From nobody Mon Sep 16 19:08:55 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org ARC-Seal: i=1; a=rsa-sha256; t=1721231357; cv=none; d=zohomail.com; s=zohoarc; b=nifH3w4oEfqQuQqQ06VHFu9Abd5U3loQ8fiCxZ4bwWekiPg9kbXcXnFACb2k1og3cpsHBxcXQsbrdCNLTj7+KDp3+vGK9J/J1r7xzgJtnAIxAjjYT5s/BevQpVd6hpBDEQEvy/e+Bgk5s8POTN3k0XI5e00CEUCSRZYN12xynbo= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1721231357; h=Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=dBJVgoGbVEBWCKNsYWowDSG5FlAgb+PmFPWTiOzZRq8=; b=YZkJP6OeZHP67hyUNTUruk4iht8szsVREBd88OKH5SFy33bYfLD2hlc0vuxqYOaAl+zw8p8NohuR1mmtxKu3I1mM3u8n41UOf/AqsnquSXJPEQwuvNLPN65Wn1BImd4b67wzTuZgGmUvkY64isXbneT+YcOlVD3xM1WLjaASlt4= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 172123135790478.00689220663196; Wed, 17 Jul 2024 08:49:17 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1sU6sp-0006Wd-Dm; Wed, 17 Jul 2024 11:47:59 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1sU6fm-0004d1-Gk for qemu-devel@nongnu.org; Wed, 17 Jul 2024 11:34:31 -0400 Received: from mail-wr1-x434.google.com ([2a00:1450:4864:20::434]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1sU6fk-0004pA-Lc for qemu-devel@nongnu.org; Wed, 17 Jul 2024 11:34:30 -0400 Received: by mail-wr1-x434.google.com with SMTP id ffacd0b85a97d-367990aaef3so4022696f8f.0 for ; Wed, 17 Jul 2024 08:34:28 -0700 (PDT) Received: from paolo-ThinkPad-T490.sou.embecosm-corp.com ([212.69.42.53]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-427c779839bsm2369475e9.7.2024.07.17.08.34.26 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 17 Jul 2024 08:34:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=embecosm.com; s=google; t=1721230467; x=1721835267; darn=nongnu.org; h=references:in-reply-to:message-id:date:subject:cc:to:from:from:to :cc:subject:date:message-id:reply-to; bh=dBJVgoGbVEBWCKNsYWowDSG5FlAgb+PmFPWTiOzZRq8=; b=KpJWHlvIeDzjgigwjWrZW0APt/gx/REH+NbAf/fTxNBNK50lVVdqAJmZDw9Vu/Xw9r aB1W4NbP836H+Gu4feynR2G9mpBHKRalEdfNTRRcijzL5YkahT0v4RyVglpYZ2WosVXQ MsC1PHWner6Tnj0rZoE55Cv7tIlrHLA63Vr4DR50mX8NMThOP74CWgRo1v+sFy/6LINY QYamBR3CHT267wzmDaIsrjf0SPw+q6Lsg+/Q0clJdDyrpiW3At3IwAsnlzzlT0vedHcB S4brd1x54VtambrJ1taaGrhZPN5XK8k0j+NPscDg/tKcYbp1dUjqPeNOWCBbJuKF2oyd owmA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1721230467; x=1721835267; h=references:in-reply-to:message-id:date:subject:cc:to:from :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=dBJVgoGbVEBWCKNsYWowDSG5FlAgb+PmFPWTiOzZRq8=; b=SR9/ypMJLuy9UiY8O4jVMhiEFhGO67GUuvk2oHJh3vGHHwohNZzbOZJUgxoD2Vioj8 takYa7ASWmPXVLhZUDtTxokeSqdaiB6EpPpPo2K1VxQC3vkpV1apFzY2Ku7XxkxBTHv5 5kBgS/9ZQeuGokWhOM15+QQzogvb3owILIXvwSPxyErw/TRF90xXIkWxlB83q/mAN5P8 ih6eJumKDdjmnU0c8zIpK4+T7j5hgUFhooWNCXrUtjzpwTfcdinA23iEBwHoaAX/yeAe ro6YgdMJw6d3cN5ck9taiCsa8WzlwekuJjJ7g1ALQ35z3BtXiuvohFACyBNcOG1tZbuw Lf2w== X-Gm-Message-State: AOJu0YyWwxAqY4Dy/vWYO6tv+qIWOpprUEKyl+aNX3NKnc4GK9V5aiv1 dJVa9mRRszBMYt7bWjAysi3wOt5z+gBTiNlvObnQw4moM+dEYAh4WeSgSpJYPwR+xQCrNFJuzRw W8IP9vA== X-Google-Smtp-Source: AGHT+IFZY/LMmxUFu5e7QOXL/isaeNSjq2sEVYQiXcwxYqZwkUV9XayT8RKL1nPF76r/y5EIw36+WQ== X-Received: by 2002:a05:6000:1449:b0:367:9224:9621 with SMTP id ffacd0b85a97d-36831652dbemr2032820f8f.31.1721230466778; Wed, 17 Jul 2024 08:34:26 -0700 (PDT) From: Paolo Savini To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Cc: Paolo Savini , Richard Handerson , Palmer Dabbelt , Alistair Francis , Bin Meng , Weiwei Li , Daniel Henrique Barboza , Liu Zhiwei , Helene Chelin , Max Chou Subject: [RFC 1/2] target/riscv: rvv: reduce the overhead for simple RISC-V vector unit-stride loads and stores Date: Wed, 17 Jul 2024 16:30:39 +0100 Message-Id: <20240717153040.11073-2-paolo.savini@embecosm.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20240717153040.11073-1-paolo.savini@embecosm.com> References: <20240717153040.11073-1-paolo.savini@embecosm.com> Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=2a00:1450:4864:20::434; envelope-from=paolo.savini@embecosm.com; helo=mail-wr1-x434.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-Mailman-Approved-At: Wed, 17 Jul 2024 11:47:57 -0400 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @embecosm.com) X-ZM-MESSAGEID: 1721231358261116600 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Helene CHELIN This patch improves the performance of the emulation of the RVV unit-stride loads and stores in the following cases: - when the data being loaded/stored per iteration amounts to 8 bytes or les= s. - when the vector length is 16 bytes (VLEN=3D128) and there's no grouping o= f the vector registers (LMUL=3D1). The optimization consists of avoiding the overhead of probing the RAM of the host machine and doing a loop load/store on the input data grouped in chunks of as many bytes as possible (8,4,2,1 bytes). Co-authored-by: Helene CHELIN Co-authored-by: Paolo Savini Signed-off-by: Helene CHELIN --- target/riscv/vector_helper.c | 46 ++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c index 29849a8b66..4b444c6bc5 100644 --- a/target/riscv/vector_helper.c +++ b/target/riscv/vector_helper.c @@ -633,6 +633,52 @@ vext_ldst_us(void *vd, target_ulong base, CPURISCVStat= e *env, uint32_t desc, =20 VSTART_CHECK_EARLY_EXIT(env); =20 + /* For data sizes <=3D 64 bits and for LMUL=3D1 with VLEN=3D128 bits w= e get a + * better performance by doing a simple simulation of the load/store + * without the overhead of prodding the host RAM */ + if ((nf =3D=3D 1) && ((evl << log2_esz) <=3D 8 || + ((vext_lmul(desc) =3D=3D 0) && (simd_maxsz(desc) =3D=3D 16)))) { + + uint32_t evl_b =3D evl << log2_esz; + + for (uint32_t j =3D env->vstart; j < evl_b;) { + addr =3D base + j; + if ((evl_b - j) >=3D 8) { + if (is_load) + lde_d_tlb(env, adjust_addr(env, addr), j, vd, ra); + else + ste_d_tlb(env, adjust_addr(env, addr), j, vd, ra); + j +=3D 8; + } + else if ((evl_b - j) >=3D 4) { + if (is_load) + lde_w_tlb(env, adjust_addr(env, addr), j, vd, ra); + else + ste_w_tlb(env, adjust_addr(env, addr), j, vd, ra); + j +=3D 4; + } + else if ((evl_b - j) >=3D 2) { + if (is_load) + lde_h_tlb(env, adjust_addr(env, addr), j, vd, ra); + else + ste_h_tlb(env, adjust_addr(env, addr), j, vd, ra); + j +=3D 