From patchwork Wed Jun 5 06:22:45 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Richard Biener X-Patchwork-Id: 1943655 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=KcueHOao; dkim=pass header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=Pa19foLt; dkim=pass (1024-bit key) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=KcueHOao; dkim=neutral header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=Pa19foLt; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=2620:52:3:1:0:246e:9693:128c; helo=server2.sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=patchwork.ozlabs.org) Received: from server2.sourceware.org (server2.sourceware.org [IPv6:2620:52:3:1:0:246e:9693:128c]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4VvHTJ09nxz20KL for ; Wed, 5 Jun 2024 16:23:15 +1000 (AEST) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 4FB333BF90FF for ; Wed, 5 Jun 2024 06:23:14 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from smtp-out2.suse.de (smtp-out2.suse.de [IPv6:2a07:de40:b251:101:10:150:64:2]) by sourceware.org (Postfix) with ESMTPS id 11ED63806BE4 for ; Wed, 5 Jun 2024 06:22:51 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 11ED63806BE4 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=suse.de Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=suse.de ARC-Filter: OpenARC Filter v1.0.0 sourceware.org 11ED63806BE4 Authentication-Results: server2.sourceware.org; arc=none smtp.remote-ip=2a07:de40:b251:101:10:150:64:2 ARC-Seal: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568573; cv=none; b=g7HuglPnA0kR7LvXA9fqBQ3VgQxYNyvBLd3pdRigodtdg6FeyYO7MpfpKP/qrJWTMaTQCBDObayNDAnqRHPhEas/0+/gM5PQWAuz4l1BSbEEO4c7Dwqcieh4DOBULj2xMvJst9mmoMaOIQG35MMNq3BfBxUoZeUNECX3mnpb9/4= ARC-Message-Signature: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568573; c=relaxed/simple; bh=SWWTkMa2pSNf7Wa9BuwX5dWbSH5D81OBEUzLwQU/FMI=; h=DKIM-Signature:DKIM-Signature:DKIM-Signature:DKIM-Signature:Date: From:To:Subject:MIME-Version; b=tMkqoT157u9Bjb75FBuX2qQNRj4goAeWaHj0FfAUcqDSgrE7d6/sAeZDFbzXvc+p8rXkgylkXEDA57ZceK7bjuo7DBHuH822qZR12BwXVG4HzOQ3Z/vp37fGQQJJGbdWf3XANgTMAf3ZdnbAr0wufX+00D7/T/88yAf3TN+oxNY= ARC-Authentication-Results: i=1; server2.sourceware.org Received: from murzim.nue2.suse.org (unknown [10.168.4.243]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id 7A32C1F7BA for ; Wed, 5 Jun 2024 06:22:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568569; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fQYMrCXQU3pl9ChzSfCJJpvIO3Uzviv+pcgNi7j9BP0=; b=KcueHOaorUEs7gbLW/biBl78aPflZUHdNu0sT2KFvB9+ogR5YnQc+GNNLaj+d+npnH/fCB 5ApsPQb1IXewlMFREe3e7uRj3H/10+gfqG0n1iTX/2d0X60D1Z6bXq6t+LE7mEoGiTQcNs EAHisBbO3gHuauh/HY3ENCi85d8YsbU= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568569; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fQYMrCXQU3pl9ChzSfCJJpvIO3Uzviv+pcgNi7j9BP0=; b=Pa19foLtBOWvdVGY3kBc/K1X2CHN2577PAAWoK4O07M2j/rX2x2Sj4U2CftK+LqS1VRCsJ CRbiQq4n3mymPJCQ== Authentication-Results: smtp-out2.suse.de; none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568569; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fQYMrCXQU3pl9ChzSfCJJpvIO3Uzviv+pcgNi7j9BP0=; b=KcueHOaorUEs7gbLW/biBl78aPflZUHdNu0sT2KFvB9+ogR5YnQc+GNNLaj+d+npnH/fCB 5ApsPQb1IXewlMFREe3e7uRj3H/10+gfqG0n1iTX/2d0X60D1Z6bXq6t+LE7mEoGiTQcNs EAHisBbO3gHuauh/HY3ENCi85d8YsbU= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568569; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fQYMrCXQU3pl9ChzSfCJJpvIO3Uzviv+pcgNi7j9BP0=; b=Pa19foLtBOWvdVGY3kBc/K1X2CHN2577PAAWoK4O07M2j/rX2x2Sj4U2CftK+LqS1VRCsJ CRbiQq4n3mymPJCQ== Date: Wed, 5 Jun 2024 08:22:45 +0200 (CEST) From: Richard Biener To: gcc-patches@gcc.gnu.org Subject: [PATCH 1/4] Relax COND_EXPR reduction vectorization SLP restriction MIME-Version: 1.0 X-Spam-Score: 0.74 X-Spam-Level: X-Spamd-Result: default: False [0.74 / 50.00]; BAYES_HAM(-3.00)[99.99%]; MISSING_MID(2.50)[]; NEURAL_SPAM_SHORT(2.19)[0.730]; NEURAL_HAM_LONG(-0.85)[-0.850]; MIME_GOOD(-0.10)[text/plain]; ARC_NA(0.00)[]; RCPT_COUNT_ONE(0.00)[1]; MISSING_XM_UA(0.00)[]; RCVD_COUNT_ZERO(0.00)[0]; FROM_HAS_DN(0.00)[]; FUZZY_BLOCKED(0.00)[rspamd.com]; TO_MATCH_ENVRCPT_ALL(0.00)[]; DKIM_SIGNED(0.00)[suse.de:s=susede2_rsa,suse.de:s=susede2_ed25519]; FROM_EQ_ENVFROM(0.00)[]; TO_DN_NONE(0.00)[]; MIME_TRACE(0.00)[0:+] X-Spam-Status: No, score=-10.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, MISSING_MID, SPF_HELO_NONE, SPF_PASS, TXREP, T_SCC_BODY_TEXT_LINE 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: 2.1.30 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Message-Id: <20240605062314.4FB333BF90FF@sourceware.org> Allow one-lane SLP but for the case where we need to swap the arms. * tree-vect-stmts.cc (vectorizable_condition): Allow single-lane SLP, but not when we need to swap then and else clause. --- gcc/tree-vect-stmts.cc | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index b26cc74f417..c82381e799e 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -12116,7 +12116,7 @@ vectorizable_condition (vec_info *vinfo, = STMT_VINFO_REDUC_DEF (vect_orig_stmt (stmt_info)) != NULL; if (for_reduction) { - if (slp_node) + if (slp_node && SLP_TREE_LANES (slp_node) > 1) return false; reduc_info = info_for_reduction (vinfo, stmt_info); reduction_type = STMT_VINFO_REDUC_TYPE (reduc_info); @@ -12205,6 +12205,10 @@ vectorizable_condition (vec_info *vinfo, cond_expr = NULL_TREE; } } + /* ??? The vectorized operand query below doesn't allow swapping + this way for SLP. */ + if (slp_node) + return false; std::swap (then_clause, else_clause); } From patchwork Wed Jun 5 06:22:57 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Richard Biener X-Patchwork-Id: 1943656 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=Bv4cLsQr; dkim=pass header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=hlkuqFWG; dkim=pass (1024-bit key) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=Bv4cLsQr; dkim=neutral header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=hlkuqFWG; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=2620:52:3:1:0:246e:9693:128c; helo=server2.sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=patchwork.ozlabs.org) Received: from server2.sourceware.org (server2.sourceware.org [IPv6:2620:52:3:1:0:246e:9693:128c]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4VvHTY2s4kz20KL for ; Wed, 5 Jun 2024 16:23:29 +1000 (AEST) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id ADF9F38F14D8 for ; Wed, 5 Jun 2024 06:23:27 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2a07:de40:b251:101:10:150:64:1]) by sourceware.org (Postfix) with ESMTPS id 053ED38CCBCA for ; Wed, 5 Jun 2024 06:22:59 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 053ED38CCBCA Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=suse.de Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=suse.de ARC-Filter: OpenARC Filter v1.0.0 sourceware.org 053ED38CCBCA Authentication-Results: server2.sourceware.org; arc=none smtp.remote-ip=2a07:de40:b251:101:10:150:64:1 ARC-Seal: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568581; cv=none; b=Jb7y1f19oAsipGe5g6cCFywdfgLcfn//ls2UTnUNHRsFjNvkbaF3Gp1bnBROlARdrvGxSiABwhFyvfvVgy3jPrtyCV63HfsFqQSBYLdGGqHoKgZgIESnWJhFmdDgRrBjUFIHrhH+tt3WSFL6m4UW1by4RgGyoj6IU9LQCYQweuA= ARC-Message-Signature: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568581; c=relaxed/simple; bh=Wa3Ceo7P9U88nRkRIJr+EHINX2hPhqT0qBwkGDmZ9mU=; h=DKIM-Signature:DKIM-Signature:DKIM-Signature:DKIM-Signature:Date: From:To:Subject:MIME-Version; b=dW2ZZFGCl5C1aJa3qSWMduUNjcfuF1IvxVXHeDJyphQhPnz+OViZZmDnDxI+GFzpgCq4OUDjIj0ucFU4zfAw7KWbyUVxg9wv2F2yifX9buwGTIxUbs1oeSiP0n0bP12S3zGyxUvnWU6ZyGsICrqSmYDPsyG2buSYyK6x9M+Z6IE= ARC-Authentication-Results: i=1; server2.sourceware.org Received: from murzim.nue2.suse.org (unknown [10.168.4.243]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id 04F2921A31 for ; Wed, 5 Jun 2024 06:22:58 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568578; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fh8blZRen59vAnTKRkabG8TsIQsSMw1Yv8hdif2Oq48=; b=Bv4cLsQrVszOm1hxekmsHL/0LyTdJkhSpnhbZvJWfC2Q56BRkPVTNkBe1VTyduHeMgr92M SUqE6/h6Vplkcy8hWelVy1lys4AsgstLHzOEDmuhZas31MAo+mQFBk6XsTTebjY5kJfQLv c17pYSEjFNU4vQRlm3zjwnfn9tNc+28= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568578; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fh8blZRen59vAnTKRkabG8TsIQsSMw1Yv8hdif2Oq48=; b=hlkuqFWGBJLiMQSh1/Sc4Ivv6idnJm5l/JRgnZWGutpWeK/8L6SIS1Hexrsdh+aQSvqYfI