[gcc r12-3767] rs6000: Parameterize some const values for density test

[gcc r12-3767] rs6000: Parameterize some const values for density test

[Kewen Lin]

https://gcc.gnu.org/g:144c4984655e9c687dfdda6b1ac39afea8b2a037

commit r12-3767-g144c4984655e9c687dfdda6b1ac39afea8b2a037
Author: Kewen Lin <linkw@linux.ibm.com>
Date:   Tue Sep 21 22:25:54 2021 -0500

    rs6000: Parameterize some const values for density test
    
    This patch follows the discussion here[1], where Segher suggested
    parameterizing those exact magic constants for density heuristics,
    to make it easier to tweak if need.
    
    The change here should be "No Functional Change".  But I verified
    it with SPEC2017 at option sets O2-vect and Ofast-unroll on Power8,
    the result is neutral as expected.
    
    [1]https://gcc.gnu.org/pipermail/gcc-patches/2021-September/579121.html
    
    gcc/ChangeLog:
    
            * config/rs6000/rs6000.opt (rs6000-density-pct-threshold,
            rs6000-density-size-threshold, rs6000-density-penalty,
            rs6000-density-load-pct-threshold,
            rs6000-density-load-num-threshold): New parameter.
            * config/rs6000/rs6000.c (rs6000_density_test): Adjust with
            corresponding parameters.

Diff:
---
 gcc/config/rs6000/rs6000.c   | 22 +++++++---------------
 gcc/config/rs6000/rs6000.opt | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+), 15 deletions(-)

diff --git a/gcc/config/rs6000/rs6000.c b/gcc/config/rs6000/rs6000.c
index 060f51abd89..ad860728169 100644
--- a/gcc/config/rs6000/rs6000.c
+++ b/gcc/config/rs6000/rs6000.c
@@ -5289,9 +5289,6 @@ struct rs6000_cost_data
 static void
 rs6000_density_test (rs6000_cost_data *data)
 {
-  const int DENSITY_PCT_THRESHOLD = 85;
-  const int DENSITY_SIZE_THRESHOLD = 70;
-  const int DENSITY_PENALTY = 10;
   struct loop *loop = data->loop_info;
   basic_block *bbs = get_loop_body (loop);
   int nbbs = loop->num_nodes;
@@ -5327,26 +5324,21 @@ rs6000_density_test (rs6000_cost_data *data)
   free (bbs);
   density_pct = (vec_cost * 100) / (vec_cost + not_vec_cost);
 
-  if (density_pct > DENSITY_PCT_THRESHOLD
-      && vec_cost + not_vec_cost > DENSITY_SIZE_THRESHOLD)
+  if (density_pct > rs6000_density_pct_threshold
+      && vec_cost + not_vec_cost > rs6000_density_size_threshold)
     {
-      data->cost[vect_body] = vec_cost * (100 + DENSITY_PENALTY) / 100;
+      data->cost[vect_body] = vec_cost * (100 + rs6000_density_penalty) / 100;
       if (dump_enabled_p ())
 	dump_printf_loc (MSG_NOTE, vect_location,
 			 "density %d%%, cost %d exceeds threshold, penalizing "
-			 "loop body cost by %d%%\n", density_pct,
-			 vec_cost + not_vec_cost, DENSITY_PENALTY);
+			 "loop body cost by %u%%\n", density_pct,
+			 vec_cost + not_vec_cost, rs6000_density_penalty);
     }
 
   /* Check whether we need to penalize the body cost to account
      for excess strided or elementwise loads.  */
   if (data->extra_ctor_cost > 0)
     {
-      /* Threshold for load stmts percentage in all vectorized stmts.  */
-      const int DENSITY_LOAD_PCT_THRESHOLD = 45;
-      /* Threshold for total number of load stmts.  */
-      const int DENSITY_LOAD_NUM_THRESHOLD = 20;
-
       gcc_assert (data->nloads <= data->nstmts);
       unsigned int load_pct = (data->nloads * 100) / data->nstmts;
 
@@ -5360,8 +5352,8 @@ rs6000_density_test (rs6000_cost_data *data)
 	      the loads.
 	 One typical case is the innermost loop of the hotspot of SPEC2017
 	 503.bwaves_r without loop interchange.  */
-      if (data->nloads > DENSITY_LOAD_NUM_THRESHOLD
-	  && load_pct > DENSITY_LOAD_PCT_THRESHOLD)
+      if (data->nloads > (unsigned int) rs6000_density_load_num_threshold
+	  && load_pct > (unsigned int) rs6000_density_load_pct_threshold)
 	{
 	  data->cost[vect_body] += data->extra_ctor_cost;
 	  if (dump_enabled_p ())
diff --git a/gcc/config/rs6000/rs6000.opt b/gcc/config/rs6000/rs6000.opt
index c1cb9ab06cd..9d7878f144a 100644
--- a/gcc/config/rs6000/rs6000.opt
+++ b/gcc/config/rs6000/rs6000.opt
@@ -639,3 +639,41 @@ Enable instructions that guard against return-oriented programming attacks.
 mprivileged
 Target Var(rs6000_privileged) Init(0)
 Generate code that will run in privileged state.
+
+-param=rs6000-density-pct-threshold=
+Target Undocumented Joined UInteger Var(rs6000_density_pct_threshold) Init(85) IntegerRange(0, 100) Param
+When costing for loop vectorization, we probably need to penalize the loop body
+cost if the existing cost model may not adequately reflect delays from
+unavailable vector resources.  We collect the cost for vectorized statements
+and non-vectorized statements separately, check the proportion of vec_cost to
+total cost of vec_cost and non vec_cost, and penalize only if the proportion
+exceeds the threshold specified by this parameter.  The default value is 85.
+
+-param=rs6000-density-size-threshold=
+Target Undocumented Joined UInteger Var(rs6000_density_size_threshold) Init(70) IntegerRange(0, 1000) Param
+Like parameter rs6000-density-pct-threshold, we also check the total sum of
+vec_cost and non vec_cost, and penalize only if the sum exceeds the threshold
+specified by this parameter.  The default value is 70.
+
+-param=rs6000-density-penalty=
+Target Undocumented Joined UInteger Var(rs6000_density_penalty) Init(10) IntegerRange(0, 1000) Param
+When both heuristics with rs6000-density-pct-threshold and
+rs6000-density-size-threshold are satisfied, we decide to penalize the loop
+body cost by the value which is specified by this parameter.  The default
+value is 10.
+
+-param=rs6000-density-load-pct-threshold=
+Target Undocumented Joined UInteger Var(rs6000_density_load_pct_threshold) Init(45) IntegerRange(0, 100) Param
+When costing for loop vectorization, we probably need to penalize the loop body
+cost by accounting for excess strided or elementwise loads.  We collect the
+numbers for general statements and load statements according to the information
+for statements to be vectorized, check the proportion of load statements, and
+penalize only if the proportion exceeds the threshold specified by this
+parameter.  The default value is 45.
+
+-param=rs6000-density-load-num-threshold=
+Target Undocumented Joined UInteger Var(rs6000_density_load_num_threshold) Init(20) IntegerRange(0, 1000) Param
+Like parameter rs6000-density-load-pct-threshold, we also check if the total
+number of load statements exceeds the threshold specified by this parameter,
+and penalize only if it's satisfied.  The default value is 20.
+