From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 1923) id 75FBC3858438; Tue, 27 Feb 2024 13:37:28 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 75FBC3858438 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1709041048; bh=11IT3xrqEoSMxQSlCTaFxUjZUwWRYWiXs8rKNsv0Jeg=; h=From:To:Subject:Date:From; b=QuHRFei3Pqx9BEEvuz8h0V71tl9Y+YBa5pWuUPSTJUZtUItQx4rk9f22hPSpPrdQA NbHKIxebRsKK6wzImrvZSVFvTefaG0G2OvzwJw2qOzAxTQFPH0xqze1h8jcgWZpiQq YMcHeJ22Lw4OOf07f0Cl8QsoGdeSmG9pMFU+LdIY= Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit From: Philipp Tomsich To: gcc-cvs@gcc.gnu.org Subject: [gcc(refs/vendors/vrull/heads/slp-improvements)] tree-optimization: use fewer lanes on VEC_PERM_EXPR for two operators X-Act-Checkin: gcc X-Git-Author: Manolis Tsamis X-Git-Refname: refs/vendors/vrull/heads/slp-improvements X-Git-Oldrev: ed7d62118f587c660491efa06e53ab003eef4a41 X-Git-Newrev: 627830ba3d5fe1e233cc1dd88572fb1a24aed2ef Message-Id: <20240227133728.75FBC3858438@sourceware.org> Date: Tue, 27 Feb 2024 13:37:28 +0000 (GMT) List-Id: https://gcc.gnu.org/g:627830ba3d5fe1e233cc1dd88572fb1a24aed2ef commit 627830ba3d5fe1e233cc1dd88572fb1a24aed2ef Author: Manolis Tsamis Date: Fri Nov 17 17:42:30 2023 +0100 tree-optimization: use fewer lanes on VEC_PERM_EXPR for two operators Currently when SLP nodes are built with "two_operators == true" the VEC_PERM_EXPR that merges the result selects a lane only based on the operator found. In the case that the input nodes have duplicate elements there may be more than one ways to chose. This commit OBtries to use an existing lane if possible, which can free up lanes that can be used in other optimizations. For example, given two vectors with duplicates: A = {a1, a1, a2, a2} B = {b1, b1, b2, b2} a two_operator node with operators +, -, +, - can be built as: RES = VEC_PERM_EXPR(0, 4, 2, 6) and use 2 lanes with this commit. The existing implementation would have built a (0, 5, 2, 7) permutation and have used 4 lanes. This commit adds a case that if the current element can be found in another lane that has been used previously then that lane will be reused. This can happen when the ONE and TWO contain duplicate elements and reduces the number of 'active' lanes. Diff: --- gcc/tree-vect-slp.cc | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc index 238a17ca4e1..c5e9833653d 100644 --- a/gcc/tree-vect-slp.cc +++ b/gcc/tree-vect-slp.cc @@ -2906,7 +2906,25 @@ fail: gassign *ostmt = as_a (ostmt_info->stmt); if (gimple_assign_rhs_code (ostmt) != code0) { - SLP_TREE_LANE_PERMUTATION (node).safe_push (std::make_pair (1, i)); + /* If the current element can be found in another lane that has + been used previously then use that one instead. This can + happen when the ONE and TWO contain duplicate elements and + reduces the number of 'active' lanes. */ + int idx = i; + for (int alt_idx = (int) i - 1; alt_idx >= 0; alt_idx--) + { + gassign *alt_stmt = as_a (stmts[alt_idx]->stmt); + if (gimple_assign_rhs_code (alt_stmt) == code0 + && gimple_assign_rhs1 (ostmt) + == gimple_assign_rhs1 (alt_stmt) + && gimple_assign_rhs2 (ostmt) + == gimple_assign_rhs2 (alt_stmt)) + { + idx = alt_idx; + break; + } + } + SLP_TREE_LANE_PERMUTATION (node).safe_push (std::make_pair (1, idx)); ocode = gimple_assign_rhs_code (ostmt); j = i; }