procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fwprop.c: New file.

Browse source 
2006-11-03  Paolo Bonzini  <bonzini@gnu.org>
            Steven Bosscher  <stevenb.gcc@gmail.com>

        * fwprop.c: New file.
        * Makefile.in: Add fwprop.o.
        * tree-pass.h (pass_rtl_fwprop, pass_rtl_fwprop_with_addr): New.
        * passes.c (init_optimization_passes): Schedule forward propagation.
        * rtlanal.c (loc_mentioned_in_p): Support NULL value of the second
        parameter.
        * timevar.def (TV_FWPROP): New.
        * common.opt (-fforward-propagate): New.
        * opts.c (decode_options): Enable forward propagation at -O2.
        * gcse.c (one_cprop_pass): Do not run local cprop unless touching jumps.
        * cse.c (fold_rtx_subreg, fold_rtx_mem, fold_rtx_mem_1, find_best_addr,
        canon_for_address, table_size): Remove.
        (new_basic_block, insert, remove_from_table): Remove references to
        table_size.
        (fold_rtx): Process SUBREGs and MEMs with equiv_constant, make
        simplification loop more straightforward by not calling fold_rtx
        recursively.
        (equiv_constant): Move here a small part of fold_rtx_subreg,
        do not call fold_rtx.  Call avoid_constant_pool_reference
        to process MEMs.
        * recog.c (canonicalize_change_group): New.
        * recog.h (canonicalize_change_group): New.

        * doc/invoke.texi (Optimization Options): Document fwprop.
        * doc/passes.texi (RTL passes): Document fwprop.


Co-Authored-By: Steven Bosscher <stevenb.gcc@gmail.com>

From-SVN: r118475
This commit is contained in:
Paolo Bonzini 2006-11-04 08:36:45 +00:00 • committed by Paolo Bonzini
parent c7cc12b01d
commit a52b023a5f
15 changed files with 1213 additions and 893 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

30
gcc/ChangeLog
View file

@ -1,3 +1,32 @@
2006-11-03 Paolo Bonzini <bonzini@gnu.org>
Steven Bosscher <stevenb.gcc@gmail.com>
* fwprop.c: New file.
* Makefile.in: Add fwprop.o.
* tree-pass.h (pass_rtl_fwprop, pass_rtl_fwprop_with_addr): New.
* passes.c (init_optimization_passes): Schedule forward propagation.
* rtlanal.c (loc_mentioned_in_p): Support NULL value of the second
parameter.
* timevar.def (TV_FWPROP): New.
* common.opt (-fforward-propagate): New.
* opts.c (decode_options): Enable forward propagation at -O2.
* gcse.c (one_cprop_pass): Do not run local cprop unless touching jumps.
* cse.c (fold_rtx_subreg, fold_rtx_mem, fold_rtx_mem_1, find_best_addr,
canon_for_address, table_size): Remove.
(new_basic_block, insert, remove_from_table): Remove references to
table_size.
(fold_rtx): Process SUBREGs and MEMs with equiv_constant, make
simplification loop more straightforward by not calling fold_rtx
recursively.
(equiv_constant): Move here a small part of fold_rtx_subreg,
do not call fold_rtx. Call avoid_constant_pool_reference
to process MEMs.
* recog.c (canonicalize_change_group): New.
* recog.h (canonicalize_change_group): New.
* doc/invoke.texi (Optimization Options): Document fwprop.
* doc/passes.texi (RTL passes): Document fwprop.
2006-11-03 Geoffrey Keating <geoffk@apple.com>
* c-decl.c (WANT_C99_INLINE_SEMANTICS): New, set to 1.
@ -23,7 +52,6 @@
2006-11-03 Paul Brook <paul@codesourcery.com>
gcc/
* config/arm/arm.c (arm_file_start): New function.
(TARGET_ASM_FILE_START): Define.
(arm_default_cpu): New variable.