y/5cC5emTVn9JNDw== Authentication-Results: smtp-out1.suse.de; none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568578; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fh8blZRen59vAnTKRkabG8TsIQsSMw1Yv8hdif2Oq48=; b=Bv4cLsQrVszOm1hxekmsHL/0LyTdJkhSpnhbZvJWfC2Q56BRkPVTNkBe1VTyduHeMgr92M SUqE6/h6Vplkcy8hWelVy1lys4AsgstLHzOEDmuhZas31MAo+mQFBk6XsTTebjY5kJfQLv c17pYSEjFNU4vQRlm3zjwnfn9tNc+28= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568578; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=fh8blZRen59vAnTKRkabG8TsIQsSMw1Yv8hdif2Oq48=; b=hlkuqFWGBJLiMQSh1/Sc4Ivv6idnJm5l/JRgnZWGutpWeK/8L6SIS1Hexrsdh+aQSvqYfI y/5cC5emTVn9JNDw== Date: Wed, 5 Jun 2024 08:22:57 +0200 (CEST) From: Richard Biener To: gcc-patches@gcc.gnu.org Subject: [PATCH 2/4] Allow single-lane COND_REDUCTION vectorization MIME-Version: 1.0 X-Spam-Level: X-Spamd-Result: default: False [0.60 / 50.00]; BAYES_HAM(-3.00)[99.99%]; MISSING_MID(2.50)[]; NEURAL_SPAM_SHORT(2.08)[0.694]; NEURAL_HAM_LONG(-0.88)[-0.879]; MIME_GOOD(-0.10)[text/plain]; ARC_NA(0.00)[]; RCPT_COUNT_ONE(0.00)[1]; MISSING_XM_UA(0.00)[]; RCVD_COUNT_ZERO(0.00)[0]; FROM_HAS_DN(0.00)[]; FUZZY_BLOCKED(0.00)[rspamd.com]; TO_MATCH_ENVRCPT_ALL(0.00)[]; DKIM_SIGNED(0.00)[suse.de:s=susede2_rsa,suse.de:s=susede2_ed25519]; FROM_EQ_ENVFROM(0.00)[]; TO_DN_NONE(0.00)[]; MIME_TRACE(0.00)[0:+] X-Spam-Score: 0.60 X-Spam-Status: No, score=-10.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, MISSING_MID, SPF_HELO_NONE, SPF_PASS, TXREP, T_SCC_BODY_TEXT_LINE 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: 2.1.30 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Message-Id: <20240605062327.ADF9F38F14D8@sourceware.org> The following enables single-lane COND_REDUCTION vectorization. * tree-vect-loop.cc (vect_create_epilog_for_reduction): Adjust for single-lane COND_REDUCTION SLP vectorization. (vectorizable_reduction): Likewise. (vect_transform_cycle_phi): Likewise. --- gcc/tree-vect-loop.cc | 97 ++++++++++++++++++++++++++++++++++++------- 1 file changed, 81 insertions(+), 16 deletions(-) diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index 06292ed8bbe..ccd6acef5c5 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -6030,7 +6030,13 @@ vect_create_epilog_for_reduction (loop_vec_info loop_vinfo, tree induc_val = NULL_TREE; tree adjustment_def = NULL; if (slp_node) - ; + { + /* Optimize: for induction condition reduction, if we can't use zero + for induc_val, use initial_def. */ + if (STMT_VINFO_REDUC_TYPE (reduc_info) == INTEGER_INDUC_COND_REDUCTION) + induc_val = STMT_VINFO_VEC_INDUC_COND_INITIAL_VAL (reduc_info); + /* ??? Coverage for double_reduc and 'else' isn't clear. */ + } else { /* Optimize: for induction condition reduction, if we can't use zero @@ -6075,23 +6081,46 @@ vect_create_epilog_for_reduction (loop_vec_info loop_vinfo, if (STMT_VINFO_REDUC_TYPE (reduc_info) == COND_REDUCTION) { auto_vec, 2> ccompares; - stmt_vec_info cond_info = STMT_VINFO_REDUC_DEF (reduc_info); - cond_info = vect_stmt_to_vectorize (cond_info); - while (cond_info != reduc_info) + if (slp_node) { - if (gimple_assign_rhs_code (cond_info->stmt) == COND_EXPR) + slp_tree cond_node = slp_node_instance->root; + while (cond_node != slp_node_instance->reduc_phis) { - gimple *vec_stmt = STMT_VINFO_VEC_STMTS (cond_info)[0]; - gcc_assert (gimple_assign_rhs_code (vec_stmt) == VEC_COND_EXPR); - ccompares.safe_push - (std::make_pair (unshare_expr (gimple_assign_rhs1 (vec_stmt)), - STMT_VINFO_REDUC_IDX (cond_info) == 2)); + stmt_vec_info cond_info = SLP_TREE_REPRESENTATIVE (cond_node); + if (gimple_assign_rhs_code (cond_info->stmt) == COND_EXPR) + { + gimple *vec_stmt + = SSA_NAME_DEF_STMT (SLP_TREE_VEC_DEFS (cond_node)[0]); + gcc_assert (gimple_assign_rhs_code (vec_stmt) == VEC_COND_EXPR); + ccompares.safe_push + (std::make_pair (gimple_assign_rhs1 (vec_stmt), + STMT_VINFO_REDUC_IDX (cond_info) == 2)); + } + /* ??? We probably want to have REDUC_IDX on the SLP node? */ + cond_node = SLP_TREE_CHILDREN + (cond_node)[STMT_VINFO_REDUC_IDX (cond_info)]; } - cond_info - = loop_vinfo->lookup_def (gimple_op (cond_info->stmt, - 