/*
- * Copyright (C) 1995-2007 University of Karlsruhe. All right reserved.
+ * Copyright (C) 1995-2008 University of Karlsruhe. All right reserved.
*
* This file is part of libFirm.
*
* to int conversion which are specified as truncation in the C standard we have
* to spill, change and restore the fpu rounding mode between spills.
*/
-#ifdef HAVE_CONFIG_H
#include "config.h"
-#endif
#include "ia32_fpu.h"
#include "ia32_new_nodes.h"
+#include "ia32_architecture.h"
#include "gen_ia32_regalloc_if.h"
#include "ircons.h"
#include "array.h"
#include "../beirgmod.h"
-#include "../bearch_t.h"
+#include "../bearch.h"
#include "../besched.h"
#include "../beabi.h"
-#include "../benode_t.h"
+#include "../benode.h"
#include "../bestate.h"
#include "../beutil.h"
#include "../bessaconstr.h"
-#include "../beirg_t.h"
+#include "../beirg.h"
+
+static ir_entity *fpcw_round = NULL;
+static ir_entity *fpcw_truncate = NULL;
+
+static ir_entity *create_ent(int value, const char *name)
+{
+ ir_mode *mode = mode_Hu;
+ ir_type *type = new_type_primitive(mode);
+ ir_type *glob = get_glob_type();
+ ir_graph *cnst_irg;
+ ir_entity *ent;
+ ir_node *cnst;
+ tarval *tv;
+
+ set_type_alignment_bytes(type, 4);
+
+ tv = new_tarval_from_long(value, mode);
+ ent = new_entity(glob, new_id_from_str(name), type);
+ set_entity_ld_ident(ent, get_entity_ident(ent));
+ set_entity_visibility(ent, ir_visibility_local);
+ add_entity_linkage(ent, IR_LINKAGE_CONSTANT);
+
+ cnst_irg = get_const_code_irg();
+ cnst = new_r_Const(cnst_irg, tv);
+ set_atomic_ent_value(ent, cnst);
+
+ return ent;
+}
+
+static void create_fpcw_entities(void)
+{
+ fpcw_round = create_ent(0xc7f, "_fpcw_round");
+ fpcw_truncate = create_ent(0x37f, "_fpcw_truncate");
+}
static ir_node *create_fpu_mode_spill(void *env, ir_node *state, int force,
ir_node *after)
{
- ia32_code_gen_t *cg = env;
ir_node *spill = NULL;
+ (void) env;
+
+ /* we don't spill the fpcw in unsafe mode */
+ if (ia32_cg_config.use_unsafe_floatconv) {
+ ir_node *block = get_nodes_block(state);
+ if (force == 1 || !is_ia32_ChangeCW(state)) {
+ ir_node *spill = new_bd_ia32_FnstCWNOP(NULL, block, state);
+ sched_add_after(after, spill);
+ return spill;
+ }
+ return NULL;
+ }
- if(force == 1 || !is_ia32_ChangeCW(state)) {
+ if (force == 1 || !is_ia32_ChangeCW(state)) {
ir_graph *irg = get_irn_irg(state);
ir_node *block = get_nodes_block(state);
- ir_node *noreg = ia32_new_NoReg_gp(cg);
+ ir_node *noreg = ia32_new_NoReg_gp(irg);
ir_node *nomem = new_NoMem();
ir_node *frame = get_irg_frame(irg);
- spill = new_rd_ia32_FnstCW(NULL, irg, block, frame, noreg, nomem, state);
+ spill = new_bd_ia32_FnstCW(NULL, block, frame, noreg, nomem, state);
set_ia32_op_type(spill, ia32_AddrModeD);
- set_ia32_ls_mode(spill, ia32_reg_classes[CLASS_ia32_fp_cw].mode);
+ /* use mode_Iu, as movl has a shorter opcode than movw */
+ set_ia32_ls_mode(spill, mode_Iu);
set_ia32_use_frame(spill);
- sched_add_after(after, spill);
+ sched_add_after(skip_Proj(after), spill);
}
return spill;
}
+static ir_node *create_fldcw_ent(ir_node *block, ir_entity *entity)
+{
+ ir_graph *irg = get_irn_irg(block);
+ ir_node *nomem = new_NoMem();
+ ir_node *noreg = ia32_new_NoReg_gp(irg);
+ ir_node *reload;
+
+ reload = new_bd_ia32_FldCW(NULL, block, noreg, noreg, nomem);
+ set_ia32_op_type(reload, ia32_AddrModeS);
+ set_ia32_ls_mode(reload, ia32_reg_classes[CLASS_ia32_fp_cw].mode);
+ set_ia32_am_sc(reload, entity);
