDEX_op_setcond2_i32: - if (swap_commutative2(&args[1], &args[3])) { - args[5] = tcg_swap_cond(args[5]); + if (swap_commutative2(&op->args[1], &op->args[3])) { + op->args[5] = tcg_swap_cond(op->args[5]); } break; default: @@ -678,70 +704,73 @@ void tcg_optimize(TCGContext *s) CASE_OP_32_64(sar): CASE_OP_32_64(rotl): CASE_OP_32_64(rotr): - if (temps[args[1]].state == TCG_TEMP_CONST - && temps[args[1]].val == 0) { - tcg_opt_gen_movi(s, op, args, args[0], 0); + if (arg_is_const(op->args[1]) + && arg_info(op->args[1])->val == 0) { + tcg_opt_gen_movi(s, op, op->args[0], 0); continue; } break; - CASE_OP_32_64(sub): + CASE_OP_32_64_VEC(sub): { TCGOpcode neg_op; bool have_neg; - if (temps[args[2]].state == TCG_TEMP_CONST) { + if (arg_is_const(op->args[2])) { /* Proceed with possible constant folding. */ break; } if (opc == INDEX_op_sub_i32) { neg_op = INDEX_op_neg_i32; have_neg = TCG_TARGET_HAS_neg_i32; - } else { + } else if (opc == INDEX_op_sub_i64) { neg_op = INDEX_op_neg_i64; have_neg = TCG_TARGET_HAS_neg_i64; + } else { + neg_op = INDEX_op_neg_vec; + have_neg = TCG_TARGET_HAS_neg_vec; } if (!have_neg) { break; } - if (temps[args[1]].state == TCG_TEMP_CONST - && temps[args[1]].val == 0) { + if (arg_is_const(op->args[1]) + && arg_info(op->args[1])->val == 0) { op->opc = neg_op; - reset_temp(args[0]); - args[1] = args[2]; + reset_temp(op->args[0]); + op->args[1] = op->args[2]; continue; } } break; - CASE_OP_32_64(xor): + CASE_OP_32_64_VEC(xor): CASE_OP_32_64(nand): - if (temps[args[1]].state != TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST - && temps[args[2]].val == -1) { + if (!arg_is_const(op->args[1]) + && arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == -1) { i = 1; goto try_not; } break; CASE_OP_32_64(nor): - if (temps[args[1]].state != TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST - && temps[args[2]].val == 0) { + if (!arg_is_const(op->args[1]) + && arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == 0) { i = 1; goto try_not; } break; - CASE_OP_32_64(andc): - if (temps[args[2]].state != TCG_TEMP_CONST - && temps[args[1]].state == TCG_TEMP_CONST - && temps[args[1]].val == -1) { + CASE_OP_32_64_VEC(andc): + if (!arg_is_const(op->args[2]) + && arg_is_const(op->args[1]) + && arg_info(op->args[1])->val == -1) { i = 2; goto try_not; } break; - CASE_OP_32_64(orc): + CASE_OP_32_64_VEC(orc): CASE_OP_32_64(eqv): - if (temps[args[2]].state != TCG_TEMP_CONST - && temps[args[1]].state == TCG_TEMP_CONST - && temps[args[1]].val == 0) { + if (!arg_is_const(op->args[2]) + && arg_is_const(op->args[1]) + && arg_info(op->args[1])->val == 0) { i = 2; goto try_not; } @@ -751,7 +780,10 @@ void tcg_optimize(TCGContext *s) TCGOpcode not_op; bool have_not; - if (def->flags & TCG_OPF_64BIT) { + if (def->flags & TCG_OPF_VECTOR) { + not_op = INDEX_op_not_vec; + have_not = TCG_TARGET_HAS_not_vec; + } else if (def->flags & TCG_OPF_64BIT) { not_op = INDEX_op_not_i64; have_not = TCG_TARGET_HAS_not_i64; } else { @@ -762,8 +794,8 @@ void tcg_optimize(TCGContext *s) break; } op->opc = not_op; - reset_temp(args[0]); - args[1] = args[i]; + reset_temp(op->args[0]); + op->args[1] = op->args[i]; continue; } default: @@ -772,30 +804,30 @@ void tcg_optimize(TCGContext *s) /* Simplify expression for "op r, a, const => mov