2; + } + else { + if (is_load) + lde_b_tlb(env, adjust_addr(env, addr), j, vd, ra); + else + ste_b_tlb(env, adjust_addr(env, addr), j, vd, ra); + j +=3D 1; + } + } + + env->vstart =3D 0; + vext_set_tail_elems_1s(evl, vd, desc, nf, esz, max_elems); + return; + } + + vext_cont_ldst_elements(&info, base, env->vreg, env->vstart, evl, desc, log2_esz, false); /* Probe the page(s). Exit with exception for any invalid page. */ --=20 2.17.1 From nobody Mon Sep 16 19:08:55 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org ARC-Seal: i=1; a=rsa-sha256; t=1721231359; cv=none; d=zohomail.com; s=zohoarc; b=Y7ug6LC+vKm3i4kLPmno9iMlA+Nygf0JZcd9rvgr7f3ufKGMN5gEZmLMIGif3QHFZER9gebcr5ODMsvptq00uBF4d+Q/gOCNK3a5zy7X8MFE++kZR+bKG9n3a3duQs7KYybS8vZaPqSDD8TfsqHnGBCmWHAFM+zILA6HEoURaz4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1721231359; h=Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=5oDIohgS1ZQJBzOA3AwPKCN3EDp3pXKXOb8IfIED3xI=; b=FdZvMqaQvUsueEj2YOGF7R9sJC8pyLfP2VOcaZCqMYGgvdVGFKkC/Yq7q//CAYb1uuPMd5St7LPBERN8VM55AhS35YqtY39ROx1FH3+RHGfLsy19wrEUkTX3IsDApuBpcT+2vvwoYgIwZE9xQqm9HfuD4O+TFxhdwoSZQaQy1pc= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1721231359720656.4696376466522; Wed, 17 Jul 2024 08:49:19 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1sU6sp-0006VX-5w; Wed, 17 Jul 2024 11:47:59 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1sU6ft-0004el-8b for qemu-devel@nongnu.org; Wed, 17 Jul 2024 11:34:37 -0400 Received: from mail-wr1-x435.google.com ([2a00:1450:4864:20::435]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1sU6fr-0004pg-G5 for qemu-devel@nongnu.org; Wed, 17 Jul 2024 11:34:37 -0400 Received: by mail-wr1-x435.google.com with SMTP id ffacd0b85a97d-367ab76d5e1so2394815f8f.3 for ; Wed, 17 Jul 2024 08:34:34 -0700 (PDT) Received: from paolo-ThinkPad-T490.sou.embecosm-corp.com ([212.69.42.53]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-427c779839bsm2369475e9.7.2024.07.17.08.34.32 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 17 Jul 2024 08:34:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=embecosm.com; s=google; t=1721230473; x=1721835273; darn=nongnu.org; h=references:in-reply-to:message-id:date:subject:cc:to:from:from:to :cc:subject:date:message-id:reply-to; bh=5oDIohgS1ZQJBzOA3AwPKCN3EDp3pXKXOb8IfIED3xI=; b=V8Qi0DBhE/nGuNBqLFTQa8I+y4LD35lMV+syScbLgpcF8ALaPndkRF7Rf3yeITkGUa RRzgmyIfY2nvCRY2OHPSThJDcMQqJn/9u8xCoOicTHA7aK4ek+VFLewvDNmOspPlbcdG DN0HKbgirl28m9ka4zqQHAyAY/3cz/XVVuYhS5pVLhfwQEk2bmOQGeKjtGuGCW47krLJ bn2z6+ouODqQ4ZjguZk7BBqzEy6y2HQMhjVfMEn68JyoEKuEW0cdQkN2nWjpqYX1bZgZ oum+2kuz8hbkvKCGxU0Qfw7FeaDWAsHMxABDUjveZjFgbceYeoGncUpysHjpAmZvjNUx FqLg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1721230473; x=1721835273; h=references:in-reply-to:message-id:date:subject:cc:to:from :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=5oDIohgS1ZQJBzOA3AwPKCN3EDp3pXKXOb8IfIED3xI=; b=Dvq606F0FpyGOYm6yPgmCtnAvB5r0y6DiRn7gG/is9Y2enwmsqV2PJJZ3nBs826AzT okk+snPQuItmKpy0AEqrbLbl2yo4WKGz/Z4x9gRLcqSFsQgwqyADS+C5mj2L+nWTtPTY XqqcuGTu9GObBjvcWvi166kNfAdgDwrvCifI0Vm+mNIG2wLI347xNnVHvgueLQ9C3oTj