Makefile.in (tracer.o): New.

	* Makefile.in (tracer.o): New.
	* params.def (TRACER_*): New options.
	* rtl.h (tracer): Declare.
	* timevar.def (TV_TRACER): New.
	* toplev.c (dump_file_index): Add DFI_tracer.
	(dump_file_info): Add tracer.
	(flag_tracer): New.
	(lang_indepdenent_options): Add tracer.
	(rest_of_compilation): Call tracer.
	* tracer.c: New file.
	* invoke.texi (-ftracer): Document.
	(--param tracer-*): Document.

From-SVN: r54154

gcc/ChangeLog

+Sat Jun  1 23:29:51 CEST 2002  Jan Hubicka  <jh@suse.cz>
+
+	* Makefile.in (tracer.o): New.
+	* params.def (TRACER_*): New options.
+	* rtl.h (tracer): Declare.
+	* timevar.def (TV_TRACER): New.
+	* toplev.c (dump_file_index): Add DFI_tracer.
+	(dump_file_info): Add tracer.
+	(flag_tracer): New.
+	(lang_indepdenent_options): Add tracer.
+	(rest_of_compilation): Call tracer.
+	* tracer.c: New file.
+	* invoke.texi (-ftracer): Document.
+	(--param tracer-*): Document.
+
 2002-06-01  Daniel Berlin  <dberlin@dberlin.org>
 
 	* tree-inline.c (expand_call_inline): Make the statement

gcc/Makefile.in

@@ -732,7 +732,7 @@ OBJS = alias.o bb-reorder.o bitmap.o builtins.o caller-save.o calls.o	\
  reload.o reload1.o reorg.o resource.o rtl.o rtlanal.o rtl-error.o	\
  sbitmap.o sched-deps.o sched-ebb.o sched-rgn.o sched-vis.o sdbout.o	\
  sibcall.o simplify-rtx.o ssa.o ssa-ccp.o ssa-dce.o stmt.o	\
- stor-layout.o stringpool.o timevar.o toplev.o tree.o tree-dump.o 	\
+ stor-layout.o stringpool.o timevar.o toplev.o tracer.o tree.o tree-dump.o \
  tree-inline.o unroll.o varasm.o varray.o version.o vmsdbgout.o xcoffout.o \
  $(GGC) $(out_object_file) $(EXTRA_OBJS)
 
@@ -1596,6 +1596,9 @@ predict.o: predict.c $(CONFIG_H) $(SYSTEM_H) $(RTL_H) $(TREE_H) flags.h \
 lists.o: lists.c $(CONFIG_H) $(SYSTEM_H) toplev.h $(RTL_H) $(GGC_H)
 bb-reorder.o : bb-reorder.c $(CONFIG_H) $(SYSTEM_H) $(RTL_H) $(TREE_H) \
    flags.h $(BASIC_BLOCK_H) hard-reg-set.h output.h cfglayout.h $(TARGET_H)
+tracer.o : tracer.c $(CONFIG_H) $(SYSTEM_H) $(RTL_H) $(TREE_H) \
+   $(BASIC_BLOCK_H) hard-reg-set.h output.h cfglayout.h flags.h \
+   $(PARAMS_H) profile.h
 cfglayout.o : cfglayout.c $(CONFIG_H) $(SYSTEM_H) $(RTL_H) $(TREE_H) \
    insn-config.h $(BASIC_BLOCK_H) hard-reg-set.h output.h function.h \
    cfglayout.h

gcc/doc/invoke.texi

@@ -282,7 +282,7 @@ in the following sections.
 -frerun-cse-after-loop  -frerun-loop-opt @gol
 -fschedule-insns  -fschedule-insns2 @gol
 -fsingle-precision-constant  -fssa -fssa-ccp -fssa-dce @gol
--fstrength-reduce  -fstrict-aliasing  -fthread-jumps  -ftrapv @gol
+-fstrength-reduce  -fstrict-aliasing  -ftracer -fthread-jumps  -ftrapv @gol
 -funroll-all-loops  -funroll-loops  @gol
 --param @var{name}=@var{value}
 -O  -O0  -O1  -O2  -O3  -Os}
@@ -3637,6 +3637,12 @@ those which have no call-preserved registers to use instead.
 For all machines, optimization level 2 and higher enables this flag by
 default.
 