5
gcc/Makefile.in
View file

@ -997,7 +997,7 @@ OBJS-common = \
debug.o df-core.o df-problems.o df-scan.o dfp.o diagnostic.o dojump.o \
dominance.o loop-doloop.o \
dwarf2asm.o dwarf2out.o emit-rtl.o except.o explow.o loop-iv.o \
expmed.o expr.o final.o flow.o fold-const.o function.o gcse.o \
expmed.o expr.o final.o flow.o fold-const.o function.o fwprop.o gcse.o \
genrtl.o ggc-common.o global.o graph.o gtype-desc.o \
haifa-sched.o hooks.o ifcvt.o insn-attrtab.o insn-emit.o insn-modes.o \
insn-extract.o insn-opinit.o insn-output.o insn-peep.o insn-recog.o \
@ -2336,6 +2336,9 @@ cse.o : cse.c $(CONFIG_H) $(SYSTEM_H) coretypes.h $(TM_H) $(RTL_H) $(REGS_H) \
hard-reg-set.h $(FLAGS_H) insn-config.h $(RECOG_H) $(EXPR_H) toplev.h \
output.h $(FUNCTION_H) $(BASIC_BLOCK_H) $(GGC_H) $(TM_P_H) $(TIMEVAR_H) \
except.h $(TARGET_H) $(PARAMS_H) rtlhooks-def.h tree-pass.h $(REAL_H)
fwprop.o : fwprop.c $(CONFIG_H) $(SYSTEM_H) coretypes.h $(TM_H) $(RTL_H) \
toplev.h insn-config.h $(RECOG_H) $(FLAGS_H) $(OBSTACK_H) $(BASIC_BLOCK_H) \
output.h $(DF_H) alloc-pool.h $(TIMEVAR_H) tree-pass.h
web.o : web.c $(CONFIG_H) $(SYSTEM_H) coretypes.h $(TM_H) $(RTL_H) \
hard-reg-set.h $(FLAGS_H) $(BASIC_BLOCK_H) $(FUNCTION_H) output.h toplev.h \
$(DF_H) $(OBSTACK_H) $(TIMEVAR_H) tree-pass.h

4
gcc/common.opt
View file

@ -444,6 +444,10 @@ fforce-mem
Common Report Var(flag_force_mem)
Copy memory operands into registers before use
fforward-propagate
Common Report Var(flag_forward_propagate)
Perform a forward propagation pass on RTL
; Nonzero means don't put addresses of constant functions in registers.
; Used for compiling the Unix kernel, where strange substitutions are
; done on the assembly output.

971
gcc/cse.c
View file

File diff suppressed because it is too large Load diff

12
gcc/doc/invoke.texi
View file

@ -310,7 +310,7 @@ Objective-C and Objective-C++ Dialects}.
-fcse-skip-blocks -fcx-limited-range -fdata-sections @gol
-fdelayed-branch -fdelete-null-pointer-checks -fearly-inlining @gol
-fexpensive-optimizations -ffast-math -ffloat-store @gol
-fforce-addr -ffunction-sections @gol
-fforce-addr -fforward-propagate -ffunction-sections @gol
-fgcse -fgcse-lm -fgcse-sm -fgcse-las -fgcse-after-reload @gol
-fcrossjumping -fif-conversion -fif-conversion2 @gol
-finline-functions -finline-functions-called-once @gol
@ -4621,6 +4621,16 @@ register-load. This option is now a nop and will be removed in 4.2.
Force memory address constants to be copied into registers before
doing arithmetic on them.
@item -fforward-propagate
@opindex fforward-propagate
Perform a forward propagation pass on RTL. The pass tries to combine two
instructions and checks if the result can be simplified. If loop unrolling
is active, two passes are performed and the second is scheduled after
loop unrolling.
This option is enabled by default at optimization levels @option{-O2},
@option{-O3}, @option{-Os}.
@item -fomit-frame-pointer
@opindex fomit-frame-pointer
Don't keep the frame pointer in a register for functions that

9
gcc/doc/passes.texi
View file

@ -685,6 +685,15 @@ optimization pass''. The bulk of the code for this pass is in
@file{cfgcleanup.c}, and there are support routines in @file{cfgrtl.c}
and @file{jump.c}.
@item Forward propagation of single-def values
This pass attempts to remove redundant computation by substituting
variables that come from a single definition, and
seeing if the result can be simplified. It performs copy propagation
and addressing mode selection. The pass is run twice, with values
being propagated into loops only on the second run. It is located in
@file{fwprop.c}.
@item Common subexpression elimination
This pass removes redundant computation within basic blocks, and

1034
gcc/fwprop.c Normal file
View file

File diff suppressed because it is too large Load diff

3
gcc/gcse.c
View file

@ -3396,7 +3396,8 @@ one_cprop_pass (int pass, bool cprop_jumps, bool bypass_jumps)
global_const_prop_count = local_const_prop_count = 0;
global_copy_prop_count = local_copy_prop_count = 0;
local_cprop_pass (cprop_jumps);
if (cprop_jumps)
local_cprop_pass (cprop_jumps);
/* Determine implicit sets. */
implicit_sets = XCNEWVEC (rtx, last_basic_block);