r, a" cases */ switch (opc) { - CASE_OP_32_64(add): - CASE_OP_32_64(sub): + CASE_OP_32_64_VEC(add): + CASE_OP_32_64_VEC(sub): + CASE_OP_32_64_VEC(or): + CASE_OP_32_64_VEC(xor): + CASE_OP_32_64_VEC(andc): CASE_OP_32_64(shl): CASE_OP_32_64(shr): CASE_OP_32_64(sar): CASE_OP_32_64(rotl): CASE_OP_32_64(rotr): - CASE_OP_32_64(or): - CASE_OP_32_64(xor): - CASE_OP_32_64(andc): - if (temps[args[1]].state != TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST - && temps[args[2]].val == 0) { - tcg_opt_gen_mov(s, op, args, args[0], args[1]); + if (!arg_is_const(op->args[1]) + && arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == 0) { + tcg_opt_gen_mov(s, op, op->args[0], op->args[1]); continue; } break; - CASE_OP_32_64(and): - CASE_OP_32_64(orc): + CASE_OP_32_64_VEC(and): + CASE_OP_32_64_VEC(orc): CASE_OP_32_64(eqv): - if (temps[args[1]].state != TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST - && temps[args[2]].val == -1) { - tcg_opt_gen_mov(s, op, args, args[0], args[1]); + if (!arg_is_const(op->args[1]) + && arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == -1) { + tcg_opt_gen_mov(s, op, op->args[0], op->args[1]); continue; } break; @@ -809,21 +841,21 @@ void tcg_optimize(TCGContext *s) affected = -1; switch (opc) { CASE_OP_32_64(ext8s): - if ((temps[args[1]].mask & 0x80) != 0) { + if ((arg_info(op->args[1])->mask & 0x80) != 0) { break; } CASE_OP_32_64(ext8u): mask = 0xff; goto and_const; CASE_OP_32_64(ext16s): - if ((temps[args[1]].mask & 0x8000) != 0) { + if ((arg_info(op->args[1])->mask & 0x8000) != 0) { break; } CASE_OP_32_64(ext16u): mask = 0xffff; goto and_const; case INDEX_op_ext32s_i64: - if ((temps[args[1]].mask & 0x80000000) != 0) { + if ((arg_info(op->args[1])->mask & 0x80000000) != 0) { break; } case INDEX_op_ext32u_i64: @@ -831,75 +863,121 @@ void tcg_optimize(TCGContext *s) goto and_const; CASE_OP_32_64(and): - mask = temps[args[2]].mask; - if (temps[args[2]].state == TCG_TEMP_CONST) { + mask = arg_info(op->args[2])->mask; + if (arg_is_const(op->args[2])) { and_const: - affected = temps[args[1]].mask & ~mask; + affected = arg_info(op->args[1])->mask & ~mask; + } + mask = arg_info(op->args[1])->mask & mask; + break; + + case INDEX_op_ext_i32_i64: + if ((arg_info(op->args[1])->mask & 0x80000000) != 0) { + break; } - mask = temps[args[1]].mask & mask; + case INDEX_op_extu_i32_i64: + /* We do not compute affected as it is a size changing op. */ + mask = (uint32_t)arg_info(op->args[1])->mask; break; CASE_OP_32_64(andc): /* Known-zeros does not imply known-ones. Therefore unless - args[2] is constant, we can't infer anything from it. */ - if (temps[args[2]].state == TCG_TEMP_CONST) { - mask = ~temps[args[2]].mask; + op->args[2] is constant, we can't infer anything from it. */ + if (arg_is_const(op->args[2])) { + mask = ~arg_info(op->args[2])->mask; goto and_const; } /* But we certainly know nothing outside args[1] may be set. */ - mask = temps[args[1]].mask; + mask = arg_info(op->args[1])->mask; break; case INDEX_op_sar_i32: - if (temps[args[2]].state == TCG_TEMP_CONST) { - tmp = temps[args[2]].val & 31; - mask = (int32_t)temps[args[1]].mask >> tmp; + if (arg_is_const(op->args[2])) { + tmp = arg_info(op->args[2])->val & 31; + mask = (int32_t)arg_info(op->args[1])->mask >> tmp; } break; case