From patchwork Wed Apr 26 07:35:55 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: liuhongt X-Patchwork-Id: 87687 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a59:b0ea:0:b0:3b6:4342:cba0 with SMTP id b10csp65747vqo; Wed, 26 Apr 2023 00:36:47 -0700 (PDT) X-Google-Smtp-Source: AKy350YTxuk0w84zuDTO60PA0aYdwVovAkfSKpAivHH4YsBKazp/dxTO7iazAWdu/k5ZUJ6AJGyU X-Received: by 2002:aa7:d80c:0:b0:506:2c70:3066 with SMTP id v12-20020aa7d80c000000b005062c703066mr17194255edq.21.1682494607629; Wed, 26 Apr 2023 00:36:47 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1682494607; cv=none; d=google.com; s=arc-20160816; b=DOclpT11c5gKwzhD5pZQUSnjga+Y2lFcjF/MmXqAuWHROlXEQxF0U+lYDLcz2Rbmtr 4jguzo4+175yYR7qFVBB9K2KsQKsp1mw1s4aGkmeZqKED9+EGeLTyTvBlTkgvr6l4LRw uZyddAodiRoK9sPBWuTHR+eboej5is4KgSVQRy96xiJUZnGk++eEtUY5R2tueUKiqEdK Q/b3GWPqzNnET1pAVfCYXEgN12Jm45mzcZWwPIALeWN3hwepqigI4XxiTSa9P8TmENxQ vaXgiDi/DhLijP45CLzFiVDThjoARTU9NKkWKIZTCpVlf7OIVBdYjxxtqSmhz3alpBu1 Py/g== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:reply-to:from:list-subscribe:list-help:list-post :list-archive:list-unsubscribe:list-id:precedence :content-transfer-encoding:mime-version:message-id:date:subject:cc :to:dmarc-filter:delivered-to:dkim-signature:dkim-filter; bh=YpDTe18n5OfvVSegzB+7XGT+AmxB+BZnk74K1khL8LE=; b=R92bsqjgLHO8OrU8E+HU+2x49qSxz2ELhRARutobay6y1ixiBuOWAequMC94EJ+WOD QTj60lUgHEVnFv4mqn8WEdR3wPgKeS4HrVWvG+TGoBZjbpQQuNRkZDvDu7A7yZGKKXkB SA6NfjSSEHj0HqbotWe07YNsLJyX106xIw8NUg8N6Qxrl336OKeKaINmXteh234fU+at v/MNBsdBWaEFfQXZvvWxrC1G1DTSrwXddVcGhXZkrd16krCa4HbOeK/QltLzbWvwUB7B 7FPtY5wP3GzkIigPzSUJcbQjhAX4w/rrnS38WMbxqd8rIW1VV0/4yBJcMtZtMV4FwA36 BhcA== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@gcc.gnu.org header.s=default header.b=KgP19OpY; spf=pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) smtp.mailfrom="gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=gnu.org Received: from sourceware.org (server2.sourceware.org. [2620:52:3:1:0:246e:9693:128c]) by mx.google.com with ESMTPS id d24-20020aa7ce18000000b00506bd2b4a70si11907331edv.414.2023.04.26.00.36.47 for (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 26 Apr 2023 00:36:47 -0700 (PDT) Received-SPF: pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) client-ip=2620:52:3:1:0:246e:9693:128c; Authentication-Results: mx.google.com; dkim=pass header.i=@gcc.gnu.org header.s=default header.b=KgP19OpY; spf=pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) smtp.mailfrom="gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=gnu.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 4CA3D3858C39 for ; Wed, 26 Apr 2023 07:36:46 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 4CA3D3858C39 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1682494606; bh=YpDTe18n5OfvVSegzB+7XGT+AmxB+BZnk74K1khL8LE=; h=To:Cc:Subject:Date:List-Id:List-Unsubscribe:List-Archive: List-Post:List-Help:List-Subscribe:From:Reply-To:From; b=KgP19OpY0ixHUtjbXmTy9oznKDObNboyNboyA1hBqA8k/EhDta9e9H99gpGjVOVCg 1Vj+MtukJswbf3GXiJI+qFivpK4zlw8HJwZ80PkhIvoVgLeimG8aCClH4fah76TsWR lkuapJeSMBhmZWYovGfwMxcYgp79ea+kwEwKiN9Q= X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from mga18.intel.com (mga18.intel.com [134.134.136.126]) by sourceware.org (Postfix) with ESMTPS id 60F6E3858CDA for ; Wed, 26 Apr 2023 07:36:00 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 60F6E3858CDA X-IronPort-AV: E=McAfee;i="6600,9927,10691"; a="331251496" X-IronPort-AV: E=Sophos;i="5.99,227,1677571200"; d="scan'208";a="331251496" Received: from fmsmga002.fm.intel.com ([10.253.24.26]) by orsmga106.