+ set_ia32_use_frame(reload);
+ arch_set_irn_register(reload, &ia32_fp_cw_regs[REG_FPCW]);
+
+ return reload;
+}
+
static ir_node *create_fpu_mode_reload(void *env, ir_node *state,
ir_node *spill, ir_node *before,
ir_node *last_state)
{
- ia32_code_gen_t *cg = env;
- ir_graph *irg = get_irn_irg(state);
- ir_node *block = get_nodes_block(before);
- ir_node *frame = get_irg_frame(irg);
- ir_node *noreg = ia32_new_NoReg_gp(cg);
- ir_node *reload = NULL;
-
- if(spill != NULL) {
- reload = new_rd_ia32_FldCW(NULL, irg, block, frame, noreg, spill);
+ ir_graph *irg = get_irn_irg(state);
+ ir_node *block = get_nodes_block(before);
+ ir_node *frame = get_irg_frame(irg);
+ ir_node *noreg = ia32_new_NoReg_gp(irg);
+ ir_node *reload = NULL;
+ (void) env;
+
+ if (ia32_cg_config.use_unsafe_floatconv) {
+ if (fpcw_round == NULL) {
+ create_fpcw_entities();
+ }
+ if (spill != NULL) {
+ reload = create_fldcw_ent(block, fpcw_round);
+ } else {
+ reload = create_fldcw_ent(block, fpcw_truncate);
+ }
+ sched_add_before(before, reload);
+ return reload;
+ }
+
+ if (spill != NULL) {
+ reload = new_bd_ia32_FldCW(NULL, block, frame, noreg, spill);
set_ia32_op_type(reload, ia32_AddrModeS);
set_ia32_ls_mode(reload, ia32_reg_classes[CLASS_ia32_fp_cw].mode);
set_ia32_use_frame(reload);
- arch_set_irn_register(cg->arch_env, reload, &ia32_fp_cw_regs[REG_FPCW]);
+ arch_set_irn_register(reload, &ia32_fp_cw_regs[REG_FPCW]);
sched_add_before(before, reload);
} else {
ir_node *or_const;
assert(last_state != NULL);
- cwstore = new_rd_ia32_FnstCW(NULL, irg, block, frame, noreg, nomem,
+ cwstore = new_bd_ia32_FnstCW(NULL, block, frame, noreg, nomem,
last_state);
set_ia32_op_type(cwstore, ia32_AddrModeD);
set_ia32_ls_mode(cwstore, lsmode);
set_ia32_use_frame(cwstore);
sched_add_before(before, cwstore);
- load = new_rd_ia32_Load(NULL, irg, block, frame, noreg, cwstore);
+ load = new_bd_ia32_Load(NULL, block, frame, noreg, cwstore);
set_ia32_op_type(load, ia32_AddrModeS);
set_ia32_ls_mode(load, lsmode);
set_ia32_use_frame(load);
sched_add_before(before, load);
- load_res = new_r_Proj(irg, block, load, mode_Iu, pn_ia32_Load_res);
+ load_res = new_r_Proj(load, mode_Iu, pn_ia32_Load_res);
/* TODO: make the actual mode configurable in ChangeCW... */
- or_const = new_rd_ia32_Immediate(NULL, irg, get_irg_start_block(irg),
- NULL, 0, 3072);
- arch_set_irn_register(cg->arch_env, or_const,
- &ia32_gp_regs[REG_GP_NOREG]);
- or = new_rd_ia32_Or(NULL, irg, block, noreg, noreg, nomem, load_res,
+ or_const = new_bd_ia32_Immediate(NULL, get_irg_start_block(irg),
+ NULL, 0, 0, 3072);
+ arch_set_irn_register(or_const, &ia32_gp_regs[REG_GP_NOREG]);
+ or = new_bd_ia32_Or(NULL, block, noreg, noreg, nomem, load_res,
or_const);
sched_add_before(before, or);
- store = new_rd_ia32_Store(NULL, irg, block, frame, noreg, nomem, or);
+ store = new_bd_ia32_Store(NULL, block, frame, noreg, nomem, or);
set_ia32_op_type(store, ia32_AddrModeD);
- set_ia32_ls_mode(store, lsmode);
+ /* use mode_Iu, as movl has a shorter opcode than movw */
+ set_ia32_ls_mode(store, mode_Iu);
set_ia32_use_frame(store);
sched_add_before(before, store);
- fldcw = new_rd_ia32_FldCW(NULL, irg, block, frame, noreg, store);
+ fldcw = new_bd_ia32_FldCW(NULL, block, frame, noreg, store);
set_ia32_op_type(fldcw, ia32_AddrModeS);
set_ia32_ls_mode(fldcw, lsmode);
set_ia32_use_frame(fldcw);
- arch_set_irn_register(cg->arch_env, fldcw, &ia32_fp_cw_regs[REG_FPCW]);
