From 5280ddaddf71ed2f15340b7e6be12682294134ea Mon Sep 17 00:00:00 2001 From: Ajit Kumar Agarwal Date: Fri, 4 Sep 2015 18:50:48 +0200 Subject: [PATCH] [Patch,tree-optimization]: Add new path Splitting pass on tree ssa representation. Added a new pass on path splitting on tree SSA representation. The path splitting optimization does the CFG transformation when the two execution paths of the IF-THEN-ELSE merge at the latch node of loop, then duplicate the merge mode into two paths preserving the SSA semantics. ChangeLog: 2015-09-05 Ajit Agarwal * Makefile.in (OBJS): Add tree-ssa-path-split.o * common.opt (ftree-path-split): Add the new flag. * opts.c (default_options_table) : Add an entry for Path splitting optimization at -O2 and above. * passes.def (path_split): Add new path splitting pass. * timevar.def (TV_TREE_PATH_SPLIT): New. * tree-pass.h (make_pass_path_split): New declaration. * tree-ssa-path-split.c: New file. * tracer.c (transform_duplicate): New function. * tracer.h: New header file. * doc/invoke.texi (ftree-path-split): Document. (fdump-tree-path_split): Document. * testsuite/gcc.dg/path-split-1.c: New. Signed-off-by:Ajit Agarwal ajitkum@xilinx.com --- gcc/Makefile.in | 1 + gcc/common.opt | 4 + gcc/doc/invoke.texi | 16 ++- gcc/opts.c | 1 + gcc/passes.def | 1 + gcc/testsuite/gcc.dg/path-split-1.c | 69 ++++++++++ gcc/timevar.def | 1 + gcc/tracer.c | 33 +++-- gcc/tracer.h | 26 ++++ gcc/tree-pass.h | 1 + gcc/tree-ssa-path-split.c | 255 ++++++++++++++++++++++++++++++++++++ 11 files changed, 394 insertions(+), 14 deletions(-) create mode 100644 gcc/testsuite/gcc.dg/path-split-1.c create mode 100644 gcc/tracer.h create mode 100644 gcc/tree-ssa-path-split.c diff --git a/gcc/Makefile.in b/gcc/Makefile.in index 3d1c1e5..7d3abf6 100644 --- a/gcc/Makefile.in +++ b/gcc/Makefile.in @@ -1462,6 +1462,7 @@ OBJS = \ tree-ssa-loop.o \ tree-ssa-math-opts.o \ tree-ssa-operands.o \ + tree-ssa-path-split.o \ tree-ssa-phiopt.o \ tree-ssa-phiprop.o \ tree-ssa-pre.o \ diff --git a/gcc/common.opt b/gcc/common.opt index 94d1d88..da76d74 100644 --- a/gcc/common.opt +++ b/gcc/common.opt @@ -2378,6 +2378,10 @@ ftree-vrp Common Report Var(flag_tree_vrp) Init(0) Optimization Perform Value Range Propagation on trees +ftree-path-split +Common Report Var(flag_tree_path_split) Init(0) Optimization +Perform Path Splitting on trees for loop backedges + funit-at-a-time Common Report Var(flag_unit_at_a_time) Init(1) Compile whole compilation unit at a time diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index d7dc64e..c5e06e2 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -349,6 +349,7 @@ Objective-C and Objective-C++ Dialects}. -fdump-tree-fre@r{[}-@var{n}@r{]} @gol -fdump-tree-vtable-verify @gol -fdump-tree-vrp@r{[}-@var{n}@r{]} @gol +-fdump-tree-path_split@r{[}-@var{n}@r{]} @gol -fdump-tree-storeccp@r{[}-@var{n}@r{]} @gol -fdump-final-insns=@var{file} @gol -fcompare-debug@r{[}=@var{opts}@r{]} -fcompare-debug-second @gol @@ -456,7 +457,7 @@ Objective-C and Objective-C++ Dialects}. -ftree-parallelize-loops=@var{n} -ftree-pre -ftree-partial-pre -ftree-pta @gol -ftree-reassoc -ftree-sink -ftree-slsr -ftree-sra @gol -ftree-switch-conversion -ftree-tail-merge -ftree-ter @gol --ftree-vectorize -ftree-vrp @gol +-ftree-vectorize -ftree-vrp @gol -ftree-path-split @gol -funit-at-a-time -funroll-all-loops -funroll-loops @gol -funsafe-loop-optimizations -funsafe-math-optimizations -funswitch-loops @gol -fipa-ra -fvariable-expansion-in-unroller -fvect-cost-model -fvpt @gol @@ -7154,6 +7155,11 @@ is made by appending @file{.slp} to the source file name. Dump each function after Value Range Propagation (VRP). The file name is made by appending @file{.vrp} to the source file name. +@item path_split +@opindex fdump-tree-path_split +Dump each function after path splitting. The file name is made by +appending @file{.path_split} to the source file name + @item all @opindex fdump-tree-all Enable all the available tree dumps with the flags provided in this option. @@ -7656,6 +7662,7 @@ also turns on the following optimization flags: -ftree-switch-conversion -ftree-tail-merge @gol -ftree-pre @gol -ftree-vrp @gol +-ftree-path-split @gol -fipa-ra} Please note the warning under @option{-fgcse} about @@ -8957,6 +8964,13 @@ enabled by default at @option{-O2} and higher. Null pointer check elimination is only done if @option{-fdelete-null-pointer-checks} is enabled. +@item -ftree-path-split +@opindex ftree-path-split +Perform Path Splitting on trees. When two execution path of the +if-then-else merge at the loop latch node, try to duplicate the +merge node into two paths. This is enabled by default at @option{-O2} +and above. + @item -fsplit-ivs-in-unroller @opindex fsplit-ivs-in-unroller Enables expression of values of induction variables in later iterations diff --git a/gcc/opts.c b/gcc/opts.c index f1a9acd..b75becc 100644 --- a/gcc/opts.c +++ b/gcc/opts.c @@ -506,6 +506,7 @@ static const struct default_options default_options_table[] = { OPT_LEVELS_2_PLUS, OPT_fisolate_erroneous_paths_dereference, NULL, 1 }, { OPT_LEVELS_2_PLUS, OPT_fipa_ra, NULL, 1 }, { OPT_LEVELS_2_PLUS, OPT_flra_remat, NULL, 1 }, + { OPT_LEVELS_2_PLUS, OPT_ftree_path_split, NULL, 1 }, /* -O3 optimizations. */ { OPT_LEVELS_3_PLUS, OPT_ftree_loop_distribute_patterns, NULL, 1 }, diff --git a/gcc/passes.def b/gcc/passes.def index 64fc4d9..536ef32 100644 --- a/gcc/passes.def +++ b/gcc/passes.def @@ -81,6 +81,7 @@ along with GCC; see the file COPYING3. If not see NEXT_PASS (pass_ccp); /* After CCP we rewrite no longer addressed locals into SSA form if possible. */ + NEXT_PASS (pass_path_split); NEXT_PASS (pass_forwprop); NEXT_PASS (pass_sra_early); /* pass_build_ealias is a dummy pass that ensures that we diff --git a/gcc/testsuite/gcc.dg/path-split-1.c b/gcc/testsuite/gcc.dg/path-split-1.c new file mode 100644 index 0000000..af01e49 --- /dev/null +++ b/gcc/testsuite/gcc.dg/path-split-1.c @@ -0,0 +1,69 @@ +/* { dg-do run } */ +/* { dg-options "-O2 -fdump-tree-path_split " } */ + +#include +#include + +#define RGBMAX 255 + +int +test() +{ + int i, Pels; + unsigned char sum = 0; + unsigned char xr, xg, xb; + unsigned char xc, xm, xy, xk; + unsigned char *ReadPtr, *EritePtr; + + ReadPtr = ( unsigned char *) malloc (sizeof (unsigned char) * 100); + EritePtr = ( unsigned char *) malloc (sizeof (unsigned char) * 100); + + for (i = 0; i < 100;i++) + { + ReadPtr[i] = 100 - i; + } + + for (i = 0; i < 100; i++) + { + xr = *ReadPtr++; + xg = *ReadPtr++; + xb = *ReadPtr++; + + xc = (unsigned char) (RGBMAX - xr); + xm = (unsigned char) (RGBMAX - xg); + xy = (unsigned