1 + STMT_VINFO_REDUC_IDX - (cond_info))); + } + else + { + stmt_vec_info cond_info = STMT_VINFO_REDUC_DEF (reduc_info); cond_info = vect_stmt_to_vectorize (cond_info); + while (cond_info != reduc_info) + { + if (gimple_assign_rhs_code (cond_info->stmt) == COND_EXPR) + { + gimple *vec_stmt = STMT_VINFO_VEC_STMTS (cond_info)[0]; + gcc_assert (gimple_assign_rhs_code (vec_stmt) == VEC_COND_EXPR); + ccompares.safe_push + (std::make_pair (gimple_assign_rhs1 (vec_stmt), + STMT_VINFO_REDUC_IDX (cond_info) == 2)); + } + cond_info + = loop_vinfo->lookup_def (gimple_op (cond_info->stmt, + 1 + STMT_VINFO_REDUC_IDX + (cond_info))); + cond_info = vect_stmt_to_vectorize (cond_info); + } } gcc_assert (ccompares.length () != 0); @@ -7844,7 +7873,7 @@ vectorizable_reduction (loop_vec_info loop_vinfo, /* If we have a condition reduction, see if we can simplify it further. */ if (v_reduc_type == COND_REDUCTION) { - if (slp_node) + if (slp_node && SLP_TREE_LANES (slp_node) != 1) return false; /* When the condition uses the reduction value in the condition, fail. */ @@ -8050,6 +8079,18 @@ vectorizable_reduction (loop_vec_info loop_vinfo, } } + if ((reduction_type == COND_REDUCTION + || reduction_type == INTEGER_INDUC_COND_REDUCTION + || reduction_type == CONST_COND_REDUCTION) + && slp_node + && SLP_TREE_NUMBER_OF_VEC_STMTS (slp_node) > 1) + { + if (dump_enabled_p ()) + dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, + "multiple types in condition reduction reduction.\n"); + return false; + } + if ((double_reduc || reduction_type != TREE_CODE_REDUCTION) && ncopies > 1) { @@ -8765,7 +8806,31 @@ vect_transform_cycle_phi (loop_vec_info loop_vinfo, if (slp_node) { vec_initial_defs.reserve (vec_num); - if (nested_cycle) + /* Optimize: if initial_def is for REDUC_MAX smaller than the base + and we can't use zero for induc_val, use initial_def. Similarly + for REDUC_MIN and initial_def larger than the base. */ + if (STMT_VINFO_REDUC_TYPE (reduc_info) == INTEGER_INDUC_COND_REDUCTION) + { + gcc_assert (SLP_TREE_LANES (slp_node) == 1); + tree initial_def = vect_phi_initial_value (phi); + reduc_info->reduc_initial_values.safe_push (initial_def); + tree induc_val = STMT_VINFO_VEC_INDUC_COND_INITIAL_VAL (reduc_info); + if (TREE_CODE (initial_def) == INTEGER_CST + && !integer_zerop (induc_val) + && ((STMT_VINFO_REDUC_CODE (reduc_info) == MAX_EXPR + && tree_int_cst_lt (initial_def, induc_val)) + || (STMT_VINFO_REDUC_CODE (reduc_info) == MIN_EXPR + && tree_int_cst_lt (induc_val, initial_def)))) + { + induc_val = initial_def; + /* Communicate we used the initial_def to epilouge + generation. */ + STMT_VINFO_VEC_INDUC_COND_INITIAL_VAL (reduc_info) = NULL_TREE; + } + vec_initial_defs.quick_push + (build_vector_from_val (vectype_out, induc_val)); + } + else if (nested_cycle) { unsigned phi_idx = loop_preheader_edge (loop)->dest_idx; vect_get_slp_defs (SLP_TREE_CHILDREN (slp_node)[phi_idx], From patchwork Wed Jun 5 06:23:05 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Richard Biener X-Patchwork-Id: 1943657 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=PGuTnVaq; dkim=pass header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=7RZ19TfC; dkim=pass (1024-bit key) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=PGuTnVaq; dkim=neutral header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=7RZ19TfC; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=2620:52:3:1:0:246e:9693:128c; helo=server2.sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=patchwork.ozlabs.org) Received: from server2.sourceware.org (server2.sourceware.org [IPv6:2620:52:3:1:0:246e:9693:128c]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4VvHTg0mbjz20KL for ; Wed, 5 Jun 2024 16:23:35 +1000 (AEST) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 6302838F14FE for ; Wed, 5 Jun 2024 06:23:33 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from smtp-out1.suse.de (smtp-out1.suse.de [195.135.223.130]) by sourceware.org (Postfix) with ESMTPS id 9B03338CCBC5 for ; Wed, 5 Jun 2024 06:23:06 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 9B03338CCBC5 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=suse.de Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=suse.de