Skip to content

R
riscv-gcc-1
Commit 7805417a by Jan Hubicka Committed by Jan Hubicka
Options

predict.c (predict_paths_leading_to, [...]): Add in_loop parameter. 


	* predict.c (predict_paths_leading_to, predict_paths_leading_to_edge):
	Add in_loop parameter.
	(predict_loops): Add loop guard heuristics.
	* predict.def (PRED_LOOP_GUARD): New heuristics.

	* gcc.dg/predict-11.c: New testcase.
	* gfortran.dg/predict-2.f90: New testcase.

From-SVN: r237781
parent 888ed1a3
Showing with 140 additions and 11 deletions
gcc/ChangeLog
2016-06-24 Jan Hubicka <hubicka@ucw.cz>
* predict.c (predict_paths_leading_to, predict_paths_leading_to_edge):
Add in_loop parameter.
(predict_loops): Add loop guard heuristics.
* predict.def (PRED_LOOP_GUARD): New heuristics.
2016-06-24 Jan Hubicka <hubicka@ucw.cz>
* predict.c: Include ipa-utils.h
(tree_bb_level_prediction): Predict recursive calls.
(tree_estimate_probability_bb): Skip inexpensive calls for call
......
gcc/predict.c
......@@ -79,8 +79,12 @@ static sreal real_almost_one, real_br_prob_base,
static void combine_predictions_for_insn (rtx_insn *, basic_block);
static void dump_prediction (FILE *, enum br_predictor, int, basic_block,
enum predictor_reason, edge);
static void predict_paths_leading_to (basic_block, enum br_predictor, enum prediction);
static void predict_paths_leading_to_edge (edge, enum br_predictor, enum prediction);
static void predict_paths_leading_to (basic_block, enum br_predictor,
enum prediction,
struct loop *in_loop = NULL);
static void predict_paths_leading_to_edge (edge, enum br_predictor,
enum prediction,
struct loop *in_loop = NULL);
static bool can_predict_insn_p (const rtx_insn *);
/* Information we hold about each branch predictor.
......@@ -1853,6 +1857,74 @@ predict_loops (void)
tree_to_shwi (loop_bound_step));
}
/* In the following code
for (loop1)
if (cond)
for (loop2)
body;
guess that cond is unlikely. */
if (loop_outer (loop)->num)
{
basic_block bb = NULL;
edge preheader_edge = loop_preheader_edge (loop);
if (single_pred_p (preheader_edge->src)
&& single_succ_p (preheader_edge->src))
preheader_edge = single_pred_edge (preheader_edge->src);
gimple *stmt = last_stmt (preheader_edge->src);
/* Pattern match fortran loop preheader:
_16 = BUILTIN_EXPECT (_15, 1, PRED_FORTRAN_LOOP_PREHEADER);
_17 = (logical(kind=4)) _16;
if (_17 != 0)
goto <bb 11>;
else
goto <bb 13>;
Loop guard branch prediction says nothing about duplicated loop
headers produced by fortran frontend and in this case we want
to predict paths leading to this preheader. */
if (stmt
&& gimple_code (stmt) == GIMPLE_COND
&& gimple_cond_code (stmt) == NE_EXPR
&& TREE_CODE (gimple_cond_lhs (stmt)) == SSA_NAME
&& integer_zerop (gimple_cond_rhs (stmt)))
{
gimple *call_stmt = SSA_NAME_DEF_STMT (gimple_cond_lhs (stmt));
if (gimple_code (call_stmt) == GIMPLE_ASSIGN
&& gimple_expr_code (call_stmt) == NOP_EXPR
&& TREE_CODE (gimple_assign_rhs1 (call_stmt)) == SSA_NAME)
call_stmt = SSA_NAME_DEF_STMT (gimple_assign_rhs1 (call_stmt));
if (gimple_code (call_stmt) == GIMPLE_CALL
&& gimple_call_internal_p (call_stmt)
&& gimple_call_internal_fn (call_stmt) == IFN_BUILTIN_EXPECT
&& TREE_CODE (gimple_call_arg (call_stmt, 2)) == INTEGER_CST
&& tree_fits_uhwi_p (gimple_call_arg (call_stmt, 2))
&& tree_to_uhwi (gimple_call_arg (call_stmt, 2))
== PRED_FORTRAN_LOOP_PREHEADER)
bb = preheader_edge->src;
}
if (!bb)
{
if (!dominated_by_p (CDI_DOMINATORS,
loop_outer (loop)->latch, loop->header))
predict_paths_leading_to_edge (loop_preheader_edge (loop),
PRED_LOOP_GUARD,
NOT_TAKEN,
loop_outer (loop));
}
else
{
if (!dominated_by_p (CDI_DOMINATORS,
loop_outer (loop)->latch, bb))
predict_paths_leading_to (bb,
PRED_LOOP_GUARD,
NOT_TAKEN,
loop_outer (loop));
}
}
/* Free basic blocks from get_loop_body. */
free (bbs);
}
......@@ -2606,12 +2678,19 @@ static void
predict_paths_for_bb (basic_block cur, basic_block bb,
enum br_predictor pred,
enum prediction taken,
bitmap visited)
bitmap visited, struct loop *in_loop = NULL)
{
edge e;
edge_iterator ei;
basic_block son;
/* If we exited the loop or CUR is unconditional in the loop, there is