eBCaft1UEdHo0sXu3MbRQYkwXtymi9bAPRRS20bjyCvKV0RWvGAgVMKTc6/pmecwos3Z p28/pT9An4b/R55/tKPjmaj05bounugJ46ANuWcJB3guJnIr9vwj3iLj7Tzfnu9MybX/ rJzQ== X-Gm-Message-State: AOJu0YxklT9ad6f7oPC6WfPCh1lfq2JKvzOiZx06k/UHEAU2LY1bUG/g FJXMzLcO63Nx6J9XMsYqpaQfpJtGG35dCBQ+/fbOH3fJ7ruKSFOdS6gKSJZzxPE9PyqX51q+wzU llIEeDw== X-Google-Smtp-Source: AGHT+IEUQ6X29LhVbfvTrgKe7YyeB/eRPNwrRdLiZPVgpXpZtfAFV+yMb2SRMJWJ0AjQ/Y9JhCilzw== X-Received: by 2002:a5d:4bc4:0:b0:366:efa2:f2f7 with SMTP id ffacd0b85a97d-36831600b19mr1598385f8f.1.1721230473321; Wed, 17 Jul 2024 08:34:33 -0700 (PDT) From: Paolo Savini To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Cc: Paolo Savini , Richard Handerson , Palmer Dabbelt , Alistair Francis , Bin Meng , Weiwei Li , Daniel Henrique Barboza , Liu Zhiwei , Helene Chelin , Max Chou Subject: [RFC 2/2] target/riscv: rvv: improve performance of RISC-V vector loads and stores on large amounts of data. Date: Wed, 17 Jul 2024 16:30:40 +0100 Message-Id: <20240717153040.11073-3-paolo.savini@embecosm.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20240717153040.11073-1-paolo.savini@embecosm.com> References: <20240717153040.11073-1-paolo.savini@embecosm.com> Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=2a00:1450:4864:20::435; envelope-from=paolo.savini@embecosm.com; helo=mail-wr1-x435.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-Mailman-Approved-At: Wed, 17 Jul 2024 11:47:57 -0400 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @embecosm.com) X-ZM-MESSAGEID: 1721231360318116600 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" This patch optimizes the emulation of unit-stride load/store RVV instructio= ns when the data being loaded/stored per iteration amounts to 64 bytes or more. The optimization consists of calling __builtin_memcpy on chunks of data of = 128 and 256 bytes between the memory address of the simulated vector register a= nd the destination memory address and vice versa. This is done only if we have direct access to the RAM of the host machine. Signed-off-by: Paolo Savini --- target/riscv/vector_helper.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c index 4b444c6bc5..7674972784 100644 --- a/target/riscv/vector_helper.c +++ b/target/riscv/vector_helper.c @@ -486,7 +486,22 @@ vext_group_ldst_host(CPURISCVState *env, void *vd, uin= t32_t byte_end, } =20 fn =3D fns[is_load][group_size]; - fn(vd, byte_offset, host + byte_offset); + + if (byte_offset + 32 < byte_end) { + group_size =3D MO_256; + if (is_load) + __builtin_memcpy((uint8_t *)(vd + byte_offset), (uint8_t *)(host += byte_offset), 32); + else + __builtin_memcpy((uint8_t *)(host + byte_offset), (uint8_t *)(vd += byte_offset), 32); + } else if (byte_offset + 16 < byte_end) { + group_size =3D MO_128; + if (is_load) + __builtin_memcpy((uint8_t *)(vd + byte_offset), (uint8_t *)(host += byte_offset), 16); + else + __builtin_memcpy((uint8_t *)(host + byte_offset), (uint8_t *)(vd += byte_offset), 16); + } else { + fn(vd, byte_offset, host + byte_offset); + } =20 return 1 << group_size; } --=20 2.17.1