+@item -ftracer
+@opindex ftracer
+Perform tail duplication to enlarge superblock size. This transformation
+simplifies the control flow of the function allowing other optimizations to do
+better job.
+
 @item -funroll-loops
 @opindex funroll-loops
 Unroll loops whose number of iterations can be determined at compile
@@ -3941,6 +3947,39 @@ given basic block needs to have to be considered hot.
 @item hot-bb-frequency-fraction
 Select fraction of the maximal frequency of executions of basic block in
 function given basic block needs to have to be considered hot
+
+@item tracer-dynamic-coverage
+@itemx tracer-dynamic-coverage-feedback
+
+This value is used to limit superblock formation once given percentage of
+executed instructions is covered.  This limits unnecesary code size expansion.
+
+The @option{tracer-dynamic-coverage-feedback} is used only when profile
+feedback is available.  The real profiles (as opposed to statically estimated
+ones) are much less balanced allowing the threshold to be larger value.
+
+@item tracer-max-code-growth
+Stop tail duplication once code growth has reached given percentage.  This is
+rather hokey argument, as most of the duplicates will be elliminated later in
+cross jumping, so it may be set to much higher values than is the desired code
+growth.
+
+@item tracer-min-branch-ratio
+
+Stop reverse growth when the reverse probability of best edge is less than this
+threshold (in percent).
+
+@item tracer-min-branch-ratio
+@itemx tracer-min-branch-ratio-feedback
+
+Stop forward growth if the best edge do have probability lower than this
+threshold.
+
+Similary to @option{tracer-dynamic-coverage} two values are present, one for
+compilation for profile feedback and one for compilation without.  The value
+for compilation with profile feedback needs to be more conservative (higher) in
+order to make tracer effective.
+
 @end table
 @end table
 

gcc/params.def

@@ -153,12 +153,43 @@ DEFPARAM(PARAM_MAX_UNROLLED_INSNS,
 
 DEFPARAM(HOT_BB_COUNT_FRACTION,
 	 "hot-bb-count-fraction",
-	 "Select fraction of the maximal count of repetitions of basic block in program given basic block needs to have to be considered hot",
+	 "Select fraction of the maximal count of repetitions of basic block in \
+program given basic block needs to have to be considered hot",
 	 10000)
 DEFPARAM(HOT_BB_FREQUENCY_FRACTION,
 	 "hot-bb-frequency-fraction",
-	 "Select fraction of the maximal frequency of executions of basic block in function given basic block needs to have to be considered hot",
+	 "Select fraction of the maximal frequency of executions of basic \
+block in function given basic block needs to have to be considered hot",
 	 1000)
+DEFPARAM(TRACER_DYNAMIC_COVERAGE_FEEDBACK,
+	 "tracer-dynamic-coverage-feedback",
+	 "The percentage of function, weighted by execution frequency, that \
+must be covered by trace formation. Used when profile feedback is available",
+	 95)
+DEFPARAM(TRACER_DYNAMIC_COVERAGE,
+	 "tracer-dynamic-coverage",
+	 "The percentage of function, weighted by execution frequency, that \
+must be covered by trace formation. Used when profile feedback is not available",
+	 75)
+DEFPARAM(TRACER_MAX_CODE_GROWTH,
+	 "tracer-max-code-growth",
+	 "Maximal code growth caused by tail duplication (in percents)",
+	 100)
+DEFPARAM(TRACER_MIN_BRANCH_RATIO,
+	 "tracer-min-branch-ratio",
+	 "Stop reverse growth if the reverse probability of best edge is less \
+than this threshold (in percents)",
+	 10)
+DEFPARAM(TRACER_MIN_BRANCH_PROBABILITY_FEEDBACK,
+	 "tracer-min-branch-probability-feedback",
+	 "Stop forward growth if the probability of best edge is less than \
+this threshold (in percents). Used when profile feedback is available",
+	 30)
+DEFPARAM(TRACER_MIN_BRANCH_PROBABILITY,
+	 "tracer-min-branch-probability",
+	 "Stop forward growth if the probability of best edge is less than \
+this threshold (in percents). Used when profile feedback is not available",
+	 50)
 /*
 Local variables:
 mode:c

