public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Aldy Hernandez <aldyh@redhat.com>
To: GCC patches <gcc-patches@gcc.gnu.org>
Subject: [COMMITTED] Use non-null knowledge in path_range_query.
Date: Fri,  3 Sep 2021 15:29:49 +0200	[thread overview]
Message-ID: <20210903132949.729488-1-aldyh@redhat.com> (raw)

[Jeff, I'm CCing you to keep you in the loop on changes that will impact
jump threading.  Since we're using one engine, any changes to the either
ranger or the path solver, is likely to improve jump threading.]

This patch improves ranges for pointers we are interested in a path, by
using the non-null class from the ranger.  This allows us to thread more
paths with minimal effort.

Tested on x86-64 Linux.

gcc/ChangeLog:

	* gimple-range-path.cc (path_range_query::range_defined_in_block):
	Adjust for non-null.
	(path_range_query::adjust_for_non_null_uses): New.
	(path_range_query::precompute_ranges): Call
	adjust_for_non_null_uses.
	* gimple-range-path.h: Add m_non_null and
	adjust_for_non_null_uses.
---
 gcc/gimple-range-path.cc | 33 +++++++++++++++++++++++++++++++++
 gcc/gimple-range-path.h  |  2 ++
 2 files changed, 35 insertions(+)

diff --git a/gcc/gimple-range-path.cc b/gcc/gimple-range-path.cc
index 6d6e5eb6635..db15eb3ff22 100644
--- a/gcc/gimple-range-path.cc
+++ b/gcc/gimple-range-path.cc
@@ -221,6 +221,9 @@ path_range_query::range_defined_in_block (irange &r, tree name, basic_block bb)
   else if (!fold_range (r, def_stmt, this))
     r.set_varying (TREE_TYPE (name));
 
+  if (bb)
+    m_non_null.adjust_range (r, name, bb);
+
   if (DEBUG_SOLVER && (bb || !r.varying_p ()))
     {
       fprintf (dump_file, "range_defined_in_block (BB%d) for ", bb ? bb->index : -1);
@@ -302,6 +305,35 @@ path_range_query::precompute_ranges_in_block (basic_block bb)
     }
 }
 
+// Adjust all pointer imports in BB with non-null information.
+
+void
+path_range_query::adjust_for_non_null_uses (basic_block bb)
+{
+  int_range_max r;
+  bitmap_iterator bi;
+  unsigned i;
+
+  EXECUTE_IF_SET_IN_BITMAP (m_imports, 0, i, bi)
+    {
+      tree name = ssa_name (i);
+
+      if (!POINTER_TYPE_P (TREE_TYPE (name)))
+	continue;
+
+      if (get_cache (r, name))
+	{
+	  if (r.nonzero_p ())
+	    continue;
+	}
+      else
+	r.set_varying (TREE_TYPE (name));
+
+      if (m_non_null.adjust_range (r, name, bb))
+	set_cache (r, name);
+    }
+}
+
 // Precompute the ranges for IMPORTS along PATH.
 //
 // IMPORTS are the set of SSA names, any of which could potentially
@@ -332,6 +364,7 @@ path_range_query::precompute_ranges (const vec<basic_block> &path,
       basic_block bb = curr_bb ();
 
       precompute_ranges_in_block (bb);
+      adjust_for_non_null_uses (bb);
 
       if (at_exit ())
 	break;
diff --git a/gcc/gimple-range-path.h b/gcc/gimple-range-path.h
index 0d2d2e7f75d..51773131040 100644
--- a/gcc/gimple-range-path.h
+++ b/gcc/gimple-range-path.h
@@ -53,6 +53,7 @@ private:
   // Methods to precompute ranges for the given path.
   bool range_defined_in_block (irange &, tree name, basic_block bb);
   void precompute_ranges_in_block (basic_block bb);
+  void adjust_for_non_null_uses (basic_block bb);
   void ssa_range_in_phi (irange &r, gphi *phi);
 
   // Path navigation.
@@ -80,6 +81,7 @@ private:
 
   const bitmap_head *m_imports;
   gimple_ranger &m_ranger;
+  non_null_ref m_non_null;
 };
 
 #endif // GCC_TREE_SSA_THREADSOLVER_H
-- 
2.31.1


                 reply	other threads:[~2021-09-03 13:30 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210903132949.729488-1-aldyh@redhat.com \
    --to=aldyh@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).