INDEX_op_sar_i64: - if (temps[args[2]].state == TCG_TEMP_CONST) { - tmp = temps[args[2]].val & 63; - mask = (int64_t)temps[args[1]].mask >> tmp; + if (arg_is_const(op->args[2])) { + tmp = arg_info(op->args[2])->val & 63; + mask = (int64_t)arg_info(op->args[1])->mask >> tmp; } break; case INDEX_op_shr_i32: - if (temps[args[2]].state == TCG_TEMP_CONST) { - tmp = temps[args[2]].val & 31; - mask = (uint32_t)temps[args[1]].mask >> tmp; + if (arg_is_const(op->args[2])) { + tmp = arg_info(op->args[2])->val & 31; + mask = (uint32_t)arg_info(op->args[1])->mask >> tmp; } break; case INDEX_op_shr_i64: - if (temps[args[2]].state == TCG_TEMP_CONST) { - tmp = temps[args[2]].val & 63; - mask = (uint64_t)temps[args[1]].mask >> tmp; + if (arg_is_const(op->args[2])) { + tmp = arg_info(op->args[2])->val & 63; + mask = (uint64_t)arg_info(op->args[1])->mask >> tmp; } break; - case INDEX_op_trunc_shr_i32: - mask = (uint64_t)temps[args[1]].mask >> args[2]; + case INDEX_op_extrl_i64_i32: + mask = (uint32_t)arg_info(op->args[1])->mask; + break; + case INDEX_op_extrh_i64_i32: + mask = (uint64_t)arg_info(op->args[1])->mask >> 32; break; CASE_OP_32_64(shl): - if (temps[args[2]].state == TCG_TEMP_CONST) { - tmp = temps[args[2]].val & (TCG_TARGET_REG_BITS - 1); - mask = temps[args[1]].mask << tmp; + if (arg_is_const(op->args[2])) { + tmp = arg_info(op->args[2])->val & (TCG_TARGET_REG_BITS - 1); + mask = arg_info(op->args[1])->mask << tmp; } break; CASE_OP_32_64(neg): /* Set to 1 all bits to the left of the rightmost. */ - mask = -(temps[args[1]].mask & -temps[args[1]].mask); + mask = -(arg_info(op->args[1])->mask + & -arg_info(op->args[1])->mask); break; CASE_OP_32_64(deposit): - mask = deposit64(temps[args[1]].mask, args[3], args[4], - temps[args[2]].mask); + mask = deposit64(arg_info(op->args[1])->mask, + op->args[3], op->args[4], + arg_info(op->args[2])->mask); + break; + + CASE_OP_32_64(extract): + mask = extract64(arg_info(op->args[1])->mask, + op->args[2], op->args[3]); + if (op->args[2] == 0) { + affected = arg_info(op->args[1])->mask & ~mask; + } + break; + CASE_OP_32_64(sextract): + mask = sextract64(arg_info(op->args[1])->mask, + op->args[2], op->args[3]); + if (op->args[2] == 0 && (tcg_target_long)mask >= 0) { + affected = arg_info(op->args[1])->mask & ~mask; + } break; CASE_OP_32_64(or): CASE_OP_32_64(xor): - mask = temps[args[1]].mask | temps[args[2]].mask; + mask = arg_info(op->args[1])->mask | arg_info(op->args[2])->mask; + break; + + case INDEX_op_clz_i32: + case INDEX_op_ctz_i32: + mask = arg_info(op->args[2])->mask | 31; + break; + + case INDEX_op_clz_i64: + case INDEX_op_ctz_i64: + mask = arg_info(op->args[2])->mask | 63; + break; + + case INDEX_op_ctpop_i32: + mask = 32 | 31; + break; + case INDEX_op_ctpop_i64: + mask = 64 | 63; break; CASE_OP_32_64(setcond): @@ -908,7 +986,7 @@ void tcg_optimize(TCGContext *s) break; CASE_OP_32_64(movcond): - mask = temps[args[3]].mask | temps[args[4]].mask; + mask = arg_info(op->args[3])->mask | arg_info(op->args[4])->mask; break; CASE_OP_32_64(ld8u): @@ -923,7 +1001,7 @@ void tcg_optimize(TCGContext *s) CASE_OP_32_64(qemu_ld): { - TCGMemOpIdx oi = args[nb_oargs + nb_iargs]; + TCGMemOpIdx oi = op->args[nb_oargs + nb_iargs]; TCGMemOp mop = get_memop(oi); if (!