1
gcc/opts.c
View file

@ -474,6 +474,7 @@ decode_options (unsigned int argc, const char **argv)
flag_thread_jumps = 1;
flag_crossjumping = 1;
flag_optimize_sibling_calls = 1;
flag_forward_propagate = 1;
flag_cse_follow_jumps = 1;
flag_gcse = 1;
flag_expensive_optimizations = 1;

2
gcc/passes.c
View file

@ -635,6 +635,7 @@ init_optimization_passes (void)
NEXT_PASS (pass_instantiate_virtual_regs);
NEXT_PASS (pass_jump2);
NEXT_PASS (pass_cse);
NEXT_PASS (pass_rtl_fwprop);
NEXT_PASS (pass_gcse);
NEXT_PASS (pass_jump_bypass);
NEXT_PASS (pass_rtl_ifcvt);
@ -645,6 +646,7 @@ init_optimization_passes (void)
NEXT_PASS (pass_loop2);
NEXT_PASS (pass_web);
NEXT_PASS (pass_cse2);
NEXT_PASS (pass_rtl_fwprop_addr);
NEXT_PASS (pass_life);
NEXT_PASS (pass_combine);
NEXT_PASS (pass_if_after_combine);

22
gcc/recog.c
View file

@ -238,6 +238,28 @@ validate_change (rtx object, rtx *loc, rtx new, int in_group)
return apply_change_group ();
}
/* Keep X canonicalized if some changes have made it non-canonical; only
modifies the operands of X, not (for example) its code. Simplifications
are not the job of this routine.
Return true if anything was changed. */
bool
canonicalize_change_group (rtx insn, rtx x)
{
if (COMMUTATIVE_P (x)
&& swap_commutative_operands_p (XEXP (x, 0), XEXP (x, 1)))
{
/* Oops, the caller has made X no longer canonical.
Let's redo the changes in the correct order. */
rtx tem = XEXP (x, 0);
validate_change (insn, &XEXP (x, 0), XEXP (x, 1), 1);
validate_change (insn, &XEXP (x, 1), tem, 1);
return true;
}
else
return false;
}
/* This subroutine of apply_change_group verifies whether the changes to INSN
were valid; i.e. whether INSN can still be recognized. */

1
gcc/recog.h
View file

@ -74,6 +74,7 @@ extern void init_recog_no_volatile (void);
extern int check_asm_operands (rtx);
extern int asm_operand_ok (rtx, const char *);
extern int validate_change (rtx, rtx *, rtx, int);
extern bool canonicalize_change_group (rtx insn, rtx x);
extern int insn_invalid_p (rtx);
extern int verify_changes (int);
extern void confirm_change_group (void);

9
gcc/rtlanal.c
View file

@ -2837,10 +2837,15 @@ auto_inc_p (rtx x)
int
loc_mentioned_in_p (rtx *loc, rtx in)
{
enum rtx_code code = GET_CODE (in);
const char *fmt = GET_RTX_FORMAT (code);
enum rtx_code code;
const char *fmt;
int i, j;
if (!in)
return 0;
code = GET_CODE (in);
fmt = GET_RTX_FORMAT (code);
for (i = GET_RTX_LENGTH (code) - 1; i >= 0; i--)
{
if (loc == &in->u.fld[i].rt_rtx)

1
gcc/timevar.def
View file

@ -128,6 +128,7 @@ DEFTIMEVAR (TV_TEMPLATE_INSTANTIATION, "template instantiation")
DEFTIMEVAR (TV_EXPAND , "expand")
DEFTIMEVAR (TV_VARCONST , "varconst")
DEFTIMEVAR (TV_JUMP , "jump")
DEFTIMEVAR (TV_FWPROP , "forward prop")
DEFTIMEVAR (TV_CSE , "CSE")
DEFTIMEVAR (TV_LOOP , "loop analysis")
DEFTIMEVAR (TV_GCSE , "global CSE")

2
gcc/tree-pass.h
View file

@ -330,6 +330,8 @@ extern struct tree_opt_pass pass_rtl_eh;
extern struct tree_opt_pass pass_initial_value_sets;
extern struct tree_opt_pass pass_unshare_all_rtl;
extern struct tree_opt_pass pass_instantiate_virtual_regs;
extern struct tree_opt_pass pass_rtl_fwprop;
extern struct tree_opt_pass pass_rtl_fwprop_addr;
extern struct tree_opt_pass pass_jump2;
extern struct tree_opt_pass pass_cse;
extern struct tree_opt_pass pass_gcse;