+ arch_set_irn_register(fldcw, &ia32_fp_cw_regs[REG_FPCW]);
sched_add_before(before, fldcw);
reload = fldcw;
}
typedef struct collect_fpu_mode_nodes_env_t {
- const arch_env_t *arch_env;
ir_node **state_nodes;
} collect_fpu_mode_nodes_env_t;
-static
-void collect_fpu_mode_nodes_walker(ir_node *node, void *data)
+static void collect_fpu_mode_nodes_walker(ir_node *node, void *data)
{
collect_fpu_mode_nodes_env_t *env = data;
+ const arch_register_t *reg;
- if(!mode_is_data(get_irn_mode(node)))
+ if (!mode_is_data(get_irn_mode(node)))
return;
- const arch_register_t *reg = arch_get_irn_register(env->arch_env, node);
- if(reg == &ia32_fp_cw_regs[REG_FPCW] && !is_ia32_ChangeCW(node)) {
+ reg = arch_get_irn_register(node);
+ if (reg == &ia32_fp_cw_regs[REG_FPCW] && !is_ia32_ChangeCW(node)) {
ARR_APP1(ir_node*, env->state_nodes, node);
}
}
-static
-void rewire_fpu_mode_nodes(be_irg_t *birg)
+static void rewire_fpu_mode_nodes(ir_graph *irg)
{
collect_fpu_mode_nodes_env_t env;
be_ssa_construction_env_t senv;
const arch_register_t *reg = &ia32_fp_cw_regs[REG_FPCW];
- ir_graph *irg = be_get_birg_irg(birg);
ir_node *initial_value;
ir_node **phis;
- be_lv_t *lv = be_get_birg_liveness(birg);
+ be_lv_t *lv = be_get_irg_liveness(irg);
int i, len;
/* do ssa construction for the fpu modes */
- env.arch_env = be_get_birg_arch_env(birg);
env.state_nodes = NEW_ARR_F(ir_node*, 0);
irg_walk_graph(irg, collect_fpu_mode_nodes_walker, NULL, &env);
- initial_value = be_abi_get_ignore_irn(birg->abi, reg);
+ initial_value = be_abi_get_ignore_irn(be_get_irg_abi(irg), reg);
/* nothing needs to be done, in fact we must not continue as for endless
* loops noone is using the initial_value and it will point to a bad node
* now
*/
- if(ARR_LEN(env.state_nodes) == 0) {
+ if (ARR_LEN(env.state_nodes) == 0) {
DEL_ARR_F(env.state_nodes);
return;
}
- be_ssa_construction_init(&senv, birg);
+ be_ssa_construction_init(&senv, irg);
be_ssa_construction_add_copies(&senv, env.state_nodes,
ARR_LEN(env.state_nodes));
be_ssa_construction_fix_users(&senv, initial_value);
- if(lv != NULL) {
+ if (lv != NULL) {
be_ssa_construction_update_liveness_phis(&senv, lv);
be_liveness_update(lv, initial_value);
len = ARR_LEN(env.state_nodes);
- for(i = 0; i < len; ++i) {
+ for (i = 0; i < len; ++i) {
be_liveness_update(lv, env.state_nodes[i]);
}
} else {
- be_liveness_invalidate(birg->lv);
+ be_liveness_invalidate(be_get_irg_liveness(irg));
}
/* set registers for the phis */
phis = be_ssa_construction_get_new_phis(&senv);
len = ARR_LEN(phis);
- for(i = 0; i < len; ++i) {
+ for (i = 0; i < len; ++i) {
ir_node *phi = phis[i];
- be_set_phi_flags(env.arch_env, phi, arch_irn_flags_ignore);
- arch_set_irn_register(env.arch_env, phi, reg);
+ arch_set_irn_register(phi, reg);
}
be_ssa_construction_destroy(&senv);
DEL_ARR_F(env.state_nodes);
- be_liveness_invalidate(be_get_birg_liveness(birg));
+ be_liveness_invalidate(be_get_irg_liveness(irg));
}
-void ia32_setup_fpu_mode(ia32_code_gen_t *cg)
+void ia32_setup_fpu_mode(ir_graph *irg)
{
/* do ssa construction for the fpu modes */
- rewire_fpu_mode_nodes(cg->birg);
+ rewire_fpu_mode_nodes(irg);
/* ensure correct fpu mode for operations */
- be_assure_state(cg->birg, &ia32_fp_cw_regs[REG_FPCW],
- cg, create_fpu_mode_spill, create_fpu_mode_reload);
+ be_assure_state(irg, &ia32_fp_cw_regs[REG_FPCW],
+ NULL, create_fpu_mode_spill, create_fpu_mode_reload);
}
projects / libfirm / blobdiff