(mop & MO_SIGN)) { mask = (2ULL << ((8 << (mop & MO_SIZE)) - 1)) - 1; @@ -946,25 +1024,25 @@ void tcg_optimize(TCGContext *s) } if (partmask == 0) { - assert(nb_oargs == 1); - tcg_opt_gen_movi(s, op, args, args[0], 0); + tcg_debug_assert(nb_oargs == 1); + tcg_opt_gen_movi(s, op, op->args[0], 0); continue; } if (affected == 0) { - assert(nb_oargs == 1); - tcg_opt_gen_mov(s, op, args, args[0], args[1]); + tcg_debug_assert(nb_oargs == 1); + tcg_opt_gen_mov(s, op, op->args[0], op->args[1]); continue; } /* Simplify expression for "op r, a, 0 => movi r, 0" cases */ switch (opc) { - CASE_OP_32_64(and): - CASE_OP_32_64(mul): + CASE_OP_32_64_VEC(and): + CASE_OP_32_64_VEC(mul): CASE_OP_32_64(muluh): CASE_OP_32_64(mulsh): - if ((temps[args[2]].state == TCG_TEMP_CONST - && temps[args[2]].val == 0)) { - tcg_opt_gen_movi(s, op, args, args[0], 0); + if (arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == 0) { + tcg_opt_gen_movi(s, op, op->args[0], 0); continue; } break; @@ -974,10 +1052,10 @@ void tcg_optimize(TCGContext *s) /* Simplify expression for "op r, a, a => mov r, a" cases */ switch (opc) { - CASE_OP_32_64(or): - CASE_OP_32_64(and): - if (temps_are_copies(args[1], args[2])) { - tcg_opt_gen_mov(s, op, args, args[0], args[1]); + CASE_OP_32_64_VEC(or): + CASE_OP_32_64_VEC(and): + if (args_are_copies(op->args[1], op->args[2])) { + tcg_opt_gen_mov(s, op, op->args[0], op->args[1]); continue; } break; @@ -987,11 +1065,11 @@ void tcg_optimize(TCGContext *s) /* Simplify expression for "op r, a, a => movi r, 0" cases */ switch (opc) { - CASE_OP_32_64(andc): - CASE_OP_32_64(sub): - CASE_OP_32_64(xor): - if (temps_are_copies(args[1], args[2])) { - tcg_opt_gen_movi(s, op, args, args[0], 0); + CASE_OP_32_64_VEC(andc): + CASE_OP_32_64_VEC(sub): + CASE_OP_32_64_VEC(xor): + if (args_are_copies(op->args[1], op->args[2])) { + tcg_opt_gen_movi(s, op, op->args[0], 0); continue; } break; @@ -1003,32 +1081,39 @@ void tcg_optimize(TCGContext *s) folding. Constants will be substituted to arguments by register allocator where needed and possible. Also detect copies. */ switch (opc) { - CASE_OP_32_64(mov): - tcg_opt_gen_mov(s, op, args, args[0], args[1]); + CASE_OP_32_64_VEC(mov): + tcg_opt_gen_mov(s, op, op->args[0], op->args[1]); break; CASE_OP_32_64(movi): - tcg_opt_gen_movi(s, op, args, args[0], args[1]); + case INDEX_op_dupi_vec: + tcg_opt_gen_movi(s, op, op->args[0], op->args[1]); break; + case INDEX_op_dup_vec: + if (arg_is_const(op->args[1])) { + tmp = arg_info(op->args[1])->val; + tmp = dup_const(TCGOP_VECE(op), tmp); + tcg_opt_gen_movi(s, op, op->args[0], tmp); + break; + } + goto do_default; + CASE_OP_32_64(not): CASE_OP_32_64(neg): CASE_OP_32_64(ext8s): CASE_OP_32_64(ext8u): CASE_OP_32_64(ext16s): CASE_OP_32_64(ext16u): + CASE_OP_32_64(ctpop): case INDEX_op_ext32s_i64: case INDEX_op_ext32u_i64: - if (temps[args[1]].state == TCG_TEMP_CONST) { - tmp = do_constant_folding(opc, temps[args[1]].val, 0); - tcg_opt_gen_movi(s, op, args, args[0], tmp); - break; - } - goto do_default; - - case INDEX_op_trunc_shr_i32: - if (temps[args[1]].state == TCG_TEMP_CONST) { - tmp = do_constant_folding(opc, temps[args[1]].val, args[2]); - tcg_opt_gen_movi(s, op, args, args[0], tmp); + case INDEX_op_ext_i32_i64: + case INDEX_op_extu_i32_i64: + case INDEX_op_extrl_i64_i32: + case INDEX_op_extrh_i64_i32: + if (arg_is_const(op->args[1])) { + tmp = do_constant_folding(opc, arg_info(op->args[1])->val, 0); + tcg_opt_gen_movi(s, op, op->args[0], tmp); break; } goto do_default; @@ -1055,40 +1140,73 @@ void tcg_optimize(TCGContext *s) CASE_OP_32_64(divu): CASE_OP_32_64(rem): CASE_OP_32_64(remu): - if (temps[args[1]].state == TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST) { - tmp = do_constant_folding(opc, temps[args[1]].val, - temps[args[2]].val); - tcg_opt_gen_movi(s, op, args, args[0], tmp); + if (arg_is_const(op->args[1]) && arg_is_const(op->args[2])) { + tmp = do_constant_folding(opc, arg_info(op->args[1])->val, + arg_info(op->args[2])->val); + tcg_opt_gen_movi(s, op, op->args[0], tmp); + break; + } + goto do_default; + + CASE_OP_32_64(clz): + CASE_OP_32_64(ctz): + if (arg_is_const(op->args[1])) { + TCGArg v = arg_info(op->args[1])->val; + if (v != 0) { + tmp = do_constant_folding(opc, v, 0); + tcg_opt_gen_movi(s, op, op->args[0], tmp); + } else { + tcg_opt_gen_mov(s, op, op->args[0], op->args[2]); + } break; } goto do_default; CASE_OP_32_64(deposit): - if (temps[args[1]].state == TCG_TEMP_CONST - && temps[args[2]].state == TCG_TEMP_CONST) { - tmp = deposit64(temps[args[1]].val, args[3], args[4], - temps[args[2]].val); - tcg_opt_gen_movi(s, op, args, args[0], tmp); + if (arg_is_const(op->args[1]) && arg_is_const(op->args[2])) { + tmp = deposit64(arg_info(op->args[1])->val, + op->args[3], op->args[4], + arg_info(op->args[2])->val); + tcg_opt_gen_movi(s, op, op->args[0], tmp); + break; + } + goto do_default; + + CASE_OP_32_64(extract): + if (arg_is_const(op->args[1])) { + tmp = extract64(arg_info(op->args[1])->val, + op->args[2], op->args[3]); + tcg_opt_gen_movi(s, op, op->args[0], tmp); + break; + } + goto do_default; + + CASE_OP_32_64(sextract): + if (arg_is_const(op->args[1])) { + tmp = sextract64(arg_info(op->args[1])->val, + op->args[2], op->args[3]); + tcg_opt_gen_movi(s, op, op->args[0], tmp); break; } goto do_default; CASE_OP_32_64(setcond): - tmp = do_constant_folding_cond(opc, args[1], args[2], args[3]); + tmp = do_constant_folding_cond(opc, op->args[1], + op->args[2], op->args[3]); if (tmp != 2) { - tcg_opt_gen_movi(s, op, args, args[0], tmp); + tcg_opt_gen_movi(s, op, op->args[0], tmp); break; } goto do_default; CASE_OP_32_64(brcond): - tmp = do_constant_folding_cond(opc, args[0], args[1], args[2]); + tmp = do_constant_folding_cond(opc, op->args[0], + op->args[1], op->args[2]); if (tmp != 2) { if (tmp) { - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); op->opc = INDEX_op_br; - args[0] = args[3]; + op->args[0] = op->args[3]; } else { tcg_op_remove(s, op); } @@ -1097,28 +1215,41 @@ void tcg_optimize(TCGContext *s) goto do_default; CASE_OP_32_64(movcond): - tmp = do_constant_folding_cond(opc, args[1], args[2], args[5]); + tmp = do_constant_folding_cond(opc, op->args[1], + op->args[2], op->args[5]); if (tmp != 2) { - tcg_opt_gen_mov(s, op, args, args[0], args[4-tmp]); + tcg_opt_gen_mov(s, op, op->args[0], op->args[4-tmp]); break; } + if (arg_is_const(op->args[3]) && arg_is_const(op->args[4])) { + tcg_target_ulong tv = arg_info(op->args[3])->val; + tcg_target_ulong fv = arg_info(op->args[4])->val; + TCGCond cond = op->args[5]; + if (fv == 1 && tv == 0) { + cond = tcg_invert_cond(cond); + } else if (!