From patchwork Tue Aug 1 10:37:57 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Anup Patel X-Patchwork-Id: 9874167 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 17B1E60365 for ; Tue, 1 Aug 2017 10:41:54 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 229892853E for ; Tue, 1 Aug 2017 10:41:54 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 166F72869A; Tue, 1 Aug 2017 10:41:54 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.3 required=2.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI, RCVD_IN_SORBS_SPAM, T_DKIM_INVALID autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 6599D2853E for ; Tue, 1 Aug 2017 10:41:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751806AbdHAKkl (ORCPT ); Tue, 1 Aug 2017 06:40:41 -0400 Received: from mail-qt0-f177.google.com ([209.85.216.177]:35168 "EHLO mail-qt0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752064AbdHAKkj (ORCPT ); Tue, 1 Aug 2017 06:40:39 -0400 Received: by mail-qt0-f177.google.com with SMTP id p3so6434264qtg.2 for ; Tue, 01 Aug 2017 03:40:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=broadcom.com; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=WmZBW/zAyGW0L60Cv3n6w0674oxOQz6HKtYtGGfx7Xw=; b=RjEGPKdJI0cyuQt/zUMZAEcdTwxtSRkgRgw/8S3wm6sCjPja3En1ylJ2rwFKtQ84An nUWkQUneqCEi5IakLhxrOShojGC/hJPdtCAsOAjUXrzYc5WLmKW6popKbwe/ERJRCvHJ 9NBw2u+qwRzzLyuQlkcRqI9kX/RZp58FS62rk= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=WmZBW/zAyGW0L60Cv3n6w0674oxOQz6HKtYtGGfx7Xw=; b=NMNTpze9sr0nYwRf8lmbjiASGESUCvq/DM+C9QPZIhM+aNP6DoHDh2lShe/N2qgucF NiNZ39i/uBcsX3mGsER9dDfJkWsuI7tOaIC5pVYhj/mLk6w8n/Ylm663slyT2HxIT02+ ElcrAMU8NmBKlviV2QTvB6P8Ua+x/3K0N/7N1MwutQW9IPAXGKASRVxA2/wAltA5ryeh tAVUZnrTj1pujI/YSQQ62CJZQJF6dF3nahw4FbG87hrTpaHi9LuwmTjX0MjEpEDyv2II Q8RLCe0fpT+KlwxhjlMaBaxtzyAdQwT3/a2pQCg/Bykl6noibeiiWHwIjhn1XwLop/e6 uB0g== X-Gm-Message-State: AIVw1117EQWXev/0mVyDuPfyYa3pepq21dTcI1iEnxBmJGV40NGDarFI oplaiJcxMkHMmsyr X-Received: by 10.237.60.152 with SMTP id d24mr26517456qtf.126.1501584038063; Tue, 01 Aug 2017 03:40:38 -0700 (PDT) Received: from anup-HP-Compaq-8100-Elite-CMT-PC.dhcp.avagotech.net ([192.19.237.250]) by smtp.gmail.com with ESMTPSA id e32sm21784162qtb.63.2017.08.01.03.40.33 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Tue, 01 Aug 2017 03:40:37 -0700 (PDT) From: Anup Patel To: Rob Herring , Mark Rutland , Vinod Koul , Dan Williams Cc: Florian Fainelli , Scott Branden , Ray Jui , linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, devicetree@vger.kernel.org, dmaengine@vger.kernel.org, bcm-kernel-feedback-list@broadcom.com, Anup Patel Subject: [PATCH v2 13/16] dmaengine: bcm-sba-raid: Re-factor sba_process_deferred_requests() Date: Tue, 1 Aug 2017 16:07:57 +0530 Message-Id: <1501583880-32072-14-git-send-email-anup.patel@broadcom.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1501583880-32072-1-git-send-email-anup.patel@broadcom.com> References: <1501583880-32072-1-git-send-email-anup.patel@broadcom.com> Sender: dmaengine-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: dmaengine@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Currently, sba_process_deferred_requests() handles both pending and completed sba_request which is unnecessary overhead for sba_issue_pending() because completed sba_request handling is not required in sba_issue_pending(). This patch breaks sba_process_deferred_requests() into two parts sba_process_received_request() and _sba_process_pending_requests(). The sba_issue_pending() will only process pending sba_request by calling _sba_process_pending_requests(). This will improve sba_issue_pending(). The sba_receive_message() will only process received sba_request by calling sba_process_received_request() for each received sba_request. The sba_process_received_request() will also call _sba_process_pending_requests() after handling received sba_request because we might have pending sba_request not submitted by previous call to sba_issue_pending(). Signed-off-by: Anup Patel Reviewed-by: Scott Branden --- drivers/dma/bcm-sba-raid.c | 109 +++++++++++++++++++-------------------------- 1 file changed, 47 insertions(+), 62 deletions(-) diff --git a/drivers/dma/bcm-sba-raid.c b/drivers/dma/bcm-sba-raid.c index 2f444cc..cb6b2e2 100644 --- a/drivers/dma/bcm-sba-raid.c +++ b/drivers/dma/bcm-sba-raid.c @@ -419,22 +419,20 @@ static int sba_send_mbox_request(struct sba_device *sba, return 0; } -static void sba_process_deferred_requests(struct sba_device *sba) +/* Note: Must be called