nothing to do. */
if (in_loop
&& (!flow_bb_inside_loop_p (in_loop, cur)
|| dominated_by_p (CDI_DOMINATORS, in_loop->latch, cur)))
return;
/* We are looking for all edges forming edge cut induced by
set of all blocks postdominated by BB. */
FOR_EACH_EDGE (e, ei, cur->preds)
......@@ -2628,11 +2707,12 @@ predict_paths_for_bb (basic_block cur, basic_block bb,
gcc_assert (bb == cur || dominated_by_p (CDI_POST_DOMINATORS, cur, bb));
/* See if there is an edge from e->src that is not abnormal
and does not lead to BB. */
and does not lead to BB and does not exit the loop. */
FOR_EACH_EDGE (e2, ei2, e->src->succs)
if (e2 != e
&& !(e2->flags & (EDGE_EH | EDGE_FAKE))
&& !dominated_by_p (CDI_POST_DOMINATORS, e2->dest, bb))
&& !dominated_by_p (CDI_POST_DOMINATORS, e2->dest, bb)
&& (!in_loop || !loop_exit_edge_p (in_loop, e2)))
{
found = true;
break;
......@@ -2651,12 +2731,12 @@ predict_paths_for_bb (basic_block cur, basic_block bb,
predict_edge_def (e, pred, taken);
}
else if (bitmap_set_bit (visited, e->src->index))
predict_paths_for_bb (e->src, e->src, pred, taken, visited);
predict_paths_for_bb (e->src, e->src, pred, taken, visited, in_loop);
}
for (son = first_dom_son (CDI_POST_DOMINATORS, cur);
son;
son = next_dom_son (CDI_POST_DOMINATORS, son))
predict_paths_for_bb (son, bb, pred, taken, visited);
predict_paths_for_bb (son, bb, pred, taken, visited, in_loop);
}
/* Sets branch probabilities according to PREDiction and
......@@ -2664,10 +2744,10 @@ predict_paths_for_bb (basic_block cur, basic_block bb,
static void
predict_paths_leading_to (basic_block bb, enum br_predictor pred,
enum prediction taken)
enum prediction taken, struct loop *in_loop)
{
bitmap visited = BITMAP_ALLOC (NULL);
predict_paths_for_bb (bb, bb, pred, taken, visited);
predict_paths_for_bb (bb, bb, pred, taken, visited, in_loop);
BITMAP_FREE (visited);
}
......@@ -2675,7 +2755,7 @@ predict_paths_leading_to (basic_block bb, enum br_predictor pred,
static void
predict_paths_leading_to_edge (edge e, enum br_predictor pred,
enum prediction taken)
enum prediction taken, struct loop *in_loop)
{
bool has_nonloop_edge = false;
edge_iterator ei;
......@@ -2693,7 +2773,7 @@ predict_paths_leading_to_edge (edge e, enum br_predictor pred,
if (!has_nonloop_edge)
{
bitmap visited = BITMAP_ALLOC (NULL);
predict_paths_for_bb (bb, bb, pred, taken, visited);
predict_paths_for_bb (bb, bb, pred, taken, visited, in_loop);
BITMAP_FREE (visited);
}
else
......
gcc/predict.def
......@@ -151,6 +151,14 @@ DEF_PREDICTOR (PRED_LOOP_IV_COMPARE_GUESS, "guess loop iv compare",
DEF_PREDICTOR (PRED_LOOP_IV_COMPARE, "loop iv compare", PROB_VERY_LIKELY,
PRED_FLAG_FIRST_MATCH)
/* In the following code
for (loop1)
if (cond)
for (loop2)
body;
guess that cond is unlikely. */
DEF_PREDICTOR (PRED_LOOP_GUARD, "loop guard", HITRATE (66), 0)
/* Branches to hot labels are likely. */
DEF_PREDICTOR (PRED_HOT_LABEL, "hot label", HITRATE (85), 0)
......
gcc/testsuite/ChangeLog
2016-06-24 Jan Hubicka <hubicka@ucw.cz>
* gcc.dg/predict-11.c: New testcase.
* gfortran.dg/predict-2.f90: New testcase.
2016-06-24 Jan Hubicka <hubicka@ucw.cz>
* gcc.dg/predict-10.c: New test.
2016-06-24 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
......
gcc/testsuite/gcc.dg/predict-11.c 0 → 100644
/* { dg-do compile } */
/* { dg-options "-O2 -fdump-tree-profile_estimate" } */
int *a,n,m;
void test(void);
void
t(void)
{
int i,j;
for (i=0;i<n;i++)
if (a[i])
for (j=0;j<m;j++)
test();
}
/* { dg-final { scan-tree-dump-times "loop guard" 1 "profile_estimate"} } */
gcc/testsuite/gfortran.dg/predict-2.f90 0 → 100644
! { dg-do compile }
! { dg-options "-O2 -fdump-tree-profile_estimate" }
subroutine test(block, array)
integer :: i,j, block(9), array(2)
do i = array(1), array(2)
do j = array(1), array(2)
block(i) = j
end do
end do
end subroutine test
! { dg-final { scan-tree-dump-times "Fortran loop preheader heuristics of edge" 2 "profile_estimate" } }
! { dg-final { scan-tree-dump-times "loop gueard" 0 "profile_estimate" } }
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment