From patchwork Mon Dec 10 07:37:32 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mugunthan V N X-Patchwork-Id: 1855581 Return-Path: X-Original-To: patchwork-linux-arm@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork2.kernel.org Received: from merlin.infradead.org (merlin.infradead.org [205.233.59.134]) by patchwork2.kernel.org (Postfix) with ESMTP id 294BBDFB79 for ; Mon, 10 Dec 2012 07:40:37 +0000 (UTC) Received: from localhost ([::1] helo=merlin.infradead.org) by merlin.infradead.org with esmtp (Exim 4.76 #1 (Red Hat Linux)) id 1Thxw2-0001yf-L5; Mon, 10 Dec 2012 07:37:54 +0000 Received: from arroyo.ext.ti.com ([192.94.94.40]) by merlin.infradead.org with esmtps (Exim 4.76 #1 (Red Hat Linux)) id 1Thxvy-0001xX-IQ for linux-arm-kernel@lists.infradead.org; Mon, 10 Dec 2012 07:37:51 +0000 Received: from dbdp20.itg.ti.com ([172.24.170.38]) by arroyo.ext.ti.com (8.13.7/8.13.7) with ESMTP id qBA7bhvw029910; Mon, 10 Dec 2012 01:37:44 -0600 Received: from DBDE70.ent.ti.com (localhost [127.0.0.1]) by dbdp20.itg.ti.com (8.13.8/8.13.8) with ESMTP id qBA7beqh014143; Mon, 10 Dec 2012 13:07:41 +0530 (IST) Received: from dbdp32.itg.ti.com (172.24.170.251) by dbde70.ent.ti.com (172.24.170.148) with Microsoft SMTP Server id 14.1.323.3; Mon, 10 Dec 2012 13:07:40 +0530 Received: from localhost.localdomain (dbdp20.itg.ti.com [172.24.170.38]) by dbdp32.itg.ti.com (8.13.8/8.13.8) with ESMTP id qBA7bbuT001460; Mon, 10 Dec 2012 13:07:37 +0530 From: Mugunthan V N To: Subject: [PATCH 1/1] net: ethernet: davinci_cpdma: Add boundary for rx and tx descriptors Date: Mon, 10 Dec 2012 13:07:32 +0530 Message-ID: <1355125052-28448-1-git-send-email-mugunthanvnm@ti.com> X-Mailer: git-send-email 1.7.0.4 MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20121210_023750_765557_3196FC6B X-CRM114-Status: GOOD ( 18.88 ) X-Spam-Score: -7.6 (-------) X-Spam-Report: SpamAssassin version 3.3.2 on merlin.infradead.org summary: Content analysis details: (-7.6 points) pts rule name description ---- ---------------------- -------------------------------------------------- -5.0 RCVD_IN_DNSWL_HI RBL: Sender listed at http://www.dnswl.org/, high trust [192.94.94.40 listed in list.dnswl.org] -0.0 SPF_PASS SPF: sender matches SPF record -0.7 RP_MATCHES_RCVD Envelope sender domain matches handover relay domain -1.9 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000] Cc: Mugunthan V N , s.hauer@pengutronix.de, linux-omap@vger.kernel.org, davem@davemloft.net, linux-arm-kernel@lists.infradead.org X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-arm-kernel-bounces@lists.infradead.org Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org When there is heavy transmission traffic in the CPDMA, then Rx descriptors memory is also utilized as tx desc memory this leads to reduced rx desc memory which leads to poor performance. This patch adds boundary for tx and rx descriptors in bd ram dividing the descriptor memory to ensure that during heavy transmission tx doesn't use rx descriptors. This patch is already applied to davinci_emac driver, since CPSW and davici_dmac uses the same CPDMA, moving the boundry seperation from Davinci EMAC driver to CPDMA driver which was done in the following commit commit 86d8c07ff2448eb4e860e50f34ef6ee78e45c40c Author: Sascha Hauer Date: Tue Jan 3 05:27:47 2012 +0000 net/davinci: do not use all descriptors for tx packets The driver uses a shared pool for both rx and tx descriptors. During open it queues fixed number of 128 descriptors for receive packets. For each received packet it tries to queue another descriptor. If this fails the descriptor is lost for rx. The driver has