ARC-Filter: OpenARC Filter v1.0.0 sourceware.org 9B03338CCBC5 Authentication-Results: server2.sourceware.org; arc=none smtp.remote-ip=195.135.223.130 ARC-Seal: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568588; cv=none; b=XdD3bnMpzcxvUwKIq2b0zAwn1+4jnAaEeXR8EjfSByIk843HAUG/mOMYHfhK5/SF9A/vmlQ9mHOlsuhvH709OxIWtjWjXzyj2J8D/OvkVt4Nb0PihzXYz9rayFAroDpm7jQt6vJNHdqZLiD0bCjs4s8n7I7oVs6y9StOXcdkJHM= ARC-Message-Signature: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568588; c=relaxed/simple; bh=3uTafRSbDF3L9Rspfiiwdtp9mXCijcxngLFEgcnv+z8=; h=DKIM-Signature:DKIM-Signature:DKIM-Signature:DKIM-Signature:Date: From:To:Subject:MIME-Version; b=rarjm1hYg8HfQf7wy3s4gLFrHR2Ua/Lgis+bw+ZboZ02tOIpJmNIBUB/rijWa/aNbThAo6egmgZ9pIUcoK8Rb8Q2Y+7TgLWPFR3d3b5qJG53xyXYKhSVlhADDhs0hYQqlWi6qP8yuiPv4bdceCrcu7XTya71EMbmwsa27oJHn0U= ARC-Authentication-Results: i=1; server2.sourceware.org Received: from murzim.nue2.suse.org (unknown [10.168.4.243]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id 961C021A31 for ; Wed, 5 Jun 2024 06:23:05 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568585; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=9J30zRPu40kzpV+2TfdVoZ7WkRoFm3qIkWHJaZ5jCJQ=; b=PGuTnVaqyqssO+1Or6zB7yacpQQA0NRMZmh0xdOYDbNDp+QKFiolMpHlvTQCKd+nM5L0GC 5cVoII8UqKOz6XgiVtvORvBNAk5/AxVcgftZvEIfvyWYGDbnjXJa2pfU6CP/8HuvsLjsPw yPz4KnFf8lMApQKdbc+UUfUS+4cBUzs= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568585; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=9J30zRPu40kzpV+2TfdVoZ7WkRoFm3qIkWHJaZ5jCJQ=; b=7RZ19TfCxo/fO7OnxmikSp8ADs9BuAD66oijyvH1ULm2QByonFTQgIe+2+Gue9GQKpDz0A 3DMkrF2DvugN3/DQ== Authentication-Results: smtp-out1.suse.de; none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568585; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=9J30zRPu40kzpV+2TfdVoZ7WkRoFm3qIkWHJaZ5jCJQ=; b=PGuTnVaqyqssO+1Or6zB7yacpQQA0NRMZmh0xdOYDbNDp+QKFiolMpHlvTQCKd+nM5L0GC 5cVoII8UqKOz6XgiVtvORvBNAk5/AxVcgftZvEIfvyWYGDbnjXJa2pfU6CP/8HuvsLjsPw yPz4KnFf8lMApQKdbc+UUfUS+4cBUzs= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568585; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=9J30zRPu40kzpV+2TfdVoZ7WkRoFm3qIkWHJaZ5jCJQ=; b=7RZ19TfCxo/fO7OnxmikSp8ADs9BuAD66oijyvH1ULm2QByonFTQgIe+2+Gue9GQKpDz0A 3DMkrF2DvugN3/DQ== Date: Wed, 5 Jun 2024 08:23:05 +0200 (CEST) From: Richard Biener To: gcc-patches@gcc.gnu.org Subject: [PATCH 3/4] Add double reduction support for SLP vectorization MIME-Version: 1.0 X-Spam-Level: X-Spamd-Result: default: False [0.57 / 50.00]; BAYES_HAM(-3.00)[100.00%]; MISSING_MID(2.50)[]; NEURAL_SPAM_SHORT(2.06)[0.685]; NEURAL_HAM_LONG(-0.89)[-0.886]; MIME_GOOD(-0.10)[text/plain]; ARC_NA(0.00)[]; RCPT_COUNT_ONE(0.00)[1]; MISSING_XM_UA(0.00)[]; RCVD_COUNT_ZERO(0.00)[0]; FROM_HAS_DN(0.00)[]; FUZZY_BLOCKED(0.00)[rspamd.com]; TO_MATCH_ENVRCPT_ALL(0.00)[]; DKIM_SIGNED(0.00)[suse.de:s=susede2_rsa,suse.de:s=susede2_ed25519]; FROM_EQ_ENVFROM(0.00)[]; TO_DN_NONE(0.00)[]; MIME_TRACE(0.00)[0:+] X-Spam-Score: 0.57 X-Spam-Status: No, score=-10.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, MISSING_MID, SPF_HELO_NONE, SPF_PASS, TXREP, T_SCC_BODY_TEXT_LINE 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: 2.1.30 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Message-Id: <20240605062333.6302838F14FE@sourceware.org> The following makes double reduction vectorization work when using (single-lane) SLP vectorization. * tree-vect-loop.cc (vect_analyze_scalar_cycles_1): Queue double reductions in LOOP_VINFO_REDUCTIONS. (vect_create_epilog_for_reduction): Remove asserts disabling SLP for double reductions. (vectorizable_reduction): Analyze SLP double reductions only once and start off the correct places. * tree-vect-slp.cc (vect_get_and_check_slp_defs): Allow vect_double_reduction_def. (vect_build_slp_tree_2): Fix condition for the ignored reduction initial values. * tree-vect-stmts.cc (vect_analyze_stmt): Allow vect_double_reduction_def. --- gcc/tree-vect-loop.cc | 35 +++++++++++++++++++++++++---------- gcc/tree-vect-slp.cc | 3 ++- gcc/tree-vect-stmts.cc | 4 ++++ 3 files changed, 31 insertions(+), 11 deletions(-) diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index ccd6acef5c5..b9e8e9b5559 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -685,6 +685,8 @@ vect_analyze_scalar_cycles_1 (loop_vec_info loop_vinfo, class loop *loop, STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_double_reduction_def; STMT_VINFO_DEF_TYPE (reduc_stmt_info) = vect_double_reduction_def; + /* Make it accessible for SLP vectorization. */ + LOOP_VINFO_REDUCTIONS (loop_vinfo).safe_push (reduc_stmt_info); } else { @@ -5975,7 +5977,6 @@ vect_create_epilog_for_reduction (loop_vec_info loop_vinfo, stmt_vec_info rdef_info = stmt_info; if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_double_reduction_def) { - gcc_assert (!slp_node); double_reduc = true; stmt_info = loop_vinfo->lookup_def (gimple_phi_arg_def (stmt_info->stmt, 0)); @@ -6020,7 +6021,7 @@ vect_create_epilog_for_reduction (loop_vec_info loop_vinfo, { outer_loop = loop; loop = loop->inner; - gcc_assert (!slp_node && double_reduc); + gcc_assert (double_reduc); } vectype = STMT_VINFO_REDUC_VECTYPE (reduc_info); @@ -6035,7 +6036,7 @@ vect_create_epilog_for_reduction (loop_vec_info loop_vinfo, for induc_val, use initial_def. */ if (STMT_VINFO_REDUC_TYPE (reduc_info) == INTEGER_INDUC_COND_REDUCTION) induc_val = STMT_VINFO_VEC_INDUC_COND_INITIAL_VAL (reduc_info); - /* ??? Coverage for double_reduc and 'else' isn't clear. */ + /* ??? Coverage for 'else' isn't clear. */ } else { @@ -7605,15 +7606,16 @@ vectorizable_reduction (loop_vec_info loop_vinfo, STMT_VINFO_TYPE (stmt_info) = reduc_vec_info_type; return true; } - if (slp_node) - { - slp_node_instance->reduc_phis = slp_node; - /* ??? We're leaving slp_node to point to the PHIs, we only - need it to get at the number of vector stmts which wasn't - yet initialized for the instance root. */ - } if (STMT_VINFO_DEF_TYPE (stmt_info) == vect_double_reduction_def) { + if (gimple_bb (stmt_info->stmt) != loop->header) + { + /* For SLP we arrive here for both the inner loop LC PHI and + the outer loop PHI. The latter is what we want to analyze + the reduction with. */ + gcc_assert (slp_node); + return true; + } use_operand_p use_p; gimple *use_stmt; bool res = single_imm_use (gimple_phi_result (stmt_info->stmt), @@ -7622,6 +7624,14 @@ vectorizable_reduction (loop_vec_info loop_vinfo, phi_info = loop_vinfo->lookup_stmt (use_stmt); } + if (slp_node) + { + slp_node_instance->reduc_phis = slp_node; + /* ??? We're leaving slp_node to point to the PHIs, we only + need it to get at the number of vector stmts which wasn't + yet initialized for the instance root. */ + } + /* PHIs should not participate in patterns. */ gcc_assert (!STMT_VINFO_RELATED_STMT (phi_info)); gphi *reduc_def_phi = as_a (phi_info->stmt); @@ -7637,6 +7647,11 @@ vectorizable_reduction (loop_vec_info loop_vinfo, bool only_slp_reduc_chain = true; stmt_info = NULL; slp_tree slp_for_stmt_info = slp_node ? slp_node_instance->root : NULL; + /* For double-reductions we start SLP analysis at the inner loop LC PHI + which is the def of the outer loop live stmt. */ + if (STMT_VINFO_DEF_TYPE (reduc_info) == vect_double_reduction_def + && slp_node) + slp_for_stmt_info = SLP_TREE_CHILDREN (slp_for_stmt_info)[0]; while (reduc_def != PHI_RESULT (reduc_def_phi)) { stmt_vec_info def = loop_vinfo->lookup_def (reduc_def); diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc index ba1190c7155..7e3d0107b4e 100644 --- a/gcc/tree-vect-slp.cc +++ b/gcc/tree-vect-slp.cc @@ -778,6 +778,7 @@ vect_get_and_check_slp_defs (vec_info *vinfo, unsigned char swap, case vect_constant_def: case vect_internal_def: case vect_reduction_def: + case vect_double_reduction_def: case vect_induction_def: case vect_nested_cycle: case vect_first_order_recurrence: @@ -1906,7 +1907,7 @@ vect_build_slp_tree_2 (vec_info *vinfo, slp_tree node, class loop *loop = LOOP_VINFO_LOOP (loop_vinfo); /* Reduction initial values are not explicitely represented. */ if (def_type != vect_first_order_recurrence - && !nested_in_vect_loop_p (loop, stmt_info)) + && gimple_bb (stmt_info->stmt) == loop->header) skip_args[loop_preheader_edge (loop)->dest_idx] = true; /* Reduction chain backedge defs are filled manually. ??? Need a better way to identify a SLP reduction chain PHI. diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index c82381e799e..5098b7fab6a 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -13260,6 +13260,10 @@ vect_analyze_stmt (vec_info *vinfo, || relevance == vect_used_only_live)); break; + case vect_double_reduction_def: + gcc_assert (!bb_vinfo && node); + break; + case vect_induction_def: case vect_first_order_recurrence: gcc_assert (!bb_vinfo); From patchwork Wed Jun 5 06:23:15 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Richard Biener X-Patchwork-Id: 1943658 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=nUiuvAKR; dkim=pass header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=cMPbllNS; dkim=pass (1024-bit key) header.d=suse.de header.i=@suse.de header.a=rsa-sha256 header.s=susede2_rsa header.b=q+v6CHt6; dkim=neutral header.d=suse.de header.i=@suse.de header.a=ed25519-sha256 header.s=susede2_ed25519 header.b=J07SUROF; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=2620:52:3:1:0:246e:9693:128c; helo=server2.sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=patchwork.ozlabs.org) Received: from server2.sourceware.org (server2.sourceware.org [IPv6:2620:52:3:1:0:246e:9693:128c]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4VvHTs3Xnqz20KL for ; Wed, 5 Jun 2024 16:23:45 +1000 (AEST) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id C7F8138CCBF2 for ; Wed, 5 Jun 2024 06:23:43 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2a07:de40:b251:101:10:150:64:1]) by sourceware.org (Postfix) with ESMTPS id DCA0838CCBC2 for ; Wed, 5 Jun 2024 06:23:17 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org DCA0838CCBC2 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=suse.de Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=suse.de ARC-Filter: OpenARC Filter v1.0.0 sourceware.org DCA0838CCBC2 Authentication-Results: server2.sourceware.org; arc=none smtp.remote-ip=2a07:de40:b251:101:10:150:64:1 ARC-Seal: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568600; cv=none; b=ZrQqHyt+Fd1X7vVxZaFGrgfLF/v3+pOP/R4WWpqwLLABIbto+xq4DHxq1aCsZd54hOynsoja7GunQyMCuFN5mPwBmebBpmlsUZhN3zGl76xrvsT1le1Nwz2cWzVg2b4V2xrDHG3O4sRUMmHNAdXbabGKjpIXONiilNeFEqofrCE= ARC-Message-Signature: i=1; a=rsa-sha256; d=sourceware.org; s=key; t=1717568600; c=relaxed/simple; bh=oL3WauKs3cetcJ7+VMwk4KRaYBrHq4T6YfhZNrHec1Y=; h=DKIM-Signature:DKIM-Signature:DKIM-Signature:DKIM-Signature:Date: From:To:Subject:MIME-Version; b=sj0A62IYImLY37IinQyQflBT76PJed5l8n/9qsQwmEx933wITjih+4tCcd73fSDxqiinRjait8iYPOcmjhvJt6DJg+gL0fegso+HEfUX6phikjYm0l1ZOf1WZKPb3/X6iHCzDDBx+UhfE+bhQNihaHw8WTaNgvjxWEBQ0t9HLcw= ARC-Authentication-Results: i=1; server2.sourceware.org Received: from murzim.nue2.suse.org (unknown [10.168.4.243]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id 8C36221A31 for ; Wed, 5 Jun 2024 06:23:15 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568596; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=z9u8eZjmQ1J/82lhN+Az763lWE4OO3lkYFFdmgFTsaI=; b=nUiuvAKRkTxOpen20MiNkGO72gwh9uIGAXRV6RI68SHJZMh7o3AJI54B1HOd9hcjVzmiaH dOUfY3sBAWnmH20Z2ZdQkHio7RQHJIB5ch5yjcjYa1G7cbQdVJkU3XABf57AJud0AkOLnh FcS6UVQfEJme3UOiqVzZJBB4ofEGb2s= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568596; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=z9u8eZjmQ1J/82lhN+Az763lWE4OO3lkYFFdmgFTsaI=; b=cMPbllNSblFnScELXgNntZHt30k4L95tKFzlpif/yg9kBVahyZ9CAbnCiwwtaoM/jEdrKg R1o8WdFPvZQ3WEDQ== Authentication-Results: smtp-out1.suse.de; none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1717568595; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=z9u8eZjmQ1J/82lhN+Az763lWE4OO3lkYFFdmgFTsaI=; b=q+v6CHt6FJ+l/AsBgH2TjWbpvkYibm/Y5EqBIE4mY5hR4doIM9mUO17uJ8Yd20MdXCGbBk T7fGldRf4jm/YuyemYIsowpYJr1Qoyp1JV/S+DXDbVaxq5DkR0hq5TlAP+peWyVPpe3A+S rwZIesi3vPttiTCJ/HKX5m7os8NeayY= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1717568595; h=from:from:reply-to:date:date:to:to:cc:mime-version:mime-version: content-type:content-type; bh=z9u8eZjmQ1J/82lhN+Az763lWE4OO3lkYFFdmgFTsaI=; b=J07SUROFuozV1Q9aKNmbqg9bJfaUwj9So4YM3z3hdTSQiXHZYYKe7xuBoDSVmHCb9IARM6 hz1cbqH3+P3QKSBw== Date: Wed, 5 Jun 2024 08:23:15 +0200 (CEST) From: Richard Biener To: gcc-patches@gcc.gnu.org Subject: [PATCH 4/4] RISC-V: Allow single-lane