Hi,
In this patch, loop unroll adjust hook is introduced for powerpc. We can do
target related hueristic adjustment in this hook. In this patch, small loops
is unrolled 2 times for O2 and O3 by default. With this patch, we can see
some improvement for spec2017. This patch enhanced a little for [Patch V2] to
enable small loops unroll for O3 by default like O2.
Bootstrapped and regtested on powerpc64le. Is this ok for trunk?
Jiufu
BR.
gcc/
2019-11-04 Jiufu Guo <guoji...@linux.ibm.com>
PR tree-optimization/88760
* config/rs6000/rs6000.c (rs6000_option_override_internal): Remove
code which changes PARAM_MAX_UNROLL_TIMES and PARAM_MAX_UNROLLED_INSNS.
(TARGET_LOOP_UNROLL_ADJUST): Add loop unroll adjust hook.
(rs6000_loop_unroll_adjust): New hook for loop unroll adjust.
Unrolling small loop 2 times for -O2 and -O3.
(rs6000_function_specific_save): Save unroll_small_loops flag.
(rs6000_function_specific_restore): Restore unroll_small_loops flag.
* gcc/config/rs6000/rs6000.opt (unroll_small_loops): New internal flag.
gcc.testsuite/
2019-11-04 Jiufu Guo <guoji...@linux.ibm.com>
PR tree-optimization/88760
* gcc.dg/pr59643.c: Update back to r277550.
---
gcc/config/rs6000/rs6000.c | 38 ++++++++++++++++++++++++++++----------
gcc/config/rs6000/rs6000.opt | 7 +++++++
gcc/testsuite/gcc.dg/pr59643.c | 3 ---
3 files changed, 35 insertions(+), 13 deletions(-)
diff --git a/gcc/config/rs6000/rs6000.c b/gcc/config/rs6000/rs6000.c
index 9ed5151..5e1a75d 100644
--- a/gcc/config/rs6000/rs6000.c
+++ b/gcc/config/rs6000/rs6000.c
@@ -1428,6 +1428,9 @@ static const struct attribute_spec
rs6000_attribute_table[] =
#undef TARGET_VECTORIZE_DESTROY_COST_DATA
#define TARGET_VECTORIZE_DESTROY_COST_DATA rs6000_destroy_cost_data
+#undef TARGET_LOOP_UNROLL_ADJUST
+#define TARGET_LOOP_UNROLL_ADJUST rs6000_loop_unroll_adjust
+
#undef TARGET_INIT_BUILTINS
#define TARGET_INIT_BUILTINS rs6000_init_builtins
#undef TARGET_BUILTIN_DECL
@@ -4540,25 +4543,20 @@ rs6000_option_override_internal (bool global_init_p)
global_options.x_param_values,
global_options_set.x_param_values);
- /* unroll very small loops 2 time if no -funroll-loops. */
+ /* If funroll-loops is not enabled explicitly, then enable small loops
+ unrolling for -O2, and do not turn fweb or frename-registers on. */
if (!global_options_set.x_flag_unroll_loops
&& !global_options_set.x_flag_unroll_all_loops)
{
- maybe_set_param_value (PARAM_MAX_UNROLL_TIMES, 2,
- global_options.x_param_values,
- global_options_set.x_param_values);
-
- maybe_set_param_value (PARAM_MAX_UNROLLED_INSNS, 20,
- global_options.x_param_values,
- global_options_set.x_param_values);
+ unroll_small_loops = optimize >= 2 ? 1 : 0;
- /* If fweb or frename-registers are not specificed in command-line,
- do not turn them on implicitly. */
if (!global_options_set.x_flag_web)
global_options.x_flag_web = 0;
if (!global_options_set.x_flag_rename_registers)
global_options.x_flag_rename_registers = 0;
}
+ else
+ unroll_small_loops = 0;
/* If using typedef char *va_list, signal that
__builtin_va_start (&ap, 0) can be optimized to
@@ -5101,6 +5099,24 @@ rs6000_destroy_cost_data (void *data)
free (data);
}
+/* Implement targetm.loop_unroll_adjust. */
+
+static unsigned
+rs6000_loop_unroll_adjust (unsigned nunroll, struct loop * loop)
+{
+ if (unroll_small_loops)
+ {
+ /* TODO: This is hardcoded to 10 right now. It can be refined, for
+ example we may want to unroll very small loops more times (4 perhaps).
+ We also should use a PARAM for this. */
+ if (loop->ninsns <= 10)
+ return MIN (2, nunroll);
+ else
+ return 0;
+ }
+ return nunroll;
+}
+
/* Handler for the Mathematical Acceleration Subsystem (mass) interface to a
library with vectorized intrinsics. */
@@ -23472,6 +23488,7 @@ rs6000_function_specific_save (struct cl_target_option
*ptr,
{
ptr->x_rs6000_isa_flags = opts->x_rs6000_isa_flags;
ptr->x_rs6000_isa_flags_explicit = opts->x_rs6000_isa_flags_explicit;
+ ptr->x_unroll_small_loops = opts->x_unroll_small_loops;
}
/* Restore the current options */
@@ -23483,6 +23500,7 @@ rs6000_function_specific_restore (struct gcc_options
*opts,
{
opts->x_rs6000_isa_flags = ptr->x_rs6000_isa_flags;
opts->x_rs6000_isa_flags_explicit = ptr->x_rs6000_isa_flags_explicit;
+ opts->x_unroll_small_loops = ptr->x_unroll_small_loops;
(void) rs6000_option_override_internal (false);
}
diff --git a/gcc/config/rs6000/rs6000.opt b/gcc/config/rs6000/rs6000.opt
index 1f37a92..9cd5b4e 100644
--- a/gcc/config/rs6000/rs6000.opt
+++ b/gcc/config/rs6000/rs6000.opt
@@ -96,6 +96,13 @@ enum rs6000_cmodel rs6000_current_cmodel = CMODEL_SMALL
TargetVariable
unsigned int rs6000_recip_control
+;; Whether to unroll small loops only
+Variable
+unsigned char unroll_small_loops
+
+TargetSave
+unsigned char x_unroll_small_loops
+
;; Mask of what builtin functions are allowed
TargetVariable
HOST_WIDE_INT rs6000_builtin_mask
diff --git a/gcc/testsuite/gcc.dg/pr59643.c b/gcc/testsuite/gcc.dg/pr59643.c
index 4446f6e..de78d60 100644
--- a/gcc/testsuite/gcc.dg/pr59643.c
+++ b/gcc/testsuite/gcc.dg/pr59643.c
@@ -1,9 +1,6 @@
/* PR tree-optimization/59643 */
/* { dg-do compile } */
/* { dg-options "-O3 -fdump-tree-pcom-details" } */
-/* { dg-additional-options "--param max-unrolled-insns=400" { target {
powerpc*-*-* } } } */
-/* Implicit threashold of max-unrolled-insn on ppc at O3 is too small for the
- loop of this case. */
void
foo (double *a, double *b, double *c, double d, double e, int n)
--
2.7.4
