From patchwork Thu Sep 29 05:00:51 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: liuhongt X-Patchwork-Id: 1684228 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=8.43.85.97; helo=sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=gcc.gnu.org header.i=@gcc.gnu.org header.a=rsa-sha256 header.s=default header.b=m3o45ahv; dkim-atps=neutral Received: from sourceware.org (ip-8-43-85-97.sourceware.org [8.43.85.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-384) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MdLpx4cjKz1ypH for ; Thu, 29 Sep 2022 15:03:20 +1000 (AEST) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 58074385840C for ; Thu, 29 Sep 2022 05:03:17 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 58074385840C DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1664427797; bh=pBUxyE0chCM+Bij1l15O4i/e8r8dcvBkBitGz1YdSIY=; h=To:Subject:Date:List-Id:List-Unsubscribe:List-Archive:List-Post: List-Help:List-Subscribe:From:Reply-To:From; b=m3o45ahvk9lIJEalgMHiLu0mUC+NTv8KyYRDbCCO1+WN28tyaqx9K6iAWFKNyc8+R 0NLE36TvemfMduysxQ1mumEJGh6FGtBP4nzKuCbbUX0WiGFwNMEpbc9lP/3LS4baN7 KraTICYOXZXk5IUn0/Fu8jR/Yu43I/71fNrIXAWc= X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by sourceware.org (Postfix) with ESMTPS id 5A7553858C83 for ; Thu, 29 Sep 2022 05:02:55 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 5A7553858C83 X-IronPort-AV: E=McAfee;i="6500,9779,10484"; a="388075616" X-IronPort-AV: E=Sophos;i="5.93,354,1654585200"; d="scan'208";a="388075616" Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 28 Sep 2022 22:02:54 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6500,9779,10484"; a="690671219" X-IronPort-AV: E=Sophos;i="5.93,354,1654585200"; d="scan'208";a="690671219" Received: from shvmail03.sh.intel.com ([10.239.245.20]) by fmsmga004.fm.intel.com with ESMTP; 28 Sep 2022 22:02:52 -0700 Received: from shliclel320.sh.intel.com (shliclel320.sh.intel.com [10.239.240.127]) by shvmail03.sh.intel.com (Postfix) with ESMTP id 7941010056AE; Thu, 29 Sep 2022 13:02:51 +0800 (CST) To: gcc-patches@gcc.gnu.org Subject: [PATCH] Check nonlinear iv in vect_can_advance_ivs_p. Date: Thu, 29 Sep 2022 13:00:51 +0800 Message-Id: <20220929050051.30330-1-hongtao.liu@intel.com> X-Mailer: git-send-email 2.18.1 X-Spam-Status: No, score=-12.2 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_SHORT, SPF_HELO_NONE, SPF_NONE, TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: liuhongt via Gcc-patches From: liuhongt Reply-To: liuhongt Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Sender: "Gcc-patches" vectorizable_nonlinear_induction doesn't always guard vect_peel_nonlinear_iv_init when it's called by vect_update_ivs_after_vectorizer which is supposed to be guarded by vect_can_advance_ivs_p. The patch put part codes from vectorizable_nonlinear_induction into a new function vect_can_peel_nonlinear_iv_p, and make vect_can_advcance_ivs_p call it. Also this patch fix a typo in my initial patch as bellow. - if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo) - && induction_type == vect_step_op_mul) if (!INTEGRAL_TYPE_P (TREE_TYPE (vectype))) Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}. Ok for trunk? gcc/ChangeLog: PR tree-optimization/107055 * tree-vect-loop-manip.cc (vect_can_advance_ivs_p): Check for nonlinear induction variables. * tree-vect-loop.cc (vect_can_peel_nonlinear_iv_p): New functions. (vectorizable_nonlinear_induction): Put part codes into vect_can_peel_nonlinear_iv_p. * tree-vectorizer.h (vect_can_peel_nonlinear_iv_p): Declare. gcc/testsuite/ChangeLog: * gcc.target/i386/pr107055.c: New test. --- gcc/testsuite/gcc.target/i386/pr107055.c | 4 ++ gcc/tree-vect-loop-manip.cc | 10 +++ gcc/tree-vect-loop.cc | 82 ++++++++++++++---------- gcc/tree-vectorizer.h | 3 + 4 files changed, 64 insertions(+), 35 deletions(-) create mode 100644 gcc/testsuite/gcc.target/i386/pr107055.c diff --git a/gcc/testsuite/gcc.target/i386/pr107055.c b/gcc/testsuite/gcc.target/i386/pr107055.c new file mode 100644 index 00000000000..63bcb3d742e --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr107055.c @@ -0,0 +1,4 @@ +/* { dg-do compile } */ +/* { dg-options "-O1 -fno-tree-dce -fno-vect-cost-model -ftree-vectorize -fprofile-arcs" } */ + +#include "../../gcc.dg/torture/pr24257.c" diff --git a/gcc/tree-vect-loop-manip.cc b/gcc/tree-vect-loop-manip.cc index 74b221a973c..1d96130c985 100644 --- a/gcc/tree-vect-loop-manip.cc +++ b/gcc/tree-vect-loop-manip.cc @@ -1413,6 +1413,7 @@ vect_can_advance_ivs_p (loop_vec_info loop_vinfo) for (gsi = gsi_start_phis (bb); !gsi_end_p (gsi); gsi_next (&gsi)) { tree evolution_part; + enum vect_induction_op_type induction_type; gphi *phi = gsi.phi (); stmt_vec_info phi_info = loop_vinfo->lookup_stmt (phi); @@ -1432,6 +1433,15 @@ vect_can_advance_ivs_p (loop_vec_info loop_vinfo) continue; } + induction_type = STMT_VINFO_LOOP_PHI_EVOLUTION_TYPE (phi_info); + if (induction_type != vect_step_op_add) + { + if (!vect_can_peel_nonlinear_iv_p (loop_vinfo, induction_type)) + return false; + + continue; + } + /* Analyze the evolution function. */ evolution_part = STMT_VINFO_LOOP_PHI_EVOLUTION_PART (phi_info); diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index aabdc6f2d81..2536cc3cf49 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -8558,6 +8558,50 @@ vect_update_nonlinear_iv (gimple_seq* stmts, tree vectype, return vec_def; } + +/* Return true if vectorizer can peel for nonlinear iv. */ +bool +vect_can_peel_nonlinear_iv_p (loop_vec_info loop_vinfo, + enum vect_induction_op_type induction_type) +{ + tree niters_skip; + /* Init_expr will be update by vect_update_ivs_after_vectorizer, + if niters is unkown: + For shift, when shift mount >= precision, there would be UD. + For mult, don't known how to generate + init_expr * pow (step, niters) for variable niters. + For neg, it should be ok, since niters of vectorized main loop + will always be multiple of 2. */ + if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo) + && induction_type != vect_step_op_neg) + { + if (dump_enabled_p ()) + dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, + "Peeling for epilogue is not supported" + " for nonlinear induction except neg" + " when iteration count is unknown.\n"); + return false; + } + + /* Also doens't support peel for neg when niter is variable. + ??? generate something like niter_expr & 1 ? init_expr : -init_expr? */ + niters_skip = LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo); + if ((niters_skip != NULL_TREE + && TREE_CODE (niters_skip) != INTEGER_CST) + || (!vect_use_loop_mask_for_alignment_p (loop_vinfo) + && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0)) + { + if (dump_enabled_p ()) + dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, + "Peeling for alignement is not supported" + " for nonlinear induction when niters_skip" + " is not constant.\n"); + return false; + } + + return true; +} + /* Function vectorizable_induction Check if STMT_INFO performs an nonlinear induction computation that can be @@ -8628,42 +8672,9 @@ vectorizable_nonlinear_induction (loop_vec_info loop_vinfo, return false; } - /* Init_expr will be update by vect_update_ivs_after_vectorizer, - if niters is unkown: - For shift, when shift mount >= precision, there would be UD. - For mult, don't known how to generate - init_expr * pow (step, niters) for variable niters. - For neg, it should be ok, since niters of vectorized main loop - will always be multiple of 2. */ - if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo) - && induction_type != vect_step_op_neg) - { - if (dump_enabled_p ()) - dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, - "Peeling for epilogue is not supported" - " for nonlinear induction except neg" - " when iteration count is unknown.\n"); - return false; - } - - /* Also doens't support peel for neg when niter is variable. - ??? generate something like niter_expr & 1 ? init_expr : -init_expr? */ - niters_skip = LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo); - if ((niters_skip != NULL_TREE - && TREE_CODE (niters_skip) != INTEGER_CST) - || (!vect_use_loop_mask_for_alignment_p (loop_vinfo) - && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0)) - { - if (dump_enabled_p ()) - dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, - "Peeling for alignement is not supported" - " for nonlinear induction when niters_skip" - " is not constant.\n"); - return false; - } + if (!vect_can_peel_nonlinear_iv_p (loop_vinfo, induction_type)) + return false; - if (!LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo) - && induction_type == vect_step_op_mul) if (!INTEGRAL_TYPE_P (TREE_TYPE (vectype))) { if (dump_enabled_p ()) @@ -8799,6 +8810,7 @@ vectorizable_nonlinear_induction (loop_vec_info loop_vinfo, gimple_seq stmts = NULL; + niters_skip = LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo); /* If we are using the loop mask to "peel" for alignment then we need to adjust the start value here. */ if (niters_skip != NULL_TREE) diff --git a/gcc/tree-vectorizer.h b/gcc/tree-vectorizer.h index 5e75ed1532b..4870c754499 100644 --- a/gcc/tree-vectorizer.h +++ b/gcc/tree-vectorizer.h @@ -2343,6 +2343,9 @@ extern tree cse_and_gimplify_to_preheader (loop_vec_info, tree); /* Nonlinear induction. */ extern tree vect_peel_nonlinear_iv_init (gimple_seq*, tree, tree, tree, enum vect_induction_op_type); +extern bool +vect_can_peel_nonlinear_iv_p (loop_vec_info loop_vinfo, + enum vect_induction_op_type induction_type); /* In tree-vect-slp.cc. */ extern void vect_slp_init (void);