char) (RGBMAX - xb); + + if (xc < xm) + { + xk = (unsigned char) (xc < xy ? xc : xy); + } + else + { + xk = (unsigned char) (xm < xy ? xm : xy); + } + + xc = (unsigned char) (xc - xk); + xm = (unsigned char) (xm - xk); + xy = (unsigned char) (xy - xk); + + *EritePtr++ = xc; + *EritePtr++ = xm; + *EritePtr++ = xy; + *EritePtr++ = xk; + sum += *EritePtr; + } + return sum; +} + +int +main() +{ + if (test() != 33) + abort(); + + return 0; +} + +/* { dg-final { scan-tree-dump "xc_\[0-9\]\[0-9\]* -> { xc_\[0-9\]\[0-9\]* }" "path_split"} } */ +/* { dg-final { scan-tree-dump "xm_\[0-9\]\[0-9\]* -> { xm_\[0-9\]\[0-9\]* }" "path_split"} } */ +/* { dg-final { scan-tree-dump "xy_\[0-9\]\[0-9\]* -> { xy_\[0-9\]\[0-9\]* }" "path_split"} } */ diff --git a/gcc/timevar.def b/gcc/timevar.def index ac41075..e57e2ab 100644 --- a/gcc/timevar.def +++ b/gcc/timevar.def @@ -298,3 +298,4 @@ DEFTIMEVAR (TV_LINK , "link JIT code") DEFTIMEVAR (TV_LOAD , "load JIT result") DEFTIMEVAR (TV_JIT_ACQUIRING_MUTEX , "acquiring JIT mutex") DEFTIMEVAR (TV_JIT_CLIENT_CODE , "JIT client code") +DEFTIMEVAR (TV_TREE_PATH_SPLIT , "tree path split") diff --git a/gcc/tracer.c b/gcc/tracer.c index cad7ab1..fb618d6 100644 --- a/gcc/tracer.c +++ b/gcc/tracer.c @@ -56,9 +56,9 @@ #include "tree-inline.h" #include "cfgloop.h" #include "fibonacci_heap.h" +#include "tracer.h" static int count_insns (basic_block); -static bool ignore_bb_p (const_basic_block); static bool better_p (const_edge, const_edge); static edge find_best_successor (basic_block); static edge find_best_predecessor (basic_block); @@ -90,7 +90,7 @@ bb_seen_p (basic_block bb) } /* Return true if we should ignore the basic block for purposes of tracing. */ -static bool +bool ignore_bb_p (const_basic_block bb) { gimple g; @@ -224,6 +224,22 @@ find_trace (basic_block bb, basic_block *trace) return i; } +basic_block +transform_duplicate (basic_block bb, basic_block bb2) +{ + edge e; + basic_block copy; + + e = find_edge (bb, bb2); + + copy = duplicate_block (bb2, e, bb); + flush_pending_stmts (e); + + add_phi_args_after_copy (©, 1, NULL); + + return (copy); +} + /* Look for basic blocks in frequency order, construct traces and tail duplicate if profitable. */ @@ -319,17 +335,8 @@ tail_duplicate (void) entries or at least rotate the loop. */ && bb2->loop_father->header != bb2) { - edge e; - basic_block copy; - - nduplicated += counts [bb2->index]; - - e = find_edge (bb, bb2); - - copy = duplicate_block (bb2, e, bb); - flush_pending_stmts (e); - - add_phi_args_after_copy (©, 1, NULL); + nduplicated += counts [bb2->index]; + basic_block copy = transform_duplicate (bb, bb2); /* Reconsider the original copy of block we've duplicated. Removing the most common predecessor may make it to be diff --git a/gcc/tracer.h b/gcc/tracer.h new file mode 100644 index 0000000..454d3b7 --- /dev/null +++ b/gcc/tracer.h @@ -0,0 +1,26 @@ +/* Header file for Tracer. + Copyright (C) 2015 Free Software Foundation, Inc. + +This file is part of GCC. + +GCC is free software; you can redistribute it and/or modify it under +the terms of the GNU General Public License as published by the Free +Software Foundation; either version 3, or (at your option) any later +version. + +GCC is distributed in the hope that it will be useful, but WITHOUT ANY +WARRANTY; without