no limitation on tx descriptors to use, so it can happen during a nmap / ping -f attack that the driver allocates all descriptors for tx and looses all rx descriptors. The driver stops working then. To fix this limit the number of tx descriptors used to half of the descriptors available, the rx path uses the other half. Tested on a custom board using nmap / ping -f to the board from two different hosts. Signed-off-by: Mugunthan V N --- drivers/net/ethernet/ti/davinci_cpdma.c | 20 ++++++++++++++------ drivers/net/ethernet/ti/davinci_emac.c | 8 -------- 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 4995673..d37f546 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -105,13 +105,13 @@ struct cpdma_ctlr { }; struct cpdma_chan { + struct cpdma_desc __iomem *head, *tail; + void __iomem *hdp, *cp, *rxfree; enum cpdma_state state; struct cpdma_ctlr *ctlr; int chan_num; spinlock_t lock; - struct cpdma_desc __iomem *head, *tail; int count; - void __iomem *hdp, *cp, *rxfree; u32 mask; cpdma_handler_fn handler; enum dma_data_direction dir; @@ -217,7 +217,7 @@ desc_from_phys(struct cpdma_desc_pool *pool, dma_addr_t dma) } static struct cpdma_desc __iomem * -cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc) +cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc, bool is_rx) { unsigned long flags; int index; @@ -225,8 +225,14 @@ cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc) spin_lock_irqsave(&pool->lock, flags); - index = bitmap_find_next_zero_area(pool->bitmap, pool->num_desc, 0, - num_desc, 0); + if (is_rx) { + index = bitmap_find_next_zero_area(pool->bitmap, + pool->num_desc/2, 0, num_desc, 0); + } else { + index = bitmap_find_next_zero_area(pool->bitmap, + pool->num_desc, pool->num_desc/2, num_desc, 0); + } + if (index < pool->num_desc) { bitmap_set(pool->bitmap, index, num_desc); desc = pool->iomap + pool->desc_size * index; @@ -660,6 +666,7 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data, unsigned long flags; u32 mode; int ret = 0; + bool is_rx; spin_lock_irqsave(&chan->lock, flags); @@ -668,7 +675,8 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data, goto unlock_ret; } - desc = cpdma_desc_alloc(ctlr->pool, 1); + is_rx = (chan->rxfree != 0); + desc = cpdma_desc_alloc(ctlr->pool, 1, is_rx); if (!desc) { chan->stats.desc_alloc_fail++; ret = -ENOMEM; diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index fce89a0..f349273 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -120,7 +120,6 @@ static const char emac_version_string[] = "TI DaVinci EMAC Linux v6.1"; #define EMAC_DEF_TX_CH (0) /* Default 0th channel */ #define EMAC_DEF_RX_CH (0) /* Default 0th channel */ #define EMAC_DEF_RX_NUM_DESC (128) -#define EMAC_DEF_TX_NUM_DESC (128) #define EMAC_DEF_MAX_TX_CH (1) /* Max TX channels configured */ #define EMAC_DEF_MAX_RX_CH (1) /* Max RX channels configured */ #define EMAC_POLL_WEIGHT (64) /* Default NAPI poll weight */ @@ -342,7 +341,6 @@ struct emac_priv { u32 mac_hash2; u32 multicast_hash_cnt[EMAC_NUM_MULTICAST_BITS]; u32 rx_addr_type; - atomic_t cur_tx; const char *phy_id; #ifdef CONFIG_OF struct device_node *phy_node; @@ -1050,9 +1048,6 @@ static void emac_tx_handler(void *token, int len, int status) { struct sk_buff *skb = token; struct net_device *ndev = skb->dev; - struct emac_priv *priv = netdev_priv(ndev); - - atomic_dec(&priv->cur_tx); if (unlikely(netif_queue_stopped(ndev))) netif_start_queue(ndev); @@ -1101,9 +1096,6 @@ static int emac_dev_xmit(struct sk_buff *skb, struct net_device *ndev) goto fail_tx; } - if (atomic_inc_return(&priv->cur_tx) >= EMAC_DEF_TX_NUM_DESC) - netif_stop_queue(ndev); - return NETDEV_TX_OK; fail_tx: