From patchwork Wed Jul 3 13:51:32 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 810620 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id B5F4817B51F; Wed, 3 Jul 2024 13:51:42 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014702; cv=none; b=B0GmK0BrrJzJLuSKXxBrYMV425VVsHjGutM2LZEjtYQKaKNs2C7qCxDZAFONKRulZ36/+TvvAUD29rYa/LUJ0ONWyxc4pc5Mjvg32ZirxsBFwkTAM4CLsFe1HVA/Tse5frvqdj89nRdHVlfFnEJbU/c80ZNx4BQc3heCMvjurl8= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014702; c=relaxed/simple; bh=QR86VhqJ+4pxj/XNlNWPVEqriIMqPhsPmYz8sb6x0k4=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=jcjG/Hqyho1Tz0u97g9Z4D9pMtJr5FFtwKVIP/WnDrXBjXPWPxmIIEUblqWi1lQbByJYtA3+rOFVzgBTXms+NEMTdnEc9sXtCLQoubiHw37lwOpWuDK0T/gWeoCR3eYrE1ec1WEhJWaxl/1bCAh9hroAueYsXMPZsVL90btfTac= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=FH2TPs2I; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="FH2TPs2I" Received: by smtp.kernel.org (Postfix) with ESMTPSA id D0B3CC32781; Wed, 3 Jul 2024 13:51:38 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014702; bh=QR86VhqJ+4pxj/XNlNWPVEqriIMqPhsPmYz8sb6x0k4=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=FH2TPs2IiLt5IkBX1I3YrSkagwX4QTLXvfXW8qoggYjTsBIi3iQAl/cKFx23JezKD I3g0s57MBKl4LtcY0TefplHV5r6uX9L8oung2HUUV3ch2ErxJx+FfOvflVcAK1F1fi xBYBRxFjZdX7oROUKA6sW9zvccmMzJcSzx9EXszjG7dFCnMtfCamiPvgn9NLOlYrrd 3+hsfpxBr0sb0TbBT1hPW6jLtY3XOZwX4lylkvtS5qBI5G63GbUn173WDblB7lq+n1 okFeD+43ejwHYXoclC6WXsiVw2Mzct8hG95dgr5dgK3arzJxQe8zGe9Jfd9nhUgXzs lJsIL5ny0ZyRg== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:32 +0300 Subject: [PATCH net-next v3 1/6] net: ethernet: ti: am65-cpsw: Introduce multi queue Rx Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-1-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=31945; i=rogerq@kernel.org; h=from:subject:message-id; bh=QR86VhqJ+4pxj/XNlNWPVEqriIMqPhsPmYz8sb6x0k4=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdm1EbDVUasP8xyGBVgdkjQF/8nIHnI/bZEW VzpBAtcYySJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw k0SlD/4tMkav9cWXO0LukvphkYRe3+MDO2gUO85LXvz/rcLbj9qrSR1eUDzeYRPRp7zCZFiL17D 4HmRXrRHL5BVpPfHPEFr0daiCYt2UYEv/OHmRUgXIysAQ4KcjlD5jSkiLUz6u1MsdQmHmnh/SVa 8o0K1hyGnXN4ptaUW7U+PiO06nb0dSV2pe+sVxI3I7UJ0h69JD3ido01tmDf1v9AxuOIHvxhZBr lCSXDqBSBc1BTSrY/Xm5lSyTzZeSO8XDtryy86muP3jSApsp28KvgXh9Ze4YmjJvFXq+qjU25mk MKld+mhndEAsDwBVDpWnkxaaTum5kssRby2WsuV8SAcVAcsNo4E9RLrk1hKh+tmHSBMfzOHyZm1 cx1ReD3E38Tm7XYN+zHTY75bMynsM2DQyC3BNH8Y2v15RY7FfkN3V5scv/L/tRY+CzbtE4nrjWS kh4tBNe3w/bolN3er7QJF5hrWlqGLt+ZM4Psz20LEDcVM0pwla37Ymis9ttWZ9Uu2o1LTdw8Npm 98oagT5RW2aO/I2AgPeyrOYcFN/WcPRM2bUMqd1q4/SsfjQDUPabSVcBxSIrNlYgjWqkc3d7i23 7o5mt8k/3OjdoR1mJmfDHyRW/xBMvwEbEQbffBTNFKhvCXJBDT+tfDEfpShN7FrshXpjmHGCcWE YJ0f4h23R10haew== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 am65-cpsw can support up to 8 queues at Rx. Use a macro AM65_CPSW_MAX_RX_QUEUES to indicate that. As there is only one DMA channel for RX traffic, the 8 queues come as 8 flows in that channel. By default, we will start with 1 flow as defined by the macro AM65_CPSW_DEFAULT_RX_CHN_FLOWS. User can change the number of flows by ethtool like so 'ethtool -L ethx rx ' All traffic will still come on flow 0. To get traffic on different flows the Classifiers will need to be set up. Signed-off-by: Roger Quadros Reviewed-by: Simon Horman --- Changelog: v3: - style fixes: reverse xmas tree and checkpatch.pl --max-line-length=80 - typo fix: Classifer -> Classifier - added Reviewed-by Simon Horman --- drivers/net/ethernet/ti/am65-cpsw-ethtool.c | 62 +++-- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 367 ++++++++++++++++------------ drivers/net/ethernet/ti/am65-cpsw-nuss.h | 36 +-- 3 files changed, 284 insertions(+), 181 deletions(-) diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c index a1d0935d1ebe..01e3967852e0 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c +++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c @@ -429,7 +429,7 @@ static void am65_cpsw_get_channels(struct net_device *ndev, ch->max_rx = AM65_CPSW_MAX_RX_QUEUES; ch->max_tx = AM65_CPSW_MAX_TX_QUEUES; - ch->rx_count = AM65_CPSW_MAX_RX_QUEUES; + ch->rx_count = common->rx_ch_num_flows; ch->tx_count = common->tx_ch_num; } @@ -448,8 +448,10 @@ static int am65_cpsw_set_channels(struct net_device *ndev, return -EBUSY; am65_cpsw_nuss_remove_tx_chns(common); + am65_cpsw_nuss_remove_rx_chns(common); - return am65_cpsw_nuss_update_tx_chns(common, chs->tx_count); + return am65_cpsw_nuss_update_tx_rx_chns(common, chs->tx_count, + chs->rx_count); } static void @@ -920,11 +922,13 @@ static int am65_cpsw_get_coalesce(struct net_device *ndev, struct ethtool_coales struct netlink_ext_ack *extack) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_rx_flow *rx_flow; struct am65_cpsw_tx_chn *tx_chn; tx_chn = &common->tx_chns[0]; + rx_flow = &common->rx_chns.flows[0]; - coal->rx_coalesce_usecs = common->rx_pace_timeout / 1000; + coal->rx_coalesce_usecs = rx_flow->rx_pace_timeout / 1000; coal->tx_coalesce_usecs = tx_chn->tx_pace_timeout / 1000; return 0; @@ -934,14 +938,26 @@ static int am65_cpsw_get_per_queue_coalesce(struct net_device *ndev, u32 queue, struct ethtool_coalesce *coal) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_rx_flow *rx_flow; struct am65_cpsw_tx_chn *tx_chn; - if (queue >= AM65_CPSW_MAX_TX_QUEUES) + if (queue >= AM65_CPSW_MAX_TX_QUEUES && + queue >= AM65_CPSW_MAX_RX_QUEUES) return -EINVAL; - tx_chn = &common->tx_chns[queue]; + if (queue < AM65_CPSW_MAX_TX_QUEUES) { + tx_chn = &common->tx_chns[queue]; + coal->tx_coalesce_usecs = tx_chn->tx_pace_timeout / 1000; + } else { + coal->tx_coalesce_usecs = ~0; + } - coal->tx_coalesce_usecs = tx_chn->tx_pace_timeout / 1000; + if (queue < AM65_CPSW_MAX_RX_QUEUES) { + rx_flow = &common->rx_chns.flows[queue]; + coal->rx_coalesce_usecs = rx_flow->rx_pace_timeout / 1000; + } else { + coal->rx_coalesce_usecs = ~0; + } return 0; } @@ -951,9 +967,11 @@ static int am65_cpsw_set_coalesce(struct net_device *ndev, struct ethtool_coales struct netlink_ext_ack *extack) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_rx_flow *rx_flow; struct am65_cpsw_tx_chn *tx_chn; tx_chn = &common->tx_chns[0]; + rx_flow = &common->rx_chns.flows[0]; if (coal->rx_coalesce_usecs && coal->rx_coalesce_usecs < 20) return -EINVAL; @@ -961,7 +979,7 @@ static int am65_cpsw_set_coalesce(struct net_device *ndev, struct ethtool_coales if (coal->tx_coalesce_usecs && coal->tx_coalesce_usecs < 20) return -EINVAL; - common->rx_pace_timeout = coal->rx_coalesce_usecs * 1000; + rx_flow->rx_pace_timeout = coal->rx_coalesce_usecs * 1000; tx_chn->tx_pace_timeout = coal->tx_coalesce_usecs * 1000; return 0; @@ -971,20 +989,36 @@ static int am65_cpsw_set_per_queue_coalesce(struct net_device *ndev, u32 queue, struct ethtool_coalesce *coal) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_rx_flow *rx_flow; struct am65_cpsw_tx_chn *tx_chn; - if (queue >= AM65_CPSW_MAX_TX_QUEUES) + if (queue >= AM65_CPSW_MAX_TX_QUEUES && + queue >= AM65_CPSW_MAX_RX_QUEUES) return -EINVAL; - tx_chn = &common->tx_chns[queue]; + if (queue < AM65_CPSW_MAX_TX_QUEUES) { + tx_chn = &common->tx_chns[queue]; + + if (coal->tx_coalesce_usecs && coal->tx_coalesce_usecs < 20) { + dev_info(common->dev, "defaulting to min value of 20us for tx-usecs for tx-%u\n", + queue); + coal->tx_coalesce_usecs = 20; + } - if (coal->tx_coalesce_usecs && coal->tx_coalesce_usecs < 20) { - dev_info(common->dev, "defaulting to min value of 20us for tx-usecs for tx-%u\n", - queue); - coal->tx_coalesce_usecs = 20; + tx_chn->tx_pace_timeout = coal->tx_coalesce_usecs * 1000; } - tx_chn->tx_pace_timeout = coal->tx_coalesce_usecs * 1000; + if (queue < AM65_CPSW_MAX_RX_QUEUES) { + rx_flow = &common->rx_chns.flows[queue]; + + if (coal->rx_coalesce_usecs && coal->rx_coalesce_usecs < 20) { + dev_info(common->dev, "defaulting to min value of 20us for rx-usecs for rx-%u\n", + queue); + coal->rx_coalesce_usecs = 20; + } + + rx_flow->rx_pace_timeout = coal->rx_coalesce_usecs * 1000; + } return 0; } diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 81d9f21086ec..deb8e39e4648 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -138,7 +138,7 @@ AM65_CPSW_PN_TS_CTL_RX_ANX_F_EN) #define AM65_CPSW_ALE_AGEOUT_DEFAULT 30 -/* Number of TX/RX descriptors */ +/* Number of TX/RX descriptors per channel/flow */ #define AM65_CPSW_MAX_TX_DESC 500 #define AM65_CPSW_MAX_RX_DESC 500 @@ -150,6 +150,7 @@ NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) #define AM65_CPSW_DEFAULT_TX_CHNS 8 +#define AM65_CPSW_DEFAULT_RX_CHN_FLOWS 1 /* CPPI streaming packet interface */ #define AM65_CPSW_CPPI_TX_FLOW_ID 0x3FFF @@ -330,7 +331,7 @@ static void am65_cpsw_nuss_ndo_host_tx_timeout(struct net_device *ndev, } static int am65_cpsw_nuss_rx_push(struct am65_cpsw_common *common, - struct page *page) + struct page *page, u32 flow_idx) { struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; struct cppi5_host_desc_t *desc_rx; @@ -363,7 +364,8 @@ static int am65_cpsw_nuss_rx_push(struct am65_cpsw_common *common, swdata = cppi5_hdesc_get_swdata(desc_rx); *((void **)swdata) = page_address(page); - return k3_udma_glue_push_rx_chn(rx_chn->rx_chn, 0, desc_rx, desc_dma); + return k3_udma_glue_push_rx_chn(rx_chn->rx_chn, flow_idx, + desc_rx, desc_dma); } void am65_cpsw_nuss_set_p0_ptype(struct am65_cpsw_common *common) @@ -398,22 +400,27 @@ static void am65_cpsw_init_port_emac_ale(struct am65_cpsw_port *port); static void am65_cpsw_destroy_xdp_rxqs(struct am65_cpsw_common *common) { struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + struct am65_cpsw_rx_flow *flow; struct xdp_rxq_info *rxq; - int i; + int id, port; - for (i = 0; i < common->port_num; i++) { - if (!common->ports[i].ndev) - continue; + for (id = 0; id < common->rx_ch_num_flows; id++) { + flow = &rx_chn->flows[id]; - rxq = &common->ports[i].xdp_rxq; + for (port = 0; port < common->port_num; port++) { + if (!common->ports[port].ndev) + continue; - if (xdp_rxq_info_is_reg(rxq)) - xdp_rxq_info_unreg(rxq); - } + rxq = &common->ports[port].xdp_rxq[id]; - if (rx_chn->page_pool) { - page_pool_destroy(rx_chn->page_pool); - rx_chn->page_pool = NULL; + if (xdp_rxq_info_is_reg(rxq)) + xdp_rxq_info_unreg(rxq); + } + + if (flow->page_pool) { + page_pool_destroy(flow->page_pool); + flow->page_pool = NULL; + } } } @@ -427,31 +434,44 @@ static int am65_cpsw_create_xdp_rxqs(struct am65_cpsw_common *common) .nid = dev_to_node(common->dev), .dev = common->dev, .dma_dir = DMA_BIDIRECTIONAL, - .napi = &common->napi_rx, + /* .napi set dynamically */ }; + struct am65_cpsw_rx_flow *flow; struct xdp_rxq_info *rxq; struct page_pool *pool; - int i, ret; - - pool = page_pool_create(&pp_params); - if (IS_ERR(pool)) - return PTR_ERR(pool); + int id, port, ret; + + for (id = 0; id < common->rx_ch_num_flows; id++) { + flow = &rx_chn->flows[id]; + pp_params.napi = &flow->napi_rx; + pool = page_pool_create(&pp_params); + if (IS_ERR(pool)) { + ret = PTR_ERR(pool); + goto err; + } - rx_chn->page_pool = pool; + flow->page_pool = pool; - for (i = 0; i < common->port_num; i++) { - if (!common->ports[i].ndev) - continue; + /* using same page pool is allowed as no running rx handlers + * simultaneously for both ndevs + */ + for (port = 0; port < common->port_num; port++) { + if (!common->ports[port].ndev) + continue; - rxq = &common->ports[i].xdp_rxq; + rxq = &common->ports[port].xdp_rxq[id]; - ret = xdp_rxq_info_reg(rxq, common->ports[i].ndev, i, 0); - if (ret) - goto err; + ret = xdp_rxq_info_reg(rxq, common->ports[port].ndev, + id, flow->napi_rx.napi_id); + if (ret) + goto err; - ret = xdp_rxq_info_reg_mem_model(rxq, MEM_TYPE_PAGE_POOL, pool); - if (ret) - goto err; + ret = xdp_rxq_info_reg_mem_model(rxq, + MEM_TYPE_PAGE_POOL, + pool); + if (ret) + goto err; + } } return 0; @@ -496,25 +516,27 @@ static enum am65_cpsw_tx_buf_type am65_cpsw_nuss_buf_type(struct am65_cpsw_tx_ch desc_idx); } -static inline void am65_cpsw_put_page(struct am65_cpsw_rx_chn *rx_chn, +static inline void am65_cpsw_put_page(struct am65_cpsw_rx_flow *flow, struct page *page, bool allow_direct, int desc_idx) { - page_pool_put_full_page(rx_chn->page_pool, page, allow_direct); - rx_chn->pages[desc_idx] = NULL; + page_pool_put_full_page(flow->page_pool, page, allow_direct); + flow->pages[desc_idx] = NULL; } static void am65_cpsw_nuss_rx_cleanup(void *data, dma_addr_t desc_dma) { - struct am65_cpsw_rx_chn *rx_chn = data; + struct am65_cpsw_rx_flow *flow = data; struct cppi5_host_desc_t *desc_rx; + struct am65_cpsw_rx_chn *rx_chn; dma_addr_t buf_dma; u32 buf_dma_len; void *page_addr; void **swdata; int desc_idx; + rx_chn = &flow->common->rx_chns; desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); swdata = cppi5_hdesc_get_swdata(desc_rx); page_addr = *swdata; @@ -525,7 +547,7 @@ static void am65_cpsw_nuss_rx_cleanup(void *data, dma_addr_t desc_dma) desc_idx = am65_cpsw_nuss_desc_idx(rx_chn->desc_pool, desc_rx, rx_chn->dsize_log2); - am65_cpsw_put_page(rx_chn, virt_to_page(page_addr), false, desc_idx); + am65_cpsw_put_page(flow, virt_to_page(page_addr), false, desc_idx); } static void am65_cpsw_nuss_xmit_free(struct am65_cpsw_tx_chn *tx_chn, @@ -601,7 +623,8 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common) struct am65_cpsw_host *host_p = am65_common_get_host(common); struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; struct am65_cpsw_tx_chn *tx_chn = common->tx_chns; - int port_idx, i, ret, tx; + int port_idx, i, ret, tx, flow_idx; + struct am65_cpsw_rx_flow *flow; u32 val, port_mask; struct page *page; @@ -669,27 +692,32 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common) return ret; } - for (i = 0; i < rx_chn->descs_num; i++) { - page = page_pool_dev_alloc_pages(rx_chn->page_pool); - if (!page) { - ret = -ENOMEM; - if (i) - goto fail_rx; + for (flow_idx = 0; flow_idx < common->rx_ch_num_flows; flow_idx++) { + flow = &rx_chn->flows[flow_idx]; + for (i = 0; i < AM65_CPSW_MAX_RX_DESC; i++) { + page = page_pool_dev_alloc_pages(flow->page_pool); + if (!page) { + dev_err(common->dev, "cannot allocate page in flow %d\n", + flow_idx); + ret = -ENOMEM; + if (i) + goto fail_rx; - return ret; - } - rx_chn->pages[i] = page; + return ret; + } + flow->pages[i] = page; - ret = am65_cpsw_nuss_rx_push(common, page); - if (ret < 0) { - dev_err(common->dev, - "cannot submit page to channel rx: %d\n", - ret); - am65_cpsw_put_page(rx_chn, page, false, i); - if (i) - goto fail_rx; + ret = am65_cpsw_nuss_rx_push(common, page, flow_idx); + if (ret < 0) { + dev_err(common->dev, + "cannot submit page to rx channel flow %d, error %d\n", + flow_idx, ret); + am65_cpsw_put_page(flow, page, false, i); + if (i) + goto fail_rx; - return ret; + return ret; + } } } @@ -699,6 +727,14 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common) goto fail_rx; } + for (i = 0; i < common->rx_ch_num_flows ; i++) { + napi_enable(&common->rx_chns.flows[i].napi_rx); + if (common->rx_chns.flows[i].irq_disabled) { + common->rx_chns.flows[i].irq_disabled = false; + enable_irq(common->rx_chns.flows[i].irq); + } + } + for (tx = 0; tx < common->tx_ch_num; tx++) { ret = k3_udma_glue_enable_tx_chn(tx_chn[tx].tx_chn); if (ret) { @@ -710,12 +746,6 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common) napi_enable(&tx_chn[tx].napi_tx); } - napi_enable(&common->napi_rx); - if (common->rx_irq_disabled) { - common->rx_irq_disabled = false; - enable_irq(rx_chn->irq); - } - dev_dbg(common->dev, "cpsw_nuss started\n"); return 0; @@ -726,11 +756,24 @@ static int am65_cpsw_nuss_common_open(struct am65_cpsw_common *common) tx--; } + for (flow_idx = 0; i < common->rx_ch_num_flows; flow_idx++) { + flow = &rx_chn->flows[flow_idx]; + if (!flow->irq_disabled) { + disable_irq(flow->irq); + flow->irq_disabled = true; + } + napi_disable(&flow->napi_rx); + } + k3_udma_glue_disable_rx_chn(rx_chn->rx_chn); fail_rx: - k3_udma_glue_reset_rx_chn(rx_chn->rx_chn, 0, rx_chn, - am65_cpsw_nuss_rx_cleanup, 0); + for (i = 0; i < common->rx_ch_num_flows; i--) + k3_udma_glue_reset_rx_chn(rx_chn->rx_chn, i, &rx_chn->flows[i], + am65_cpsw_nuss_rx_cleanup, !!i); + + am65_cpsw_destroy_xdp_rxqs(common); + return ret; } @@ -779,12 +822,12 @@ static int am65_cpsw_nuss_common_stop(struct am65_cpsw_common *common) dev_err(common->dev, "rx teardown timeout\n"); } - napi_disable(&common->napi_rx); - hrtimer_cancel(&common->rx_hrtimer); - - for (i = 0; i < AM65_CPSW_MAX_RX_FLOWS; i++) - k3_udma_glue_reset_rx_chn(rx_chn->rx_chn, i, rx_chn, + for (i = 0; i < common->rx_ch_num_flows; i++) { + napi_disable(&common->rx_chns.flows[i].napi_rx); + hrtimer_cancel(&common->rx_chns.flows[i].rx_hrtimer); + k3_udma_glue_reset_rx_chn(rx_chn->rx_chn, i, &rx_chn->flows[i], am65_cpsw_nuss_rx_cleanup, !!i); + } k3_udma_glue_disable_rx_chn(rx_chn->rx_chn); @@ -793,10 +836,6 @@ static int am65_cpsw_nuss_common_stop(struct am65_cpsw_common *common) writel(0, common->cpsw_base + AM65_CPSW_REG_CTL); writel(0, common->cpsw_base + AM65_CPSW_REG_STAT_PORT_EN); - for (i = 0; i < rx_chn->descs_num; i++) { - if (rx_chn->pages[i]) - am65_cpsw_put_page(rx_chn, rx_chn->pages[i], false, i); - } am65_cpsw_destroy_xdp_rxqs(common); dev_dbg(common->dev, "cpsw_nuss stopped\n"); @@ -867,7 +906,7 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev) goto runtime_put; } - ret = netif_set_real_num_rx_queues(ndev, AM65_CPSW_MAX_RX_QUEUES); + ret = netif_set_real_num_rx_queues(ndev, common->rx_ch_num_flows); if (ret) { dev_err(common->dev, "cannot set real number of rx queues\n"); goto runtime_put; @@ -990,12 +1029,12 @@ static int am65_cpsw_xdp_tx_frame(struct net_device *ndev, return ret; } -static int am65_cpsw_run_xdp(struct am65_cpsw_common *common, +static int am65_cpsw_run_xdp(struct am65_cpsw_rx_flow *flow, struct am65_cpsw_port *port, struct xdp_buff *xdp, int desc_idx, int cpu, int *len) { - struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + struct am65_cpsw_common *common = flow->common; struct net_device *ndev = port->ndev; int ret = AM65_CPSW_XDP_CONSUMED; struct am65_cpsw_tx_chn *tx_chn; @@ -1055,7 +1094,7 @@ static int am65_cpsw_run_xdp(struct am65_cpsw_common *common, } page = virt_to_head_page(xdp->data); - am65_cpsw_put_page(rx_chn, page, true, desc_idx); + am65_cpsw_put_page(flow, page, true, desc_idx); out: return ret; @@ -1094,11 +1133,11 @@ static void am65_cpsw_nuss_rx_csum(struct sk_buff *skb, u32 csum_info) } } -static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, - u32 flow_idx, int cpu) +static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_rx_flow *flow, int cpu) { - struct am65_cpsw_rx_chn *rx_chn = &common->rx_chns; + struct am65_cpsw_rx_chn *rx_chn = &flow->common->rx_chns; u32 buf_dma_len, pkt_len, port_id = 0, csum_info; + struct am65_cpsw_common *common = flow->common; struct am65_cpsw_ndev_priv *ndev_priv; struct am65_cpsw_ndev_stats *stats; struct cppi5_host_desc_t *desc_rx; @@ -1108,6 +1147,7 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, struct am65_cpsw_port *port; int headroom, desc_idx, ret; struct net_device *ndev; + u32 flow_idx = flow->id; struct sk_buff *skb; struct xdp_buff xdp; void *page_addr; @@ -1161,12 +1201,13 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, } if (port->xdp_prog) { - xdp_init_buff(&xdp, AM65_CPSW_MAX_PACKET_SIZE, &port->xdp_rxq); + xdp_init_buff(&xdp, AM65_CPSW_MAX_PACKET_SIZE, + &port->xdp_rxq[flow->id]); xdp_prepare_buff(&xdp, page_addr, skb_headroom(skb), pkt_len, false); - ret = am65_cpsw_run_xdp(common, port, &xdp, desc_idx, + ret = am65_cpsw_run_xdp(flow, port, &xdp, desc_idx, cpu, &pkt_len); if (ret != AM65_CPSW_XDP_PASS) return ret; @@ -1184,7 +1225,7 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, skb_mark_for_recycle(skb); skb->protocol = eth_type_trans(skb, ndev); am65_cpsw_nuss_rx_csum(skb, csum_info); - napi_gro_receive(&common->napi_rx, skb); + napi_gro_receive(&flow->napi_rx, skb); stats = this_cpu_ptr(ndev_priv->stats); @@ -1193,21 +1234,21 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, stats->rx_bytes += pkt_len; u64_stats_update_end(&stats->syncp); - new_page = page_pool_dev_alloc_pages(rx_chn->page_pool); + new_page = page_pool_dev_alloc_pages(flow->page_pool); if (unlikely(!new_page)) return -ENOMEM; - rx_chn->pages[desc_idx] = new_page; + flow->pages[desc_idx] = new_page; if (netif_dormant(ndev)) { - am65_cpsw_put_page(rx_chn, new_page, true, desc_idx); + am65_cpsw_put_page(flow, new_page, true, desc_idx); ndev->stats.rx_dropped++; return 0; } requeue: - ret = am65_cpsw_nuss_rx_push(common, new_page); + ret = am65_cpsw_nuss_rx_push(common, new_page, flow_idx); if (WARN_ON(ret < 0)) { - am65_cpsw_put_page(rx_chn, new_page, true, desc_idx); + am65_cpsw_put_page(flow, new_page, true, desc_idx); ndev->stats.rx_errors++; ndev->stats.rx_dropped++; } @@ -1217,38 +1258,33 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, static enum hrtimer_restart am65_cpsw_nuss_rx_timer_callback(struct hrtimer *timer) { - struct am65_cpsw_common *common = - container_of(timer, struct am65_cpsw_common, rx_hrtimer); + struct am65_cpsw_rx_flow *flow = container_of(timer, + struct am65_cpsw_rx_flow, + rx_hrtimer); - enable_irq(common->rx_chns.irq); + enable_irq(flow->irq); return HRTIMER_NORESTART; } static int am65_cpsw_nuss_rx_poll(struct napi_struct *napi_rx, int budget) { - struct am65_cpsw_common *common = am65_cpsw_napi_to_common(napi_rx); - int flow = AM65_CPSW_MAX_RX_FLOWS; + struct am65_cpsw_rx_flow *flow = am65_cpsw_napi_to_rx_flow(napi_rx); + struct am65_cpsw_common *common = flow->common; int cpu = smp_processor_id(); bool xdp_redirect = false; int cur_budget, ret; int num_rx = 0; - /* process every flow */ - while (flow--) { - cur_budget = budget - num_rx; - - while (cur_budget--) { - ret = am65_cpsw_nuss_rx_packets(common, flow, cpu); - if (ret) { - if (ret == AM65_CPSW_XDP_REDIRECT) - xdp_redirect = true; - break; - } - num_rx++; - } - - if (num_rx >= budget) + /* process only this flow */ + cur_budget = budget; + while (cur_budget--) { + ret = am65_cpsw_nuss_rx_packets(flow, cpu); + if (ret) { + if (ret == AM65_CPSW_XDP_REDIRECT) + xdp_redirect = true; break; + } + num_rx++; } if (xdp_redirect) @@ -1257,14 +1293,14 @@ static int am65_cpsw_nuss_rx_poll(struct napi_struct *napi_rx, int budget) dev_dbg(common->dev, "%s num_rx:%d %d\n", __func__, num_rx, budget); if (num_rx < budget && napi_complete_done(napi_rx, num_rx)) { - if (common->rx_irq_disabled) { - common->rx_irq_disabled = false; - if (unlikely(common->rx_pace_timeout)) { - hrtimer_start(&common->rx_hrtimer, - ns_to_ktime(common->rx_pace_timeout), + if (flow->irq_disabled) { + flow->irq_disabled = false; + if (unlikely(flow->rx_pace_timeout)) { + hrtimer_start(&flow->rx_hrtimer, + ns_to_ktime(flow->rx_pace_timeout), HRTIMER_MODE_REL_PINNED); } else { - enable_irq(common->rx_chns.irq); + enable_irq(flow->irq); } } } @@ -1512,11 +1548,11 @@ static int am65_cpsw_nuss_tx_poll(struct napi_struct *napi_tx, int budget) static irqreturn_t am65_cpsw_nuss_rx_irq(int irq, void *dev_id) { - struct am65_cpsw_common *common = dev_id; + struct am65_cpsw_rx_flow *flow = dev_id; - common->rx_irq_disabled = true; + flow->irq_disabled = true; disable_irq_nosync(irq); - napi_schedule(&common->napi_rx); + napi_schedule(&flow->napi_rx); return IRQ_HANDLED; } @@ -2315,19 +2351,23 @@ static void am65_cpsw_nuss_free_rx_chns(void *data) k3_udma_glue_release_rx_chn(rx_chn->rx_chn); } -static void am65_cpsw_nuss_remove_rx_chns(void *data) +void am65_cpsw_nuss_remove_rx_chns(void *data) { struct am65_cpsw_common *common = data; struct device *dev = common->dev; struct am65_cpsw_rx_chn *rx_chn; + struct am65_cpsw_rx_flow *flows; + int i; rx_chn = &common->rx_chns; + flows = rx_chn->flows; devm_remove_action(dev, am65_cpsw_nuss_free_rx_chns, common); - if (!(rx_chn->irq < 0)) - devm_free_irq(dev, rx_chn->irq, common); - - netif_napi_del(&common->napi_rx); + for (i = 0; i < common->rx_ch_num_flows; i++) { + if (!(flows[i].irq < 0)) + devm_free_irq(dev, flows[i].irq, &flows[i]); + netif_napi_del(&flows[i].napi_rx); + } am65_cpsw_nuss_free_rx_chns(common); @@ -2340,6 +2380,7 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) struct k3_udma_glue_rx_channel_cfg rx_cfg = { 0 }; u32 max_desc_num = AM65_CPSW_MAX_RX_DESC; struct device *dev = common->dev; + struct am65_cpsw_rx_flow *flow; u32 hdesc_size, hdesc_size_out; u32 fdqring_id; int i, ret = 0; @@ -2348,12 +2389,21 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) AM65_CPSW_NAV_SW_DATA_SIZE); rx_cfg.swdata_size = AM65_CPSW_NAV_SW_DATA_SIZE; - rx_cfg.flow_id_num = AM65_CPSW_MAX_RX_FLOWS; + rx_cfg.flow_id_num = common->rx_ch_num_flows; rx_cfg.flow_id_base = common->rx_flow_id_base; /* init all flows */ rx_chn->dev = dev; - rx_chn->descs_num = max_desc_num; + rx_chn->descs_num = max_desc_num * rx_cfg.flow_id_num; + + for (i = 0; i < common->rx_ch_num_flows; i++) { + flow = &rx_chn->flows[i]; + flow->page_pool = NULL; + flow->pages = devm_kcalloc(dev, AM65_CPSW_MAX_RX_DESC, + sizeof(*flow->pages), GFP_KERNEL); + if (!flow->pages) + return -ENOMEM; + } rx_chn->rx_chn = k3_udma_glue_request_rx_chn(dev, "rx", &rx_cfg); if (IS_ERR(rx_chn->rx_chn)) { @@ -2376,13 +2426,6 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) rx_chn->dsize_log2 = __fls(hdesc_size_out); WARN_ON(hdesc_size_out != (1 << rx_chn->dsize_log2)); - rx_chn->page_pool = NULL; - - rx_chn->pages = devm_kcalloc(dev, rx_chn->descs_num, - sizeof(*rx_chn->pages), GFP_KERNEL); - if (!rx_chn->pages) - return -ENOMEM; - common->rx_flow_id_base = k3_udma_glue_rx_get_flow_id_base(rx_chn->rx_chn); dev_info(dev, "set new flow-id-base %u\n", common->rx_flow_id_base); @@ -2406,6 +2449,10 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) K3_UDMA_GLUE_SRC_TAG_LO_USE_REMOTE_SRC_TAG, }; + flow = &rx_chn->flows[i]; + flow->id = i; + flow->common = common; + rx_flow_cfg.ring_rxfdq0_id = fdqring_id; rx_flow_cfg.rx_cfg.size = max_desc_num; rx_flow_cfg.rxfdq_cfg.size = max_desc_num; @@ -2422,28 +2469,32 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) k3_udma_glue_rx_flow_get_fdq_id(rx_chn->rx_chn, i); - rx_chn->irq = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i); - - if (rx_chn->irq < 0) { + flow->irq = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i); + if (flow->irq <= 0) { dev_err(dev, "Failed to get rx dma irq %d\n", - rx_chn->irq); - ret = rx_chn->irq; + flow->irq); + ret = flow->irq; goto err; } - } - - netif_napi_add(common->dma_ndev, &common->napi_rx, - am65_cpsw_nuss_rx_poll); - hrtimer_init(&common->rx_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED); - common->rx_hrtimer.function = &am65_cpsw_nuss_rx_timer_callback; - ret = devm_request_irq(dev, rx_chn->irq, - am65_cpsw_nuss_rx_irq, - IRQF_TRIGGER_HIGH, dev_name(dev), common); - if (ret) { - dev_err(dev, "failure requesting rx irq %u, %d\n", - rx_chn->irq, ret); - goto err; + snprintf(flow->name, + sizeof(flow->name), "%s-rx%d", + dev_name(dev), i); + netif_napi_add(common->dma_ndev, &flow->napi_rx, + am65_cpsw_nuss_rx_poll); + hrtimer_init(&flow->rx_hrtimer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_PINNED); + flow->rx_hrtimer.function = &am65_cpsw_nuss_rx_timer_callback; + + ret = devm_request_irq(dev, flow->irq, + am65_cpsw_nuss_rx_irq, + IRQF_TRIGGER_HIGH, + flow->name, flow); + if (ret) { + dev_err(dev, "failure requesting rx %d irq %u, %d\n", + i, flow->irq, ret); + goto err; + } } err: @@ -3287,8 +3338,9 @@ static int am65_cpsw_nuss_register_ndevs(struct am65_cpsw_common *common) k3_udma_glue_disable_tx_chn(tx_chan[i].tx_chn); } - for (i = 0; i < AM65_CPSW_MAX_RX_FLOWS; i++) - k3_udma_glue_reset_rx_chn(rx_chan->rx_chn, i, rx_chan, + for (i = 0; i < common->rx_ch_num_flows; i++) + k3_udma_glue_reset_rx_chn(rx_chan->rx_chn, i, + &rx_chan->flows[i], am65_cpsw_nuss_rx_cleanup, !!i); k3_udma_glue_disable_rx_chn(rx_chan->rx_chn); @@ -3330,12 +3382,18 @@ static int am65_cpsw_nuss_register_ndevs(struct am65_cpsw_common *common) return ret; } -int am65_cpsw_nuss_update_tx_chns(struct am65_cpsw_common *common, int num_tx) +int am65_cpsw_nuss_update_tx_rx_chns(struct am65_cpsw_common *common, + int num_tx, int num_rx) { int ret; common->tx_ch_num = num_tx; + common->rx_ch_num_flows = num_rx; ret = am65_cpsw_nuss_init_tx_chns(common); + if (ret) + return ret; + + ret = am65_cpsw_nuss_init_rx_chns(common); return ret; } @@ -3465,6 +3523,7 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) common->rx_flow_id_base = -1; init_completion(&common->tdown_complete); common->tx_ch_num = AM65_CPSW_DEFAULT_TX_CHNS; + common->rx_ch_num_flows = AM65_CPSW_DEFAULT_RX_CHN_FLOWS; common->pf_p0_rx_ptype_rrobin = false; common->default_vlan = 1; @@ -3656,8 +3715,10 @@ static int am65_cpsw_nuss_resume(struct device *dev) return ret; /* If RX IRQ was disabled before suspend, keep it disabled */ - if (common->rx_irq_disabled) - disable_irq(common->rx_chns.irq); + for (i = 0; i < common->rx_ch_num_flows; i++) { + if (common->rx_chns.flows[i].irq_disabled) + disable_irq(common->rx_chns.flows[i].irq); + } am65_cpts_resume(common->cpts); diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h index e2ce2be320bd..7528bf7ab3fb 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.h +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h @@ -22,8 +22,7 @@ struct am65_cpts; #define HOST_PORT_NUM 0 #define AM65_CPSW_MAX_TX_QUEUES 8 -#define AM65_CPSW_MAX_RX_QUEUES 1 -#define AM65_CPSW_MAX_RX_FLOWS 1 +#define AM65_CPSW_MAX_RX_QUEUES 8 #define AM65_CPSW_PORT_VLAN_REG_OFFSET 0x014 @@ -58,7 +57,7 @@ struct am65_cpsw_port { struct am65_cpsw_qos qos; struct devlink_port devlink_port; struct bpf_prog *xdp_prog; - struct xdp_rxq_info xdp_rxq; + struct xdp_rxq_info xdp_rxq[AM65_CPSW_MAX_RX_QUEUES]; /* Only for suspend resume context */ u32 vid_context; }; @@ -94,16 +93,27 @@ struct am65_cpsw_tx_chn { u32 rate_mbps; }; +struct am65_cpsw_rx_flow { + u32 id; + struct napi_struct napi_rx; + struct am65_cpsw_common *common; + int irq; + bool irq_disabled; + struct hrtimer rx_hrtimer; + unsigned long rx_pace_timeout; + struct page_pool *page_pool; + struct page **pages; + char name[32]; +}; + struct am65_cpsw_rx_chn { struct device *dev; struct device *dma_dev; struct k3_cppi_desc_pool *desc_pool; struct k3_udma_glue_rx_channel *rx_chn; - struct page_pool *page_pool; - struct page **pages; u32 descs_num; unsigned char dsize_log2; - int irq; + struct am65_cpsw_rx_flow flows[AM65_CPSW_MAX_RX_QUEUES]; }; #define AM65_CPSW_QUIRK_I2027_NO_TX_CSUM BIT(0) @@ -149,12 +159,8 @@ struct am65_cpsw_common { struct completion tdown_complete; atomic_t tdown_cnt; + int rx_ch_num_flows; struct am65_cpsw_rx_chn rx_chns; - struct napi_struct napi_rx; - - bool rx_irq_disabled; - struct hrtimer rx_hrtimer; - unsigned long rx_pace_timeout; u32 nuss_ver; u32 cpsw_ver; @@ -203,8 +209,8 @@ struct am65_cpsw_ndev_priv { #define am65_common_get_host(common) (&(common)->host) #define am65_common_get_port(common, id) (&(common)->ports[(id) - 1]) -#define am65_cpsw_napi_to_common(pnapi) \ - container_of(pnapi, struct am65_cpsw_common, napi_rx) +#define am65_cpsw_napi_to_rx_flow(pnapi) \ + container_of(pnapi, struct am65_cpsw_rx_flow, napi_rx) #define am65_cpsw_napi_to_tx_chn(pnapi) \ container_of(pnapi, struct am65_cpsw_tx_chn, napi_tx) @@ -216,7 +222,9 @@ extern const struct ethtool_ops am65_cpsw_ethtool_ops_slave; void am65_cpsw_nuss_set_p0_ptype(struct am65_cpsw_common *common); void am65_cpsw_nuss_remove_tx_chns(struct am65_cpsw_common *common); -int am65_cpsw_nuss_update_tx_chns(struct am65_cpsw_common *common, int num_tx); +void am65_cpsw_nuss_remove_rx_chns(void *data); +int am65_cpsw_nuss_update_tx_rx_chns(struct am65_cpsw_common *common, + int num_tx, int num_rx); bool am65_cpsw_port_dev_check(const struct net_device *dev); From patchwork Wed Jul 3 13:51:33 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 809623 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id CEC5F8C1F; Wed, 3 Jul 2024 13:51:46 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014706; cv=none; b=NnZjG1ROPTXtP1WbJQNz9gEKGyQqsNbwULiZfA+XO5HY9DTchD2RTbbBYBaTIjjQhveifX6oJisYbzxMV7/yCgoNG4Vey9ahWF0Mvtnbq0Ak5fjEWYLWxkQjI+klhxDAZejdZU8H6Dxnuq8FeWl0OMRRV7ICheENWmioQvTsmmc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014706; c=relaxed/simple; bh=O97XTUmrKRR0a5m/khz4qd/xvH7b9hbhH1hyFDxFXD4=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=GDsOrchaz2oVkmPUBE2ZGwtXNW6Cse1ctCy1tfoMwJQ3ZSy8whFd4QPEXoOQp4JozskpDG5061I0nhnnDKVpgNH1/QGV708/5sBgULsqbqqPpAv45BQbK7LGNa/4AxOM07AJ264fKtjIvqpEt36LGJx+aBsTuAnHFejO2iIz90U= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=hp/vm05e; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="hp/vm05e" Received: by smtp.kernel.org (Postfix) with ESMTPSA id F0025C4AF07; Wed, 3 Jul 2024 13:51:42 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014706; bh=O97XTUmrKRR0a5m/khz4qd/xvH7b9hbhH1hyFDxFXD4=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=hp/vm05eVIhEuDpS3+GUrDzRKgUaEqge3zeUJaN14BFR2h+jgiMyGSReQp23Ua7aH 9r59vcM6hGR7USl/npy2lEofMhaFZTGKBenhx/be+eBzsu3MqTgoaDVbUKSp34wsoe jWK/p7Rm0TwzoTz2o74i8dznIGcPFdvWh8zDpyriz1mfLfA4/FaIPhZ7GiYxEkz8RM 1GFrxYckTwiXpfsTkxe1aG9cAuB7dp58Vcm1Glh216N76iTUGZimE3rqCLCe+lvusy w0F9zmELJOctHF7OfyzZCOrovFI6ahK8Wlll6DOQyqTozRlEZBYD/D/YlVvZq04cd+ Le2ehYB1unObw== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:33 +0300 Subject: [PATCH net-next v3 2/6] net: ethernet: ti: cpsw_ale: use regfields for ALE registers Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-2-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=7543; i=rogerq@kernel.org; h=from:subject:message-id; bh=O97XTUmrKRR0a5m/khz4qd/xvH7b9hbhH1hyFDxFXD4=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdmK8+jEmeV4YmSKuRFEhuRLYY8IQQqOgjFq qjUddkjwOWJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw k/KyD/41nQ3qAhPYGdwyw72oyJlsfMTne28Rllq+33UZHjJlapEWA3rP905eU/AHIw9PcPb6Ewl PYCg5e9dayoRKnkG5a2O9AMsuWoBt0xjFTIyre6czDHIHI9/ukoQQBkgPuDAGGqOQC+qAESuh7B o6lSvVRQxi6jdd1ICcOb9z9gZ5SdkQamvaQWQP3N9qkNccBB5VhR7HNIRdBVntexEMS/2YCqX3R VujSocwZ3y22WBjXCQL+vguJMFqL+mrP1qokUMMzmvx826hb6LFeDRgqX6xdci5C8l27yRZ9+Uq wKDjqOxi11TawS+p/P9jd2ta9igJkALPC6MhN3IMhW7bMljDHVALItdxnOVF76IFskhqFpEx2Bx C7i7pUtAtE04P4ZUeslEtwGP0eO2KU/949YMR70SxKWyX4g75TIP4gk0JuxXE22n17Kf4V1hqWM p6nWP48wmv2D/zjJmuEi0IwyGgh5zUwYIcnaGoa4BHl4cnmGmXildnojb24ENliGJuMnf1dEien vYHIpo1Q5q0t+KzhClxwDCAcQJqmr+lpjknyysAskyVl3vQQ5c9UeNXtfvKeRyw3QSRKCsVDn4P QQZFxLm/noJCAJKX26zFJreE00UhMDKpnn2neMRRjCd6HaXk9OogiS27AXi/N3QA7XWAvLy067C 2G12I6qsEk4ZviQ== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 Map the entire ALE registerspace