Received: by 2002:a05:6358:4e97:b0:b3:742d:4702 with SMTP id ce23csp514125rwb; Sat, 13 Aug 2022 03:02:19 -0700 (PDT) X-Google-Smtp-Source: AA6agR52v+LEbqDfsyixaiKp76x+0uaT8r/gPrgCSsqpzs0Wo2zUicQCiSz/27x+GbJFlE6HeMIk X-Received: by 2002:a17:907:8a14:b0:731:2198:627b with SMTP id sc20-20020a1709078a1400b007312198627bmr5096351ejc.635.1660384939014; Sat, 13 Aug 2022 03:02:19 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1660384939; cv=none; d=google.com; s=arc-20160816; b=AMooHtYLTKh4f9lGonmWmbe+9GrR9V6k59i6WLBzUC4ghJgvjWjkRFa3Pv2q3CM4C7 PZHbZhZojkQy0voHC2/QKMPoTdptebOnXZSA8SXP1cOgy5CEYZs8kjEAVNCytpN9Y7EA uPx5Mkq8aNpCrjtL+Fu/2OdJqGXmQiPM4GMFKOcrf5fhVv38FS3IYbVt4MWvLkeufOyN FjUMtglz0WuNubh/CZqd/1zxYX3IPpVD/TbVd0bHPSNyUnc1LJTli9A4Pah/s44CY4Os Av0kAn1jSIKwgz6PyeJIps3h/oFp1ZJrfhxlbylEUN84cJDqy9mDCPuCDJJ8b81Xwj4h Kg/Q== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :message-id:date:subject:cc:to:from; bh=ep17o1HlT+10Iz++lht3jf2Nl2ImCa4it5rgWGSiUVc=; b=yFra0KJbSM1kCWvETE7Nq/DwJw6TgiE2r73Ngxu3L2Kk8saWefEoyzLIYGjSA1zilV CB9+JGw9TuIxh6uCCaOD0HzAYVDjMnQyIFi1YtUVO84XeCdaHMFuSmJHr872fOjJpLc7 34Zj5zfkrMwnVKo5FFjV/kSAT2BgR5hC0OhjflZQPrs+RoUXfUauye/6DJDEPytDC16a ElvKm7BI5ndcPC3Bs+TVFlArWcWxfcbf9WU1lDnJHKRvK0TPtntzoSjlaAizpdKcgKQt SGu+EKXevlds9uJV6GYN9bjjM7fbfd3QHq5iWZEnDGQjaupLTabmlg0ad7qLlV/rSmDK d5gg== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-crypto-owner@vger.kernel.org; dmarc=fail (p=QUARANTINE sp=QUARANTINE dis=NONE) header.from=huawei.com Return-Path: Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id b17-20020a056402351100b0043d2f04a703si3986841edd.40.2022.08.13.03.01.30; Sat, 13 Aug 2022 03:02:19 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; spf=pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-crypto-owner@vger.kernel.org; dmarc=fail (p=QUARANTINE sp=QUARANTINE dis=NONE) header.from=huawei.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234663AbiHMJ7x (ORCPT + 99 others); Sat, 13 Aug 2022 05:59:53 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:49322 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229507AbiHMJ7w (ORCPT ); Sat, 13 Aug 2022 05:59:52 -0400 Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [45.249.212.187]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6966F8B9BB; Sat, 13 Aug 2022 02:59:50 -0700 (PDT) Received: from dggpemm500021.china.huawei.com (unknown [172.30.72.56]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4M4bZC6TTXzmVLW; Sat, 13 Aug 2022 17:57:39 +0800 (CST) Received: from dggpemm500005.china.huawei.com (7.185.36.74) by dggpemm500021.china.huawei.com (7.185.36.109) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.24; Sat, 13 Aug 2022 17:59:47 +0800 Received: from localhost.localdomain (10.67.164.66) by dggpemm500005.china.huawei.com (7.185.36.74) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.24; Sat, 13 Aug 2022 17:59:47 +0800 From: Yang Shen To: , CC: , , , Subject: [PATCH] crypto: hisilicon/zip - optimization for performance Date: Sat, 13 Aug 2022 17:57:52 +0800 Message-ID: <20220813095752.39941-1-shenyang39@huawei.com> X-Mailer: git-send-email 2.24.0 MIME-Version: 1.0 Content-Transfer-Encoding: 7BIT Content-Type: text/plain; charset=US-ASCII X-Originating-IP: [10.67.164.66] X-ClientProxiedBy: dggems705-chm.china.huawei.com (10.3.19.182) To dggpemm500005.china.huawei.com (7.185.36.74) X-CFilter-Loop: Reflected X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_MED, SPF_HELO_NONE,SPF_PASS,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 lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-crypto@vger.kernel.org 1.Remove some useless steps during doing requests. 