SLP in-order reductions MIME-Version: 1.0 X-Spam-Level: X-Spamd-Result: default: False [0.57 / 50.00]; BAYES_HAM(-3.00)[100.00%]; MISSING_MID(2.50)[]; NEURAL_SPAM_SHORT(2.06)[0.687]; NEURAL_HAM_LONG(-0.89)[-0.886]; MIME_GOOD(-0.10)[text/plain]; ARC_NA(0.00)[]; FUZZY_BLOCKED(0.00)[rspamd.com]; RCPT_COUNT_ONE(0.00)[1]; DKIM_SIGNED(0.00)[suse.de:s=susede2_rsa,suse.de:s=susede2_ed25519]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FROM_HAS_DN(0.00)[]; MISSING_XM_UA(0.00)[]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+]; TO_DN_NONE(0.00)[]; RCVD_COUNT_ZERO(0.00)[0]; DBL_BLOCKED_OPENRESOLVER(0.00)[gcc.target:url,murzim.nue2.suse.org:helo] X-Spam-Score: 0.57 X-Spam-Status: No, score=-10.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_SHORT, MISSING_MID, SPF_HELO_NONE, SPF_PASS, TXREP, T_SCC_BODY_TEXT_LINE 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: 2.1.30 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Message-Id: <20240605062343.C7F8138CCBF2@sourceware.org> The single-lane case isn't different from non-SLP, no re-association implied. But the transform stage cannot handle a conditional reduction op which isn't checked during analysis - this makes it work, exercised with a single-lane non-reduction-chain by gcc.target/i386/pr112464.c * tree-vect-loop.cc (vectorizable_reduction): Allow single-lane SLP in-order reductions. (vectorize_fold_left_reduction): Handle SLP reduction with conditional reduction op. --- gcc/tree-vect-loop.cc | 48 +++++++++++++++++-------------------------- 1 file changed, 19 insertions(+), 29 deletions(-) diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index b9e8e9b5559..ceb92156b58 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -7139,56 +7139,46 @@ vectorize_fold_left_reduction (loop_vec_info loop_vinfo, gcc_assert (TREE_CODE_LENGTH (tree_code (code)) == binary_op); if (slp_node) - { - if (is_cond_op) - { - if (dump_enabled_p ()) - dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, - "fold-left reduction on SLP not supported.\n"); - return false; - } - - gcc_assert (known_eq (TYPE_VECTOR_SUBPARTS (vectype_out), - TYPE_VECTOR_SUBPARTS (vectype_in))); - } + gcc_assert (known_eq (TYPE_VECTOR_SUBPARTS (vectype_out), + TYPE_VECTOR_SUBPARTS (vectype_in))); /* The operands either come from a binary operation or an IFN_COND operation. The former is a gimple assign with binary rhs and the latter is a gimple call with four arguments. */ gcc_assert (num_ops == 2 || num_ops == 4); - tree op0, opmask; - if (!is_cond_op) - op0 = ops[1 - reduc_index]; - else - { - op0 = ops[2 + (1 - reduc_index)]; - opmask = ops[0]; - gcc_assert (!slp_node); - } int group_size = 1; stmt_vec_info scalar_dest_def_info; auto_vec vec_oprnds0, vec_opmask; if (slp_node) { - auto_vec > vec_defs (2); - vect_get_slp_defs (loop_vinfo, slp_node, &vec_defs); - vec_oprnds0.safe_splice (vec_defs[1 - reduc_index]); - vec_defs[0].release (); - vec_defs[1].release (); + vect_get_slp_defs (SLP_TREE_CHILDREN (slp_node)[(is_cond_op ? 2 : 0) + + (1 - reduc_index)], + &vec_oprnds0); group_size = SLP_TREE_SCALAR_STMTS (slp_node).length (); scalar_dest_def_info = SLP_TREE_SCALAR_STMTS (slp_node)[group_size - 1]; + /* For an IFN_COND_OP we also need the vector mask operand. */ + if (is_cond_op) + vect_get_slp_defs (SLP_TREE_CHILDREN (slp_node)[0], &vec_opmask); } else { + tree op0, opmask; + if (!is_cond_op) + op0 = ops[1 - reduc_index]; + else + { + op0 = ops[2 + (1 - reduc_index)]; + opmask = ops[0]; + } vect_get_vec_defs_for_operand (loop_vinfo, stmt_info, 1, op0, &vec_oprnds0); scalar_dest_def_info = stmt_info; /* For an IFN_COND_OP we also need the vector mask operand. */ if (is_cond_op) - vect_get_vec_defs_for_operand (loop_vinfo, stmt_info, 1, - opmask, &vec_opmask); + vect_get_vec_defs_for_operand (loop_vinfo, stmt_info, 1, + opmask, &vec_opmask); } gimple *sdef = vect_orig_stmt (scalar_dest_def_info)->stmt; @@ -8210,7 +8200,7 @@ vectorizable_reduction (loop_vec_info loop_vinfo, } if (reduction_type == FOLD_LEFT_REDUCTION - && slp_node + && (slp_node && SLP_TREE_LANES (slp_node) > 1) && !REDUC_GROUP_FIRST_ELEMENT (stmt_info)) { /* We cannot use in-order reductions in this case because there is