using regmap. Add regfields for Major and Minor Version fields. Signed-off-by: Roger Quadros Reviewed-by: Simon Horman --- Changelog: v3: - added Reviewed-by Simon Horman --- drivers/net/ethernet/ti/cpsw_ale.c | 84 ++++++++++++++++++++++++++++++-------- drivers/net/ethernet/ti/cpsw_ale.h | 17 ++++++-- 2 files changed, 79 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 64bf22cd860c..f638aa63edee 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -76,7 +77,7 @@ enum { * @dev_id: ALE version/SoC id * @features: features supported by ALE * @tbl_entries: number of ALE entries - * @major_ver_mask: mask of ALE Major Version Value in ALE_IDVER reg. + * @reg_fields: pointer to array of register field configuration * @nu_switch_ale: NU Switch ALE * @vlan_entry_tbl: ALE vlan entry fields description tbl */ @@ -84,7 +85,7 @@ struct cpsw_ale_dev_id { const char *dev_id; u32 features; u32 tbl_entries; - u32 major_ver_mask; + const struct reg_field *reg_fields; bool nu_switch_ale; const struct ale_entry_fld *vlan_entry_tbl; }; @@ -1292,25 +1293,37 @@ void cpsw_ale_stop(struct cpsw_ale *ale) cpsw_ale_control_set(ale, 0, ALE_ENABLE, 0); } +static const struct reg_field ale_fields_cpsw[] = { + /* CPSW_ALE_IDVER_REG */ + [MINOR_VER] = REG_FIELD(ALE_IDVER, 0, 7), + [MAJOR_VER] = REG_FIELD(ALE_IDVER, 8, 15), +}; + +static const struct reg_field ale_fields_cpsw_nu[] = { + /* CPSW_ALE_IDVER_REG */ + [MINOR_VER] = REG_FIELD(ALE_IDVER, 0, 7), + [MAJOR_VER] = REG_FIELD(ALE_IDVER, 8, 10), +}; + static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { { /* am3/4/5, dra7. dm814x, 66ak2hk-gbe */ .dev_id = "cpsw", .tbl_entries = 1024, - .major_ver_mask = 0xff, + .reg_fields = ale_fields_cpsw, .vlan_entry_tbl = vlan_entry_cpsw, }, { /* 66ak2h_xgbe */ .dev_id = "66ak2h-xgbe", .tbl_entries = 2048, - .major_ver_mask = 0xff, + .reg_fields = ale_fields_cpsw, .vlan_entry_tbl = vlan_entry_cpsw, }, { .dev_id = "66ak2el", .features = CPSW_ALE_F_STATUS_REG, - .major_ver_mask = 0x7, + .reg_fields = ale_fields_cpsw_nu, .nu_switch_ale = true, .vlan_entry_tbl = vlan_entry_nu, }, @@ -1318,7 +1331,7 @@ static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { .dev_id = "66ak2g", .features = CPSW_ALE_F_STATUS_REG, .tbl_entries = 64, - .major_ver_mask = 0x7, + .reg_fields = ale_fields_cpsw_nu, .nu_switch_ale = true, .vlan_entry_tbl = vlan_entry_nu, }, @@ -1326,20 +1339,20 @@ static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { .dev_id = "am65x-cpsw2g", .features = CPSW_ALE_F_STATUS_REG | CPSW_ALE_F_HW_AUTOAGING, .tbl_entries = 64, - .major_ver_mask = 0x7, + .reg_fields = ale_fields_cpsw_nu, .nu_switch_ale = true, .vlan_entry_tbl = vlan_entry_nu, }, { .dev_id = "j721e-cpswxg", .features = CPSW_ALE_F_STATUS_REG | CPSW_ALE_F_HW_AUTOAGING, - .major_ver_mask = 0x7, + .reg_fields = ale_fields_cpsw_nu, .vlan_entry_tbl = vlan_entry_k3_cpswxg, }, { .dev_id = "am64-cpswxg", .features = CPSW_ALE_F_STATUS_REG | CPSW_ALE_F_HW_AUTOAGING, - .major_ver_mask = 0x7, + .reg_fields = ale_fields_cpsw_nu, .vlan_entry_tbl = vlan_entry_k3_cpswxg, .tbl_entries = 512, }, @@ -1361,41 +1374,76 @@ cpsw_ale_dev_id *cpsw_ale_match_id(const struct cpsw_ale_dev_id *id, return NULL; } +static const struct regmap_config ale_regmap_cfg = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 4, + .name = "cpsw-ale", +}; + +static int cpsw_ale_regfield_init(struct cpsw_ale *ale) +{ + struct regmap *regmap = ale->regmap; + struct device *dev = ale->params.dev; + const struct reg_field *reg_fields = ale->params.reg_fields; + int i; + + for (i = 0; i < ALE_FIELDS_MAX; i++) { + ale->fields[i] = devm_regmap_field_alloc(dev, regmap, + reg_fields[i]); + if (IS_ERR(ale->fields[i])) { + dev_err(dev, "Unable to allocate regmap field %d\n", i); + return PTR_ERR(ale->fields[i]); + } + } + + return 0; +} + struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) { const struct cpsw_ale_dev_id *ale_dev_id; struct cpsw_ale *ale; - u32 rev, ale_entries; + u32 ale_entries, rev_major, rev_minor; + int ret; ale_dev_id = cpsw_ale_match_id(cpsw_ale_id_match, params->dev_id); if (!ale_dev_id) return ERR_PTR(-EINVAL); params->ale_entries = ale_dev_id->tbl_entries; - params->major_ver_mask = ale_dev_id->major_ver_mask; params->nu_switch_ale = ale_dev_id->nu_switch_ale; + params->reg_fields = ale_dev_id->reg_fields; ale = devm_kzalloc(params->dev, sizeof(*ale), GFP_KERNEL); if (!ale) return ERR_PTR(-ENOMEM); + ale->regmap = devm_regmap_init_mmio(params->dev, params->ale_regs, + &ale_regmap_cfg); + if (IS_ERR(ale->regmap)) { + dev_err(params->dev, "Couldn't create CPSW ALE regmap\n"); + return ERR_PTR(-ENOMEM); + } + + ale->params = *params; + ret = cpsw_ale_regfield_init(ale); + if (ret) + return ERR_PTR(ret); ale->p0_untag_vid_mask = devm_bitmap_zalloc(params->dev, VLAN_N_VID, GFP_KERNEL); if (!ale->p0_untag_vid_mask) return ERR_PTR(-ENOMEM); - ale->params = *params; ale->ageout = ale->params.ale_ageout * HZ; ale->features = ale_dev_id->features; ale->vlan_entry_tbl = ale_dev_id->vlan_entry_tbl; - rev = readl_relaxed(ale->params.ale_regs + ALE_IDVER); - ale->version = - (ALE_VERSION_MAJOR(rev, ale->params.major_ver_mask) << 8) | - ALE_VERSION_MINOR(rev); + regmap_field_read(ale->fields[MINOR_VER], &rev_minor); + regmap_field_read(ale->fields[MAJOR_VER], &rev_major); + ale->version = rev_major << 8 | rev_minor; dev_info(ale->params.dev, "initialized cpsw ale version %d.%d\n", - ALE_VERSION_MAJOR(rev, ale->params.major_ver_mask), - ALE_VERSION_MINOR(rev)); + rev_major, rev_minor); if (ale->features & CPSW_ALE_F_STATUS_REG && !ale->params.ale_entries) { diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index 6779ee111d57..58d377dd7496 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -8,6 +8,8 @@ #ifndef __TI_CPSW_ALE_H__ #define __TI_CPSW_ALE_H__ +struct reg_fields; + struct cpsw_ale_params { struct device *dev; void __iomem *ale_regs; @@ -20,19 +22,26 @@ struct cpsw_ale_params { * to identify this hardware. */ bool nu_switch_ale; - /* mask bit used in NU Switch ALE is 3 bits instead of 8 bits. So - * pass it from caller. - */ - u32 major_ver_mask; + const struct reg_field *reg_fields; const char *dev_id; unsigned long bus_freq; }; struct ale_entry_fld; +struct regmap; + +enum ale_fields { + MINOR_VER, + MAJOR_VER, + /* terminator */ + ALE_FIELDS_MAX, +}; struct cpsw_ale { struct cpsw_ale_params params; struct timer_list timer; + struct regmap *regmap; + struct regmap_field *fields[ALE_FIELDS_MAX]; unsigned long ageout; u32 version; u32 features; From patchwork Wed Jul 3 13:51:34 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 810619 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6422617CA1B; Wed, 3 Jul 2024 13:51:50 