2.Adjust the possibility of branch prediction. Signed-off-by: Yang Shen --- drivers/crypto/hisilicon/zip/zip_crypto.c | 27 +++++++++++------------ 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/drivers/crypto/hisilicon/zip/zip_crypto.c b/drivers/crypto/hisilicon/zip/zip_crypto.c index ad35434a3fdb..12b6d86c54d7 100644 --- a/drivers/crypto/hisilicon/zip/zip_crypto.c +++ b/drivers/crypto/hisilicon/zip/zip_crypto.c @@ -183,7 +183,7 @@ static int add_comp_head(struct scatterlist *dst, u8 req_type) int ret; ret = sg_copy_from_buffer(dst, sg_nents(dst), head, head_size); - if (ret != head_size) { + if (unlikely(ret != head_size)) { pr_err("the head size of buffer is wrong (%d)!\n", ret); return -ENOMEM; } @@ -193,11 +193,11 @@ static int add_comp_head(struct scatterlist *dst, u8 req_type) static int get_comp_head_size(struct acomp_req *acomp_req, u8 req_type) { - if (!acomp_req->src || !acomp_req->slen) + if (unlikely(!acomp_req->src || !acomp_req->slen)) return -EINVAL; - if (req_type == HZIP_ALG_TYPE_GZIP && - acomp_req->slen < GZIP_HEAD_FEXTRA_SHIFT) + if (unlikely(req_type == HZIP_ALG_TYPE_GZIP && + acomp_req->slen < GZIP_HEAD_FEXTRA_SHIFT)) return -EINVAL; switch (req_type) { @@ -230,6 +230,8 @@ static struct hisi_zip_req *hisi_zip_create_req(struct acomp_req *req, } set_bit(req_id, req_q->req_bitmap); + write_unlock(&req_q->req_lock); + req_cache = q + req_id; req_cache->req_id = req_id; req_cache->req = req; @@ -242,8 +244,6 @@ static struct hisi_zip_req *hisi_zip_create_req(struct acomp_req *req, req_cache->dskip = 0; } - write_unlock(&req_q->req_lock); - return req_cache; } @@ -254,7 +254,6 @@ static void hisi_zip_remove_req(struct hisi_zip_qp_ctx *qp_ctx, write_lock(&req_q->req_lock); clear_bit(req->req_id, req_q->req_bitmap); - memset(req, 0, sizeof(struct hisi_zip_req)); write_unlock(&req_q->req_lock); } @@ -339,7 +338,7 @@ static int hisi_zip_do_work(struct hisi_zip_req *req, struct hisi_zip_sqe zip_sqe; int ret; - if (!a_req->src || !a_req->slen || !a_req->dst || !a_req->dlen) + if (unlikely(!a_req->src || !a_req->slen || !a_req->dst || !a_req->dlen)) return -EINVAL; req->hw_src = hisi_acc_sg_buf_map_to_hw_sgl(dev, a_req->src, pool, @@ -365,7 +364,7 @@ static int hisi_zip_do_work(struct hisi_zip_req *req, /* send command to start a task */ atomic64_inc(&dfx->send_cnt); ret = hisi_qp_send(qp, &zip_sqe); - if (ret < 0) { + if (unlikely(ret < 0)) { atomic64_inc(&dfx->send_busy_cnt); ret = -EAGAIN; dev_dbg_ratelimited(dev, "failed to send request!\n"); @@ -417,7 +416,7 @@ static void hisi_zip_acomp_cb(struct hisi_qp *qp, void *data) atomic64_inc(&dfx->recv_cnt); status = ops->get_status(sqe); - if (status != 0 && status != HZIP_NC_ERR) { + if (unlikely(status != 0 && status != HZIP_NC_ERR)) { dev_err(dev, "%scompress fail in qp%u: %u, output: %u\n", (qp->alg_type == 0) ? "" : "de", qp->qp_id, status, sqe->produced); @@ -450,7 +449,7 @@ static int hisi_zip_acompress(struct acomp_req *acomp_req) /* let's output compression head now */ head_size = add_comp_head(acomp_req->dst, qp_ctx->qp->req_type); - if (head_size < 0) { + if (unlikely(head_size < 0)) { dev_err_ratelimited(dev, "failed to add comp head (%d)!\n", head_size); return head_size; @@ -461,7 +460,7 @@ static int hisi_zip_acompress(struct acomp_req *acomp_req) return PTR_ERR(req); ret = hisi_zip_do_work(req, qp_ctx); - if (ret != -EINPROGRESS) { + if (unlikely(ret != -EINPROGRESS)) { dev_info_ratelimited(dev, "failed to do compress (%d)!\n", ret); hisi_zip_remove_req(qp_ctx, req); } @@ -478,7 +477,7 @@ static int hisi_zip_adecompress(struct acomp_req *acomp_req) int head_size, ret; head_size = get_comp_head_size(acomp_req, qp_ctx->qp->req_type); - if (head_size < 0) { + if (unlikely(head_size < 0)) { dev_err_ratelimited(dev, "failed to get comp head size (%d)!\n", head_size); return head_size; @@ -489,7 +488,7 @@ static int hisi_zip_adecompress(struct acomp_req *acomp_req) return PTR_ERR(req); ret = hisi_zip_do_work(req, qp_ctx); - if (ret != -EINPROGRESS) { + if (unlikely(ret != -EINPROGRESS)) { dev_info_ratelimited(dev, "failed to do decompress (%d)!\n", ret); hisi_zip_remove_req(qp_ctx, req); -- 2.24.0