with sba->reqs_lock held */ +static void _sba_process_pending_requests(struct sba_device *sba) { int ret; u32 count; - unsigned long flags; struct sba_request *req; - struct dma_async_tx_descriptor *tx; - - spin_lock_irqsave(&sba->reqs_lock, flags); - - /* Count pending requests */ - count = 0; - list_for_each_entry(req, &sba->reqs_pending_list, node) - count++; - /* Process pending requests */ + /* + * Process few pending requests + * + * For now, we process ( * 8) + * number of requests at a time. + */ + count = sba->mchans_count * 8; while (!list_empty(&sba->reqs_pending_list) && count) { /* Get the first pending request */ req = list_first_entry(&sba->reqs_pending_list, @@ -445,11 +443,7 @@ static void sba_process_deferred_requests(struct sba_device *sba) break; /* Send request to mailbox channel */ - spin_unlock_irqrestore(&sba->reqs_lock, flags); ret = sba_send_mbox_request(sba, req); - spin_lock_irqsave(&sba->reqs_lock, flags); - - /* If something went wrong then keep request pending */ if (ret < 0) { _sba_pending_request(sba, req); break; @@ -457,20 +451,18 @@ static void sba_process_deferred_requests(struct sba_device *sba) count--; } +} - /* Count completed requests */ - count = 0; - list_for_each_entry(req, &sba->reqs_completed_list, node) - count++; - - /* Process completed requests */ - while (!list_empty(&sba->reqs_completed_list) && count) { - req = list_first_entry(&sba->reqs_completed_list, - struct sba_request, node); - list_del_init(&req->node); - tx = &req->tx; +static void sba_process_received_request(struct sba_device *sba, + struct sba_request *req) +{ + unsigned long flags; + struct dma_async_tx_descriptor *tx; + struct sba_request *nreq, *first = req->first; - spin_unlock_irqrestore(&sba->reqs_lock, flags); + /* Process only after all chained requests are received */ + if (!atomic_dec_return(&first->next_pending_count)) { + tx = &first->tx; WARN_ON(tx->cookie < 0); if (tx->cookie > 0) { @@ -485,41 +477,34 @@ static void sba_process_deferred_requests(struct sba_device *sba) spin_lock_irqsave(&sba->reqs_lock, flags); - /* If waiting for 'ack' then move to completed list */ - if (!async_tx_test_ack(&req->tx)) - _sba_complete_request(sba, req); - else - _sba_free_request(sba, req); + /* Free all requests chained to first request */ + list_for_each_entry(nreq, &first->next, next) + _sba_free_request(sba, nreq); + INIT_LIST_HEAD(&first->next); - count--; - } + /* Mark request as received */ + _sba_received_request(sba, first); - /* Re-check pending and completed work */ - count = 0; - if (!list_empty(&sba->reqs_pending_list) || - !list_empty(&sba->reqs_completed_list)) - count = 1; - - spin_unlock_irqrestore(&sba->reqs_lock, flags); -} - -static void sba_process_received_request(struct sba_device *sba, - struct sba_request *req) -{ - unsigned long flags; - - spin_lock_irqsave(&sba->reqs_lock, flags); + /* The client is allowed to attach dependent operations + * until 'ack' is set + */ + if (!async_tx_test_ack(tx)) + _sba_complete_request(sba, first); + else + _sba_free_request(sba, first); - /* Mark request as received */ - _sba_received_request(sba, req); + /* Cleanup completed requests */ + list_for_each_entry_safe(req, nreq, + &sba->reqs_completed_list, node) { + if (async_tx_test_ack(&req->tx)) + _sba_free_request(sba, req); + } - /* Update request */ - if (!atomic_dec_return(&req->first->next_pending_count)) - _sba_complete_request(sba, req->first); - if (req->first != req) - _sba_free_request(sba, req); + /* Process pending requests */ + _sba_process_pending_requests(sba); - spin_unlock_irqrestore(&sba->reqs_lock, flags); + spin_unlock_irqrestore(&sba->reqs_lock, flags); + } } /* ====== DMAENGINE callbacks ===== */ @@ -544,10 +529,13 @@ static int sba_device_terminate_all(struct dma_chan *dchan) static void sba_issue_pending(struct dma_chan *dchan) { + unsigned long flags; struct sba_device *sba = to_sba_device(dchan); - /* Process deferred requests */ - sba_process_deferred_requests(sba); + /* Process pending requests */ + spin_lock_irqsave(&sba->reqs_lock, flags); + _sba_process_pending_requests(sba); + spin_unlock_irqrestore(&sba->reqs_lock, flags); } static dma_cookie_t sba_tx_submit(struct dma_async_tx_descriptor *tx) @@ -1482,9 +1470,6 @@ static void sba_receive_message(struct mbox_client *cl, void *msg) /* Process received request */ sba_process_received_request(sba, req); - - /* Process deferred requests */ - sba_process_deferred_requests(sba); } /* ====== Platform driver routines ===== */