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014710; cv=none; b=p4X23ix6dhCeMq7Jk3k1QGsZssSxOBusdhcdrIRNt3h83DflBakF/U4ZO+lHTjD5UjNWOB+3xoSdtEH/3jyEbywt1Ga4fQ7WZ6PVPYkykwtnPDxnPcJ2/DpZwbcRrYua9UhyBGmN2TOgzvVf1KfgSrSU9D38Hbb60BRhxSgdCNg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014710; c=relaxed/simple; bh=JeOJtLTbcGj6xPCF5ZeTJIWS2mvZBY3kd+RN0BWRcHw=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=Jqs3VOPeqNmMfJbDZWMKbgrUo+dXQ8hfj2cQBOWIISVQpLS6yE7qMxXxVFtq3m5clt66vI5gOOYBWsXoiW7HWDVnWYeE6llvqhrRB+k0yUeT0Hw4c09mKAki31uAIay8tlpQ099bKSrlul8VpztesTFwigexpqSX4Cj7GHKGdCU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=gNBaVU8D; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="gNBaVU8D" Received: by smtp.kernel.org (Postfix) with ESMTPSA id D6238C3277B; Wed, 3 Jul 2024 13:51:46 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014710; bh=JeOJtLTbcGj6xPCF5ZeTJIWS2mvZBY3kd+RN0BWRcHw=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=gNBaVU8D0mDvFt6w8npig95oTJ2jxtngLd2KOuK7E0kGliU3wQdS0XSI9TvBT2+p1 JruK6CkjAhxfEBK1eV9VJ4Sq4DxIUjenab0EgQK3PGK0A31G+5tRRth8GdWthWLElx 1vo457SGZScAa0TFpOCWmHC4B4eSj/XcLA7XOChbUERDNQ5bDAULabsWK535aZoZng LrMjN62Z1sTC3ba2oIvQR1oxlPpqs9nEqyoLbaT6Epp65xrsyLzWYrBksqpm8FGQoD hd3xMVuh0/j+Cwnnvosr8/enB7TN+f2A0JRhLT3NyV2rGEXAMe3d2DDA1P1WNwyKQ0 oYxPsEfDGRDwQ== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:34 +0300 Subject: [PATCH net-next v3 3/6] net: ethernet: ti: cpsw_ale: use regfields for number of Entries and Policers Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-3-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=3806; i=rogerq@kernel.org; h=from:subject:message-id; bh=JeOJtLTbcGj6xPCF5ZeTJIWS2mvZBY3kd+RN0BWRcHw=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdmHFzbCImmd3YtgVwaLH7A3aNwQHJAzOSMN iwFvozW3cmJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw k9FAD/9+N/+nHYj61/piisJP0F6f0fS28Ou1ha5bVV7Fw823mdKTn9DgBtEsXykRxWF1qPplDXr XibWml1N3NxnVGdAzcPQs34A8X+nCqwebHCc6knzBbH+RySxS4XcDq8Rnt6XWACZRx6koFJMbAz +/ymcqXNar7EudTPp2cYjv3EKXCOvWjGY3k/Pqch671LlhOr7UlOascyyivlycfHvR8TyXTWvoR AnbLKLIVlvYgLhMs/+3ZV1V+njOBCh3SzZgHqsomWDh3W09Hxru4ZTAyVeo4MgiS117MGYwf067 QTuvfJf1zHDyQUXhmkjX2usFcvHUL70SgiDbNZXYH1cIcIzaDjDysnHslmGldnyPaoyEDW4F62k dbWCKpVL3mM+HkDHKgcVNO0VT2ZoUoN7XZdEFRUY6fGG9o5KPaWYgcJyclMiq2C1m/vjSoz+t7D iEOmFTgNXDV7qrORFjBIyLWMnXHKKZdwAH42GSfDtO/17WKuIcyPcjWQ9vsau6IBiI3NcWjO2tI f8f0JC6GeIKzv5SR9ho0mh6vw34i0daqoloJR9wQ3hOXeBE+De5Ch284I+Cn0hflpGj2Qz8Oife E9f/7PHcXM/oRj/RzHOqQFDvnAqYPogtEkbXYS1ZsxJjb0g7CEmCzlY5JDj3XhTIX1nCKscBvBD bx+zKwX9PqAPMpA== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 Use regfields for number of ALE Entries and Policers. The variants that support Policers/Classifiers have the number of policers encoded in the ALE_STATUS register. Use that and show the number of Policers in the ALE info message. Signed-off-by: Roger Quadros Reviewed-by: Simon Horman --- Changelog: v3: - added Reviewed-by Simon Horman --- drivers/net/ethernet/ti/cpsw_ale.c | 25 +++++++++++++++++++------ drivers/net/ethernet/ti/cpsw_ale.h | 3 +++ 2 files changed, 22 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index f638aa63edee..c11326e8fbd4 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -103,7 +103,7 @@ struct cpsw_ale_dev_id { #define ALE_UCAST_TOUCHED 3 #define ALE_TABLE_SIZE_MULTIPLIER 1024 -#define ALE_STATUS_SIZE_MASK 0x1f +#define ALE_POLICER_SIZE_MULTIPLIER 8 static inline int cpsw_ale_get_field(u32 *ale_entry, u32 start, u32 bits) { @@ -1303,6 +1303,9 @@ static const struct reg_field ale_fields_cpsw_nu[] = { /* CPSW_ALE_IDVER_REG */ [MINOR_VER] = REG_FIELD(ALE_IDVER, 0, 7), [MAJOR_VER] = REG_FIELD(ALE_IDVER, 8, 10), + /* CPSW_ALE_STATUS_REG */ + [ALE_ENTRIES] = REG_FIELD(ALE_STATUS, 0, 7), + [ALE_POLICERS] = REG_FIELD(ALE_STATUS, 8, 15), }; static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { @@ -1404,7 +1407,7 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) { const struct cpsw_ale_dev_id *ale_dev_id; struct cpsw_ale *ale; - u32 ale_entries, rev_major, rev_minor; + u32 ale_entries, rev_major, rev_minor, policers; int ret; ale_dev_id = cpsw_ale_match_id(cpsw_ale_id_match, params->dev_id); @@ -1447,9 +1450,7 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) if (ale->features & CPSW_ALE_F_STATUS_REG && !ale->params.ale_entries) { - ale_entries = - readl_relaxed(ale->params.ale_regs + ALE_STATUS) & - ALE_STATUS_SIZE_MASK; + regmap_field_read(ale->fields[ALE_ENTRIES], &ale_entries); /* ALE available on newer NetCP switches has introduced * a register, ALE_STATUS, to indicate the size of ALE * table which shows the size as a multiple of 1024 entries. @@ -1463,8 +1464,20 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) ale_entries *= ALE_TABLE_SIZE_MULTIPLIER; ale->params.ale_entries = ale_entries; } + + if (ale->features & CPSW_ALE_F_STATUS_REG && + !ale->params.num_policers) { + regmap_field_read(ale->fields[ALE_POLICERS], &policers); + if (!policers) + return ERR_PTR(-EINVAL); + + policers *= ALE_POLICER_SIZE_MULTIPLIER; + ale->params.num_policers = policers; + } + dev_info(ale->params.dev, - "ALE Table size %ld\n", ale->params.ale_entries); + "ALE Table size %ld, Policers %ld\n", ale->params.ale_entries, + ale->params.num_policers); /* set default bits for existing h/w */ ale->port_mask_bits = ale->params.ale_ports; diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index 58d377dd7496..e12bb2caf016 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -15,6 +15,7 @@ struct cpsw_ale_params { void __iomem *ale_regs; unsigned long ale_ageout; /* in secs */ unsigned long ale_entries; + unsigned long num_policers; unsigned long ale_ports; /* NU Switch has specific handling as number of bits in ALE entries * are different than other versions of ALE. Also there are specific @@ -33,6 +34,8 @@ struct regmap; enum ale_fields { MINOR_VER, MAJOR_VER, + ALE_ENTRIES, + ALE_POLICERS, /* terminator */ ALE_FIELDS_MAX, }; From patchwork Wed Jul 3 13:51:35 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 809622 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id A5D4E17B4E8; Wed, 3 Jul 2024 13:51:54 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014714; cv=none; b=DU8+zclThgDQMACQNGzpEc/IQeR1DZP/TqZNSW1UaXIcs0Q+infvUtQwpxoiZgkXu5xosNUHAEg1gV/s7dwSimoU78BdrtMsN3PsKsjbsQF+PnpFFwvWhYaxXaz3rgmRYlTJbVoDDQzBkMvRYXcXRS3vTBgT2CqOzK/PYvDXUbA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014714; c=relaxed/simple; bh=89MKu2GvUohqdOqYrtP/c08hc+UbOCEVHc8lLtk0VUY=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=qDtOFxJDwHB1Xuqs2/4JfpAdF0LpuZRdkNlbq3G7KUWEAz+1HfAZqJi1v/9ZrQLPDVIBVEl5xQZK/zhPeSCcLY0Gt0J8efNhhShnjfs6cASWStf35bc3zIPtqbcRdrupkgzAFHcwhyZUw+n7pa88S++sEr00A+xHpaKTXezSGtM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=CSlBgZwi; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="CSlBgZwi" Received: by smtp.kernel.org (Postfix) with ESMTPSA id BDEC4C3277B; Wed, 3 Jul 2024 13:51:50 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014714; bh=89MKu2GvUohqdOqYrtP/c08hc+UbOCEVHc8lLtk0VUY=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=CSlBgZwihwufdSqFGr/HONY6bRS8XPNgGXgoqOzjtE56Y8MEJOixXLrYWF4H/9IgR