gcc/rtl.h

@@ -2281,4 +2281,6 @@ extern void if_convert			PARAMS ((int));
 /* In predict.c */
 extern void invert_br_probabilities	PARAMS ((rtx));
 extern bool expensive_function_p	PARAMS ((int));
+/* In tracer.c */
+extern void tracer			PARAMS ((void));
 #endif /* ! GCC_RTL_H */

gcc/timevar.def

@@ -58,6 +58,7 @@ DEFTIMEVAR (TV_JUMP                  , "jump")
 DEFTIMEVAR (TV_CSE                   , "CSE")
 DEFTIMEVAR (TV_GCSE                  , "global CSE")
 DEFTIMEVAR (TV_LOOP                  , "loop analysis")
+DEFTIMEVAR (TV_TRACER                , "tracer")
 DEFTIMEVAR (TV_CSE2                  , "CSE 2")
 DEFTIMEVAR (TV_BRANCH_PROB           , "branch prediction")
 DEFTIMEVAR (TV_FLOW                  , "flow analysis")

gcc/toplev.c

@@ -224,6 +224,7 @@ enum dump_file_index
   DFI_loop,
   DFI_cfg,
   DFI_bp,
+  DFI_tracer,
   DFI_cse2,
   DFI_life,
   DFI_combine,
@@ -271,6 +272,7 @@ static struct dump_file_info dump_file[DFI_MAX] =
   { "loop",	'L', 1, 0, 0 },
   { "cfg",	'f', 1, 0, 0 },
   { "bp",	'b', 1, 0, 0 },
+  { "tracer",	'T', 1, 0, 0 },
   { "cse2",	't', 1, 0, 0 },
   { "life",	'f', 1, 0, 0 },	/* Yes, duplicate enable switch.  */
   { "combine",	'c', 1, 0, 0 },
@@ -866,6 +868,10 @@ int flag_merge_constants = 1;
    one, unconditionally renumber instruction UIDs.  */
 int flag_renumber_insns = 1;
 
+/* Nonzero if we perform superblock formation.  */
+
+int flag_tracer = 0;
+
 /* Values of the -falign-* flags: how much to align labels in code.
    0 means `use default', 1 means `don't align'.
    For each variable, there is an _log variant which is the power
@@ -977,6 +983,8 @@ static const lang_independent_options f_options[] =
    N_("When possible do not generate stack frames") },
   {"optimize-sibling-calls", &flag_optimize_sibling_calls, 1,
    N_("Optimize sibling and tail recursive calls") },
+  {"tracer", &flag_tracer, 1,
+   N_("Perform superblock formation via tail duplication") },
   {"cse-follow-jumps", &flag_cse_follow_jumps, 1,
    N_("When running CSE, follow jumps to their targets") },
   {"cse-skip-blocks", &flag_cse_skip_blocks, 1,
@@ -2958,6 +2966,19 @@ rest_of_compilation (decl)
       close_dump_file (DFI_bp, print_rtl_with_bb, insns);
       timevar_pop (TV_BRANCH_PROB);
     }
+  if (flag_tracer)
+    {
+      timevar_push (TV_TRACER);
+      open_dump_file (DFI_tracer, decl);
+      if (rtl_dump_file)
+	dump_flow_info (rtl_dump_file);
+      cleanup_cfg (CLEANUP_EXPENSIVE);
+      tracer ();
+      cleanup_cfg (CLEANUP_EXPENSIVE);
+      close_dump_file (DFI_tracer, print_rtl_with_bb, get_insns ());
+      timevar_pop (TV_TRACER);
+      reg_scan (get_insns (), max_reg_num (), 0);
+    }
 
   if (optimize > 0)
     {