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Apr 2023 00:36:00 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10691"; a="805383514" X-IronPort-AV: E=Sophos;i="5.99,227,1677571200"; d="scan'208";a="805383514" Received: from shvmail03.sh.intel.com ([10.239.245.20]) by fmsmga002.fm.intel.com with ESMTP; 26 Apr 2023 00:35:56 -0700 Received: from shliclel4217.sh.intel.com (shliclel4217.sh.intel.com [10.239.240.127]) by shvmail03.sh.intel.com (Postfix) with ESMTP id EA1641005188; Wed, 26 Apr 2023 15:35:55 +0800 (CST) To: gcc-patches@gcc.gnu.org Cc: crazylht@gmail.com, hjl.tools@gmail.com Subject: [PATCH] [vect]Enhance NARROW FLOAT_EXPR vectorization by truncating integer to lower precision. Date: Wed, 26 Apr 2023 15:35:55 +0800 Message-Id: <20230426073555.4073530-1-hongtao.liu@intel.com> X-Mailer: git-send-email 2.39.1.388.g2fc9e9ca3c MIME-Version: 1.0 X-Spam-Status: No, score=-12.1 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, 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: 2.1.29 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+ouuuleilei=gmail.com@gcc.gnu.org Sender: "Gcc-patches" X-getmail-retrieved-from-mailbox: =?utf-8?q?INBOX?= X-GMAIL-THRID: =?utf-8?q?1764223465675826514?= X-GMAIL-MSGID: =?utf-8?q?1764223465675826514?= Similar like WIDEN FLOAT_EXPR, when direct_optab is not existed, try intermediate integer type whenever gimple ranger can tell it's safe. .i.e. When there's no direct optab for vector long long -> vector float, but the value range of integer can be represented as int, try vector int -> vector float if availble. Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}. Ok for trunk? gcc/ChangeLog: PR tree-optimization/108804 * tree-vect-patterns.cc (vect_get_range_info): Remove static. * tree-vect-stmts.cc (vect_create_vectorized_demotion_stmts): Add new parameter last_stmt_p. (vectorizable_conversion): Enhance NARROW FLOAT_EXPR vectorization by truncating to lower precision. * tree-vectorizer.h (vect_get_range_info): New declare. gcc/testsuite/ChangeLog: * gcc.target/i386/pr108804.c: New test. --- gcc/testsuite/gcc.target/i386/pr108804.c | 15 ++++ gcc/tree-vect-patterns.cc | 2 +- gcc/tree-vect-stmts.cc | 106 ++++++++++++++++++----- gcc/tree-vectorizer.h | 1 + 4 files changed, 100 insertions(+), 24 deletions(-) create mode 100644 gcc/testsuite/gcc.target/i386/pr108804.c diff --git a/gcc/testsuite/gcc.target/i386/pr108804.c b/gcc/testsuite/gcc.target/i386/pr108804.c new file mode 100644 index 00000000000..2a43c1e1848 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr108804.c @@ -0,0 +1,15 @@ +/* { dg-do compile } */ +/* { dg-options "-mavx2 -Ofast -fdump-tree-vect-details" } */ +/* { dg-final { scan-tree-dump-times "vectorized \[1-3] loops" 1 "vect" } } */ + +typedef unsigned long long uint64_t; +uint64_t d[512]; +float f[1024]; + +void foo() { + for (int i=0; i<512; ++i) { + uint64_t k = d[i]; + f[i]=(k & 0x3F30FFFF); + } +} + diff --git a/gcc/tree-vect-patterns.cc b/gcc/tree-vect-patterns.cc index a49b0953977..dd546b488a4 100644 --- a/gcc/tree-vect-patterns.cc +++ b/gcc/tree-vect-patterns.cc @@ -61,7 +61,7 @@ along with GCC; see the file COPYING3. If not see /* Return true if we have a useful VR_RANGE range for VAR, storing it in *MIN_VALUE and *MAX_VALUE if so. Note the range in the dump files. */ -static bool +bool vect_get_range_info (tree var, wide_int *min_value, wide_int *max_value) { value_range vr; diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index 6b7dbfd4a23..d79a1409d24 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -51,6 +51,7 @@ along with GCC; see the file COPYING3. If not see #include "internal-fn.h" #include "tree-vector-builder.h" #include "vec-perm-indices.h" +#include "gimple-range.h" #include "tree-ssa-loop-niter.h" #include "gimple-fold.h" #include "regs.h" @@ -4799,7 +4800,8 @@ vect_create_vectorized_demotion_stmts (vec_info *vinfo, vec *vec_oprnds, stmt_vec_info stmt_info, vec &vec_dsts, gimple_stmt_iterator *gsi, - slp_tree slp_node, enum tree_code code) + slp_tree slp_node, enum tree_code code, + bool last_stmt_p) { unsigned int i; tree vop0, vop1, new_tmp, vec_dest; @@ -4815,9 +4817,9 @@ vect_create_vectorized_demotion_stmts (vec_info *vinfo, vec *vec_oprnds, new_tmp = make_ssa_name (vec_dest, new_stmt); gimple_assign_set_lhs (new_stmt, new_tmp); vect_finish_stmt_generation (vinfo, stmt_info, new_stmt, gsi); - - if (multi_step_cvt) - /* Store the resulting vector for next recursive call. */ + if (multi_step_cvt || !last_stmt_p) + /* Store the resulting vector for next recursive call, + or return the resulting vector_tmp for NARROW FLOAT_EXPR. */ (*vec_oprnds)[i/2] = new_tmp; else { @@ -4843,7 +4845,8 @@ vect_create_vectorized_demotion_stmts (vec_info *vinfo, vec *vec_oprnds, vect_create_vectorized_demotion_stmts (vinfo, vec_oprnds, multi_step_cvt - 1, stmt_info, vec_dsts, gsi, - slp_node, VEC_PACK_TRUNC_EXPR); + slp_node, VEC_PACK_TRUNC_EXPR, + last_stmt_p); } vec_dsts.quick_push (vec_dest); @@ -5248,22 +5251,53 @@ vectorizable_conversion (vec_info *vinfo, &interm_types)) break; - if (code != FIX_TRUNC_EXPR - || GET_MODE_SIZE (lhs_mode) >= GET_MODE_SIZE (rhs_mode)) + if (GET_MODE_SIZE (lhs_mode) >= GET_MODE_SIZE (rhs_mode)) goto unsupported; - cvt_type - = build_nonstandard_integer_type (GET_MODE_BITSIZE (rhs_mode), 0); - cvt_type = get_same_sized_vectype (cvt_type, vectype_in); - if (cvt_type == NULL_TREE) - goto unsupported; - if (!supportable_convert_operation (code, cvt_type, vectype_in, - &codecvt1)) - goto unsupported; - if (supportable_narrowing_operation (NOP_EXPR, vectype_out, cvt_type, - &code1, &multi_step_cvt, - &interm_types)) - break; + if (code == FIX_TRUNC_EXPR) + { + cvt_type + = build_nonstandard_integer_type (GET_MODE_BITSIZE (rhs_mode), 0); + cvt_type = get_same_sized_vectype (cvt_type, vectype_in); + if (cvt_type == NULL_TREE) + goto unsupported; + if (!supportable_convert_operation (code, cvt_type, vectype_in, + &codecvt1)) + goto unsupported; + if (supportable_narrowing_operation (NOP_EXPR, vectype_out, cvt_type, + &code1, &multi_step_cvt, + &interm_types)) + break; + } + /* If op0 can be represented with low precision integer, + truncate it to cvt_type and the do FLOAT_EXPR. */ + else if (code == FLOAT_EXPR) + { + wide_int op_min_value, op_max_value; + if (!vect_get_range_info (op0, &op_min_value, &op_max_value)) + goto unsupported; + + cvt_type + = build_nonstandard_integer_type (GET_MODE_BITSIZE (lhs_mode), 0); + if (cvt_type == NULL_TREE + || (wi::min_precision (op_max_value, SIGNED) + > TYPE_PRECISION (cvt_type)) + || (wi::min_precision (op_min_value, SIGNED) + > TYPE_PRECISION (cvt_type))) + goto unsupported; + + cvt_type = get_same_sized_vectype (cvt_type, vectype_out); + if (cvt_type == NULL_TREE) + goto unsupported; + if (!supportable_narrowing_operation (NOP_EXPR, cvt_type, vectype_in, + &code1, &multi_step_cvt, + &interm_types)) + goto unsupported; + if (supportable_convert_operation (code, vectype_out, + cvt_type, &codecvt1)) + break; + } + goto unsupported; default: @@ -5335,8 +5369,11 @@ vectorizable_conversion (vec_info *vinfo, from supportable_*_operation, and store them in the correct order for future use in vect_create_vectorized_*_stmts (). */ auto_vec vec_dsts (multi_step_cvt + 1); + bool widen_or_narrow_float_p + = cvt_type && (modifier == WIDEN + || (modifier == NARROW && code == FLOAT_EXPR)); vec_dest = vect_create_destination_var (scalar_dest, - (cvt_type && modifier == WIDEN) + widen_or_narrow_float_p ? cvt_type : vectype_out); vec_dsts.quick_push (vec_dest); @@ -5353,7 +5390,7 @@ vectorizable_conversion (vec_info *vinfo, if (cvt_type) vec_dest = vect_create_destination_var (scalar_dest, - modifier == WIDEN + widen_or_narrow_float_p ? vectype_out : cvt_type); int ninputs = 1; @@ -5456,7 +5493,7 @@ vectorizable_conversion (vec_info *vinfo, vect_get_vec_defs (vinfo, stmt_info, slp_node, ncopies * ninputs, op0, &vec_oprnds0); /* Arguments are ready. Create the new vector stmts. */ - if (cvt_type) + if (cvt_type && code != FLOAT_EXPR) FOR_EACH_VEC_ELT (vec_oprnds0, i, vop0) { gcc_assert (TREE_CODE_LENGTH (codecvt1) == unary_op); @@ -5470,7 +5507,30 @@ vectorizable_conversion (vec_info *vinfo, vect_create_vectorized_demotion_stmts (vinfo, &vec_oprnds0, multi_step_cvt, stmt_info, vec_dsts, gsi, - slp_node, code1); + slp_node, code1, + !cvt_type || code != FLOAT_EXPR); + /* After demoting op0 to cvt_type, convert it to dest. */ + if (cvt_type && code == FLOAT_EXPR) + { + for (unsigned int i = 0; i != vec_oprnds0.length() / 2; i++) + { + /* Arguments are ready, create the new vector stmt. */ + gcc_assert (TREE_CODE_LENGTH (codecvt1) == unary_op); + gassign *new_stmt + = gimple_build_assign (vec_dest, codecvt1, vec_oprnds0[i]); + new_temp = make_ssa_name (vec_dest, new_stmt); + gimple_assign_set_lhs (new_stmt, new_temp); + vect_finish_stmt_generation (vinfo, stmt_info, new_stmt, gsi); + + /* This is the last step of the conversion sequence. Store the + vectors in SLP_NODE or in vector info of the scalar statement + (or in STMT_VINFO_RELATED_STMT chain). */ + if (slp_node) + SLP_TREE_VEC_STMTS (slp_node).quick_push (new_stmt); + else + STMT_VINFO_VEC_STMTS (stmt_info).safe_push (new_stmt); + } + } break; } if (!slp_node) diff --git a/gcc/tree-vectorizer.h b/gcc/tree-vectorizer.h index 9cf2fb23fe3..27fa2306a5e 100644 --- a/gcc/tree-vectorizer.h +++ b/gcc/tree-vectorizer.h @@ -2384,6 +2384,7 @@ extern bool compatible_calls_p (gcall *, gcall *); /* In tree-vect-patterns.cc. */ extern void vect_mark_pattern_stmts (vec_info *, stmt_vec_info, gimple *, tree); +extern bool vect_get_range_info (tree, wide_int*, wide_int*); /* Pattern recognition functions. Additional pattern recognition functions can (and will) be added