(tv == 1 && fv == 0)) { + goto do_default; + } + op->args[3] = cond; + op->opc = opc = (opc == INDEX_op_movcond_i32 + ? INDEX_op_setcond_i32 + : INDEX_op_setcond_i64); + nb_iargs = 2; + } goto do_default; case INDEX_op_add2_i32: case INDEX_op_sub2_i32: - if (temps[args[2]].state == TCG_TEMP_CONST - && temps[args[3]].state == TCG_TEMP_CONST - && temps[args[4]].state == TCG_TEMP_CONST - && temps[args[5]].state == TCG_TEMP_CONST) { - uint32_t al = temps[args[2]].val; - uint32_t ah = temps[args[3]].val; - uint32_t bl = temps[args[4]].val; - uint32_t bh = temps[args[5]].val; + if (arg_is_const(op->args[2]) && arg_is_const(op->args[3]) + && arg_is_const(op->args[4]) && arg_is_const(op->args[5])) { + uint32_t al = arg_info(op->args[2])->val; + uint32_t ah = arg_info(op->args[3])->val; + uint32_t bl = arg_info(op->args[4])->val; + uint32_t bh = arg_info(op->args[5])->val; uint64_t a = ((uint64_t)ah << 32) | al; uint64_t b = ((uint64_t)bh << 32) | bl; TCGArg rl, rh; - TCGOp *op2 = insert_op_before(s, op, INDEX_op_movi_i32, 2); - TCGArg *args2 = &s->gen_opparam_buf[op2->args]; + TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_movi_i32, 2); if (opc == INDEX_op_add2_i32) { a += b; @@ -1126,99 +1257,97 @@ void tcg_optimize(TCGContext *s) a -= b; } - rl = args[0]; - rh = args[1]; - tcg_opt_gen_movi(s, op, args, rl, (int32_t)a); - tcg_opt_gen_movi(s, op2, args2, rh, (int32_t)(a >> 32)); - - /* We've done all we need to do with the movi. Skip it. */ - oi_next = op2->next; + rl = op->args[0]; + rh = op->args[1]; + tcg_opt_gen_movi(s, op, rl, (int32_t)a); + tcg_opt_gen_movi(s, op2, rh, (int32_t)(a >> 32)); break; } goto do_default; case INDEX_op_mulu2_i32: - if (temps[args[2]].state == TCG_TEMP_CONST - && temps[args[3]].state == TCG_TEMP_CONST) { - uint32_t a = temps[args[2]].val; - uint32_t b = temps[args[3]].val; + if (arg_is_const(op->args[2]) && arg_is_const(op->args[3])) { + uint32_t a = arg_info(op->args[2])->val; + uint32_t b = arg_info(op->args[3])->val; uint64_t r = (uint64_t)a * b; TCGArg rl, rh; - TCGOp *op2 = insert_op_before(s, op, INDEX_op_movi_i32, 2); - TCGArg *args2 = &s->gen_opparam_buf[op2->args]; - - rl = args[0]; - rh = args[1]; - tcg_opt_gen_movi(s, op, args, rl, (int32_t)r); - tcg_opt_gen_movi(s, op2, args2, rh, (int32_t)(r >> 32)); + TCGOp *op2 = tcg_op_insert_before(s, op, INDEX_op_movi_i32, 2); - /* We've done all we need to do with the movi. Skip it. */ - oi_next = op2->next; + rl = op->args[0]; + rh = op->args[1]; + tcg_opt_gen_movi(s, op, rl, (int32_t)r); + tcg_opt_gen_movi(s, op2, rh, (int32_t)(r >> 32)); break; } goto do_default; case INDEX_op_brcond2_i32: - tmp = do_constant_folding_cond2(&args[0], &args[2], args[4]); + tmp = do_constant_folding_cond2(&op->args[0], &op->args[2], + op->args[4]); if (tmp != 2) { if (tmp) { do_brcond_true: - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); op->opc = INDEX_op_br; - args[0] = args[5]; + op->args[0] = op->args[5]; } else { do_brcond_false: tcg_op_remove(s, op); } - } else if ((args[4] == TCG_COND_LT || args[4] == TCG_COND_GE) - && temps[args[2]].state == TCG_TEMP_CONST - && temps[args[3]].state == TCG_TEMP_CONST - && temps[args[2]].val == 0 - && temps[args[3]].val == 0) { + } else if ((op->args[4] == TCG_COND_LT + || op->args[4] == TCG_COND_GE) + && arg_is_const(op->args[2]) + && arg_info(op->args[2])->val == 0 + && arg_is_const(op->args[3]) + && arg_info(op->args[3])->val == 0) { /* Simplify LT/GE comparisons vs zero to a single compare vs the high word of the input. */ do_brcond_high: - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); op->opc = INDEX_op_brcond_i32; - args[0] = args[1]; - args[1] = args[3]; - args[2] = args[4]; - args[3] = args[5]; - } else if (args[4] == TCG_COND_EQ) { + op->args[0] = op->args[1]; + op->args[1] = op->args[3]; + op->args[2] = op->args[4]; + op->args[3] = op->args[5]; + } else if (op->args[4] == TCG_COND_EQ) { /* Simplify EQ comparisons where one of the pairs can be simplified. */ tmp = do_constant_folding_cond(INDEX_op_brcond_i32, - args[0], args[2], TCG_COND_EQ); + op->args[0], op->args[2], + TCG_COND_EQ); if (tmp == 0) { goto do_brcond_false; } else if (tmp == 1) { goto do_brcond_high; } tmp = do_constant_folding_cond(INDEX_op_brcond_i32, - args[1], args[3], TCG_COND_EQ); + op->args[1], op->args[3], + TCG_COND_EQ); if (tmp == 0) { goto do_brcond_false; } else if (tmp != 1) { goto do_default; } do_brcond_low: - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); op->opc = INDEX_op_brcond_i32; - args[1] = args[2]; - args[2] = args[4]; - args[3] = args[5]; - } else if (args[4] == TCG_COND_NE) { + op->args[1] = op->args[2]; + op->args[2] = op->args[4]; + op->args[3] = op->args[5]; + } else if (op->args[4] == TCG_COND_NE) { /* Simplify NE comparisons where one of the pairs can be simplified. */ tmp = do_constant_folding_cond(INDEX_op_brcond_i32, - args[0], args[2], TCG_COND_NE); + op->args[0], op->args[2], + TCG_COND_NE); if (tmp == 0) { goto do_brcond_high; } else if (tmp == 1) { goto do_brcond_true; } tmp = do_constant_folding_cond(INDEX_op_brcond_i32, - args[1], args[3], TCG_COND_NE); + op->args[1], op->args[3], + TCG_COND_NE); if (tmp == 0) { goto do_brcond_low; } else if (tmp == 1) { @@ -1231,59 +1360,65 @@ void tcg_optimize(TCGContext *s) break; case INDEX_op_setcond2_i32: - tmp = do_constant_folding_cond2(&args[1], &args[3], args[5]); + tmp = do_constant_folding_cond2(&op->args[1], &op->args[3], + op->args[5]); if (tmp != 2) { do_setcond_const: - tcg_opt_gen_movi(s, op, args, args[0], tmp); - } else if ((args[5] == TCG_COND_LT || args[5] == TCG_COND_GE) - && temps[args[3]].state == TCG_TEMP_CONST - && temps[args[4]].state == TCG_TEMP_CONST - && temps[args[3]].val == 0 - && temps[args[4]].val == 0) { + tcg_opt_gen_movi(s, op, op->args[0], tmp); + } else if ((op->args[5] == TCG_COND_LT + || op->args[5] == TCG_COND_GE) + && arg_is_const(op->args[3]) + && arg_info(op->args[3])->val == 0 + && arg_is_const(op->args[4]) + && arg_info(op->args[4])->val == 0) { /* Simplify LT/GE comparisons vs zero to a single compare vs the high word of the input. */ do_setcond_high: - reset_temp(args[0]); - temps[args[0]].mask = 1; + reset_temp(op->args[0]); + arg_info(op->args[0])->mask = 1; op->opc = INDEX_op_setcond_i32; - args[1] = args[2]; - args[2] = args[4]; - args[3] = args[5]; - } else if (args[5] == TCG_COND_EQ) { + op->args[1] = op->args[2]; + op->args[2] = op->args[4]; + op->args[3] = op->args[5]; + } else if (op->args[5] == TCG_COND_EQ) { /* Simplify EQ comparisons where