even the implied warranty of MERCHANTABILITY or +FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + for more details. + +You should have received a copy of the GNU General Public License +along with GCC; see the file COPYING3. If not see +. */ + +#ifndef GCC_TRACER_H +#define GCC_TRACER_H + +extern basic_block transform_duplicate (basic_block bb, basic_block bb2); +extern bool ignore_bb_p (const_basic_block bb); + +#endif /* GCC_TRaCER_H */ diff --git a/gcc/tree-pass.h b/gcc/tree-pass.h index 7b66a1c..6af7f0d 100644 --- a/gcc/tree-pass.h +++ b/gcc/tree-pass.h @@ -383,6 +383,7 @@ extern gimple_opt_pass *make_pass_tree_loop_done (gcc::context *ctxt); extern gimple_opt_pass *make_pass_ch (gcc::context *ctxt); extern gimple_opt_pass *make_pass_ch_vect (gcc::context *ctxt); extern gimple_opt_pass *make_pass_ccp (gcc::context *ctxt); +extern gimple_opt_pass *make_pass_path_split (gcc::context *ctxt); extern gimple_opt_pass *make_pass_phi_only_cprop (gcc::context *ctxt); extern gimple_opt_pass *make_pass_build_ssa (gcc::context *ctxt); extern gimple_opt_pass *make_pass_build_alias (gcc::context *ctxt); diff --git a/gcc/tree-ssa-path-split.c b/gcc/tree-ssa-path-split.c new file mode 100644 index 0000000..f8fd098 --- /dev/null +++ b/gcc/tree-ssa-path-split.c @@ -0,0 +1,255 @@ +/* Support routines for Path Splitting. + Copyright (C) 2015 Free Software Foundation, Inc. + Contributed by Ajit Kumar Agarwal . + + This file is part of GCC. + + GCC is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 3, or (at your option) + any later version. + +GCC is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with GCC; see the file COPYING3. If not see +. */ + +#include "config.h" +#include "system.h" +#include "coretypes.h" +#include "backend.h" +#include "cfghooks.h" +#include "tree.h" +#include "gimple.h" +#include "rtl.h" +#include "ssa.h" +#include "flags.h" +#include "alias.h" +#include "fold-const.h" +#include "stor-layout.h" +#include "calls.h" +#include "cfganal.h" +#include "internal-fn.h" +#include "gimple-fold.h" +#include "tree-eh.h" +#include "gimple-iterator.h" +#include "gimple-walk.h" +#include "tree-cfg.h" +#include "tree-ssa-loop-manip.h" +#include "tree-ssa-loop-niter.h" +#include "tree-ssa-loop.h" +#include "tree-into-ssa.h" +#include "tree-ssa.h" +#include "tree-pass.h" +#include "tree-dump.h" +#include "cfgloop.h" +#include "tree-scalar-evolution.h" +#include "tree-ssa-propagate.h" +#include "tree-chrec.h" +#include "tracer.h" + +/* Get the latch node and predecessor of the latch node of loop if + the immediate dominator of the latch node of loop is the block + with conditional branch. */ + +static void +find_trace_loop_latch_same_as_join_blk (loop_p loop, basic_block *trace) +{ + edge_iterator ei; + edge e1; + + basic_block latch = loop->latch; + + if (EDGE_COUNT (latch->preds) == 2) + { + basic_block bb + = get_immediate_dominator (CDI_DOMINATORS, latch); + gimple last = gsi_stmt (gsi_last_bb (bb)); + + if (last && gimple_code (last) != GIMPLE_COND) + return ; + + trace[1] = latch; + + FOR_EACH_EDGE (e1, ei, latch->preds) + { + if (!single_succ_p(e1->src) || + !(single_succ_edge (e1->src)->flags & EDGE_FALLTHRU)) + break; + else + { + trace[0] = e1->src; + break; + } + } + } +} + +/* Return false when there is cast assign stmt. Return true + if the number of stmt greater than 1 otherwise false. */ + +static bool +is_feasible_trace (basic_block bb) +{ + int num_stmt = 0; + gimple_stmt_iterator gsi; + + for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi)) + { + gimple stmt = gsi_stmt (gsi); + if (gimple_assign_cast_p (stmt)) + return false; + if (!is_gimple_debug (stmt)) + num_stmt++; + } + + if (num_stmt > 1) + return true; + + return false; +} + +/* If the immediate dominator of the latch of the loop is + block with conditional branch, then the loop latch is + duplicated to its predecessors path preserving the SSA + semantics. + + CFG before transformation. + + : + xk_35 = MIN_EXPR ; + goto ; + + : + xk_36 = MIN_EXPR ; + + : + # xk_4 = PHI + xc_37 = xc_32 - xk_4; + xm_38 = xm_33 - xk_4; + xy_39 = xy_34 - xk_4; + + CFG After Path Splitting transformation + before cleanup phase. + + : + xk_35 = MIN_EXPR ; + + : + # xk_29 = PHI + xc_56 = xc_32 - xk_29; + xm_57 = xm_33 - xk_29; + xy_58 = xy_34 - xk_29; + goto ; + + : + xk_36 = MIN_EXPR ; + + : + # xk_4 = PHI + xc_37 = xc_32 - xk_4; + xm_38 = xm_33 - xk_4; + xy_39 = xy_34 - xk_4; + + : ....... */ + +static bool +perform_path_splitting () +{ + bool changed = false; + basic_block trace[2] = {NULL, NULL}; + loop_p loop; + + loop_optimizer_init (LOOPS_NORMAL | LOOPS_HAVE_RECORDED_EXITS); + initialize_original_copy_tables (); + calculate_dominance_info (CDI_DOMINATORS); + + FOR_EACH_LOOP (loop, LI_FROM_INNERMOST) + { + if (ignore_bb_p (loop->latch)) + continue; + find_trace_loop_latch_same_as_join_blk (loop, trace); + + if (trace[0] && trace[1] && is_feasible_trace (trace[1])) + { + transform_duplicate (trace[0], trace[1]); + trace[0] = NULL; + trace[1] = NULL; + changed = true; + } + } + loop_optimizer_finalize (); + free_original_copy_tables (); + return changed; +} + +static unsigned int +execute_path_split (void) +{ + bool changed; + + if (n_basic_blocks_for_fn (cfun) <= NUM_FIXED_BLOCKS + 1) + return 0; + + if (!(mark_dfs_back_edges ())) + return 0; + + changed = perform_path_splitting(); + + if (changed) + { + free_dominance_info (CDI_DOMINATORS); + /* If we changed the CFG schedule loops for fixup by cleanup_cfg. */ + if (current_loops) + loops_state_set (LOOPS_NEED_FIXUP); + } + + return changed ? TODO_cleanup_cfg : 0; + +} + +static bool +gate_path_split(void) +{ + return flag_tree_path_split != 0; +} + +namespace { + +const pass_data pass_data_path_split = +{ + GIMPLE_PASS, /* type */ + "path_split", /* name */ + OPTGROUP_NONE, /* optinfo_flags */ + TV_TREE_PATH_SPLIT, /* tv_id */ + PROP_ssa, /* properties_required */ + 0, /* properties_provided */ + 0, /* properties_destroyed */ + 0, /* todo_flags_start */ + TODO_update_ssa, /* todo_flags_finish */ +}; + +class pass_path_split : public gimple_opt_pass +{ + public: + pass_path_split (gcc::context *ctxt) + : gimple_opt_pass (pass_data_path_split, ctxt) + {} + /* opt_pass methods: */ + opt_pass * clone () { return new pass_path_split (m_ctxt); } + virtual bool gate (function *) { return gate_path_split (); } + virtual unsigned int execute (function *) { return execute_path_split (); } + +}; // class pass_path_split + +} // anon namespace + +gimple_opt_pass * +make_pass_path_split (gcc::context *ctxt) +{ + return new pass_path_split (ctxt); +} -- 1.8.2.1