one of the pairs can be simplified. */ tmp = do_constant_folding_cond(INDEX_op_setcond_i32, - args[1], args[3], TCG_COND_EQ); + op->args[1], op->args[3], + TCG_COND_EQ); if (tmp == 0) { goto do_setcond_const; } else if (tmp == 1) { goto do_setcond_high; } tmp = do_constant_folding_cond(INDEX_op_setcond_i32, - args[2], args[4], TCG_COND_EQ); + op->args[2], op->args[4], + TCG_COND_EQ); if (tmp == 0) { goto do_setcond_high; } else if (tmp != 1) { goto do_default; } do_setcond_low: - reset_temp(args[0]); - temps[args[0]].mask = 1; + reset_temp(op->args[0]); + arg_info(op->args[0])->mask = 1; op->opc = INDEX_op_setcond_i32; - args[2] = args[3]; - args[3] = args[5]; - } else if (args[5] == TCG_COND_NE) { + op->args[2] = op->args[3]; + op->args[3] = op->args[5]; + } else if (op->args[5] == TCG_COND_NE) { /* Simplify NE comparisons where one of the pairs can be simplified. */ tmp = do_constant_folding_cond(INDEX_op_setcond_i32, - args[1], args[3], TCG_COND_NE); + op->args[1], op->args[3], + TCG_COND_NE); if (tmp == 0) { goto do_setcond_high; } else if (tmp == 1) { goto do_setcond_const; } tmp = do_constant_folding_cond(INDEX_op_setcond_i32, - args[2], args[4], TCG_COND_NE); + op->args[2], op->args[4], + TCG_COND_NE); if (tmp == 0) { goto do_setcond_low; } else if (tmp == 1) { @@ -1296,10 +1431,12 @@ void tcg_optimize(TCGContext *s) break; case INDEX_op_call: - if (!(args[nb_oargs + nb_iargs + 1] + if (!(op->args[nb_oargs + nb_iargs + 1] & (TCG_CALL_NO_READ_GLOBALS | TCG_CALL_NO_WRITE_GLOBALS))) { for (i = 0; i < nb_globals; i++) { - reset_temp(i); + if (test_bit(i, temps_used.l)) { + reset_ts(&s->temps[i]); + } } } goto do_reset_output; @@ -1312,19 +1449,57 @@ void tcg_optimize(TCGContext *s) block, otherwise we only trash the output args. "mask" is the non-zero bits mask for the first output arg. */ if (def->flags & TCG_OPF_BB_END) { - reset_all_temps(nb_temps); + bitmap_zero(temps_used.l, nb_temps); } else { do_reset_output: for (i = 0; i < nb_oargs; i++) { - reset_temp(args[i]); + reset_temp(op->args[i]); /* Save the corresponding known-zero bits mask for the first output argument (only one supported so far). */ if (i == 0) { - temps[args[i]].mask = mask; + arg_info(op->args[i])->mask = mask; } } } break; } + + /* Eliminate duplicate and redundant fence instructions. */ + if (prev_mb) { + switch (opc) { + case INDEX_op_mb: + /* Merge two barriers of the same type into one, + * or a weaker barrier into a stronger one, + * or two weaker barriers into a stronger one. + * mb X; mb Y => mb X|Y + * mb; strl => mb; st + * ldaq; mb => ld; mb + * ldaq; strl => ld; mb; st + * Other combinations are also merged into a strong + * barrier. This is stricter than specified but for + * the purposes of TCG is better than not optimizing. + */ + prev_mb->args[0] |= op->args[0]; + tcg_op_remove(s, op); + break; + + default: + /* Opcodes that end the block stop the optimization. */ + if ((def->flags & TCG_OPF_BB_END) == 0) { + break; + } + /* fallthru */ + case INDEX_op_qemu_ld_i32: + case INDEX_op_qemu_ld_i64: + case INDEX_op_qemu_st_i32: + case INDEX_op_qemu_st_i64: + case INDEX_op_call: + /* Opcodes that touch guest memory stop the optimization. */ + prev_mb = NULL; + break; + } + } else if (opc == INDEX_op_mb) { + prev_mb = op; + } } }