QZSg+PycChruRthJD20su23Q0V0WfsfSr3lKF+6VQPpG9+14wGVDYdBAlgAB4++qRO 0FcZMsxKoBMAjkE5Z9HMkpJGC88ZkzNjkbFUa5maUVmdhHOXcS0b/Wb6Bhb5zcrDud WKzGNKrUe1umowXf6KJENKiM6GooaB6qHxicT2IhE0w+9RYkxkYCaDD6z6ARPTe2et Pdy3JKb1mJcG4Nx0LRcU7Ukjd5V4lzn/dth2J3kMB2j+ETCpWmlGH4ySYQ8CKalW1T Td6IX/j9cuuUA== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:35 +0300 Subject: [PATCH net-next v3 4/6] net: ethernet: ti: cpsw_ale: add Policer and Thread control register fields Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-4-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=6034; i=rogerq@kernel.org; h=from:subject:message-id; bh=89MKu2GvUohqdOqYrtP/c08hc+UbOCEVHc8lLtk0VUY=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdmWyFr3GPd/iE4fnYocDShrr4s2jVUzTT2m SaPmIbubRWJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw k8HGEACxhD0FRGey9+O6ciW6Aubnc/TiAof/RI93DiAF907gMRfgO6tr47Kim0MXi5MGLGtAvwa uJQn2HOZ5lJcKWT0NV6AiBnUz84OQpgCZt/kzgdZtGqn1xBY/X0tM7s9Zqf7a7NLv+ynHntVLAR 0sgz40j1W6PE7QW74++RWQJr5ldnYzN+oDTmfeI45zJXkoy8kpgqyyRiseI5PKh0yBwfphjK+yD ev3nzPOzQEwrBesE1akrDKcCjd298wvT8oxDgHM2MpAiV8e5CCDPXmd9oM7rMh3pB6JaUWEUWIW nQqKReE4dNsG8gfId9sCNge+n87pCUdKbJ58du7LrBXMjsSyvR0g1DqN54o15MpcUNMO3C4PTmx qHV93E3CHYZXN6qCD2yMcCkgPGWI+zAGgsJ06UOFC9ltc43rtB1Dz4DoTpKf5avOmdKu0DxG6BB oKAsdfKnwIYqwEXmVBrrK0J62jX+HfS0AU/hAf/+Z7nnKea0XmBOnJBwhSI7j4r44Z0cSr8g4UV xhCpnLXqtO0hcpN2WGWVauX6mVT3JrzVPeliRzYOwKFrlp8o6KiS3knBHsvP72LE5yfQNRl12Rb T/NS9fIR6nCUuuZreO8V9LPbJTKpiJdIdO66Mq9Dm9C9StD04DjxUrlqh8O5lLmFx9fZQmstAj2 2zOEQMhL5OE+f7A== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 Adds regfileds for Policer registers and Thread mapping/control registers. Signed-off-by: Roger Quadros Reviewed-by: Simon Horman --- Changelog: v3: - added Reviewed-by Simon Horman --- drivers/net/ethernet/ti/cpsw_ale.c | 86 ++++++++++++++++++++++++++++++++++++++ drivers/net/ethernet/ti/cpsw_ale.h | 41 ++++++++++++++++++ 2 files changed, 127 insertions(+) diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index c11326e8fbd4..6b64ec78d1b6 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -46,6 +46,24 @@ #define ALE_UNKNOWNVLAN_FORCE_UNTAG_EGRESS 0x9C #define ALE_VLAN_MASK_MUX(reg) (0xc0 + (0x4 * (reg))) +#define ALE_POLICER_PORT_OUI 0x100 +#define ALE_POLICER_DA_SA 0x104 +#define ALE_POLICER_VLAN 0x108 +#define ALE_POLICER_ETHERTYPE_IPSA 0x10c +#define ALE_POLICER_IPDA 0x110 +#define ALE_POLICER_PIR 0x118 +#define ALE_POLICER_CIR 0x11c +#define ALE_POLICER_TBL_CTL 0x120 +#define ALE_POLICER_CTL 0x124 +#define ALE_POLICER_TEST_CTL 0x128 +#define ALE_POLICER_HIT_STATUS 0x12c +#define ALE_THREAD_DEF 0x134 +#define ALE_THREAD_CTL 0x138 +#define ALE_THREAD_VAL 0x13c + +#define ALE_POLICER_TBL_WRITE_ENABLE BIT(31) +#define ALE_POLICER_TBL_INDEX_MASK GENMASK(4, 0) + #define AM65_CPSW_ALE_THREAD_DEF_REG 0x134 /* ALE_AGING_TIMER */ @@ -1306,6 +1324,74 @@ static const struct reg_field ale_fields_cpsw_nu[] = { /* CPSW_ALE_STATUS_REG */ [ALE_ENTRIES] = REG_FIELD(ALE_STATUS, 0, 7), [ALE_POLICERS] = REG_FIELD(ALE_STATUS, 8, 15), + /* CPSW_ALE_POLICER_PORT_OUI_REG */ + [POL_PORT_MEN] = REG_FIELD(ALE_POLICER_PORT_OUI, 31, 31), + [POL_TRUNK_ID] = REG_FIELD(ALE_POLICER_PORT_OUI, 30, 30), + [POL_PORT_NUM] = REG_FIELD(ALE_POLICER_PORT_OUI, 25, 25), + [POL_PRI_MEN] = REG_FIELD(ALE_POLICER_PORT_OUI, 19, 19), + [POL_PRI_VAL] = REG_FIELD(ALE_POLICER_PORT_OUI, 16, 18), + [POL_OUI_MEN] = REG_FIELD(ALE_POLICER_PORT_OUI, 15, 15), + [POL_OUI_INDEX] = REG_FIELD(ALE_POLICER_PORT_OUI, 0, 5), + + /* CPSW_ALE_POLICER_DA_SA_REG */ + [POL_DST_MEN] = REG_FIELD(ALE_POLICER_DA_SA, 31, 31), + [POL_DST_INDEX] = REG_FIELD(ALE_POLICER_DA_SA, 16, 21), + [POL_SRC_MEN] = REG_FIELD(ALE_POLICER_DA_SA, 15, 15), + [POL_SRC_INDEX] = REG_FIELD(ALE_POLICER_DA_SA, 0, 5), + + /* CPSW_ALE_POLICER_VLAN_REG */ + [POL_OVLAN_MEN] = REG_FIELD(ALE_POLICER_VLAN, 31, 31), + [POL_OVLAN_INDEX] = REG_FIELD(ALE_POLICER_VLAN, 16, 21), + [POL_IVLAN_MEN] = REG_FIELD(ALE_POLICER_VLAN, 15, 15), + [POL_IVLAN_INDEX] = REG_FIELD(ALE_POLICER_VLAN, 0, 5), + + /* CPSW_ALE_POLICER_ETHERTYPE_IPSA_REG */ + [POL_ETHERTYPE_MEN] = REG_FIELD(ALE_POLICER_ETHERTYPE_IPSA, 31, 31), + [POL_ETHERTYPE_INDEX] = REG_FIELD(ALE_POLICER_ETHERTYPE_IPSA, 16, 21), + [POL_IPSRC_MEN] = REG_FIELD(ALE_POLICER_ETHERTYPE_IPSA, 15, 15), + [POL_IPSRC_INDEX] = REG_FIELD(ALE_POLICER_ETHERTYPE_IPSA, 0, 5), + + /* CPSW_ALE_POLICER_IPDA_REG */ + [POL_IPDST_MEN] = REG_FIELD(ALE_POLICER_IPDA, 31, 31), + [POL_IPDST_INDEX] = REG_FIELD(ALE_POLICER_IPDA, 16, 21), + + /* CPSW_ALE_POLICER_TBL_CTL_REG */ + /** + * REG_FIELDS not defined for this as fields cannot be correctly + * used independently + */ + + /* CPSW_ALE_POLICER_CTL_REG */ + [POL_EN] = REG_FIELD(ALE_POLICER_CTL, 31, 31), + [POL_RED_DROP_EN] = REG_FIELD(ALE_POLICER_CTL, 29, 29), + [POL_YELLOW_DROP_EN] = REG_FIELD(ALE_POLICER_CTL, 28, 28), + [POL_YELLOW_THRESH] = REG_FIELD(ALE_POLICER_CTL, 24, 26), + [POL_POL_MATCH_MODE] = REG_FIELD(ALE_POLICER_CTL, 22, 23), + [POL_PRIORITY_THREAD_EN] = REG_FIELD(ALE_POLICER_CTL, 21, 21), + [POL_MAC_ONLY_DEF_DIS] = REG_FIELD(ALE_POLICER_CTL, 20, 20), + + /* CPSW_ALE_POLICER_TEST_CTL_REG */ + [POL_TEST_CLR] = REG_FIELD(ALE_POLICER_TEST_CTL, 31, 31), + [POL_TEST_CLR_RED] = REG_FIELD(ALE_POLICER_TEST_CTL, 30, 30), + [POL_TEST_CLR_YELLOW] = REG_FIELD(ALE_POLICER_TEST_CTL, 29, 29), + [POL_TEST_CLR_SELECTED] = REG_FIELD(ALE_POLICER_TEST_CTL, 28, 28), + [POL_TEST_ENTRY] = REG_FIELD(ALE_POLICER_TEST_CTL, 0, 4), + + /* CPSW_ALE_POLICER_HIT_STATUS_REG */ + [POL_STATUS_HIT] = REG_FIELD(ALE_POLICER_HIT_STATUS, 31, 31), + [POL_STATUS_HIT_RED] = REG_FIELD(ALE_POLICER_HIT_STATUS, 30, 30), + [POL_STATUS_HIT_YELLOW] = REG_FIELD(ALE_POLICER_HIT_STATUS, 29, 29), + + /* CPSW_ALE_THREAD_DEF_REG */ + [ALE_DEFAULT_THREAD_EN] = REG_FIELD(ALE_THREAD_DEF, 15, 15), + [ALE_DEFAULT_THREAD_VAL] = REG_FIELD(ALE_THREAD_DEF, 0, 5), + + /* CPSW_ALE_THREAD_CTL_REG */ + [ALE_THREAD_CLASS_INDEX] = REG_FIELD(ALE_THREAD_CTL, 0, 4), + + /* CPSW_ALE_THREAD_VAL_REG */ + [ALE_THREAD_ENABLE] = REG_FIELD(ALE_THREAD_VAL, 15, 15), + [ALE_THREAD_VALUE] = REG_FIELD(ALE_THREAD_VAL, 0, 5), }; static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index e12bb2caf016..2cb76acc6d16 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -36,6 +36,47 @@ enum ale_fields { MAJOR_VER, ALE_ENTRIES, ALE_POLICERS, + POL_PORT_MEN, + POL_TRUNK_ID, + POL_PORT_NUM, + POL_PRI_MEN, + POL_PRI_VAL, + POL_OUI_MEN, + POL_OUI_INDEX, + POL_DST_MEN, + POL_DST_INDEX, + POL_SRC_MEN, + POL_SRC_INDEX, + POL_OVLAN_MEN, + POL_OVLAN_INDEX, + POL_IVLAN_MEN, + POL_IVLAN_INDEX, + POL_ETHERTYPE_MEN, + POL_ETHERTYPE_INDEX, + POL_IPSRC_MEN, + POL_IPSRC_INDEX, + POL_IPDST_MEN, + POL_IPDST_INDEX, + POL_EN, + POL_RED_DROP_EN, + POL_YELLOW_DROP_EN, + POL_YELLOW_THRESH, + POL_POL_MATCH_MODE, + POL_PRIORITY_THREAD_EN, + POL_MAC_ONLY_DEF_DIS, + POL_TEST_CLR, + POL_TEST_CLR_RED, + POL_TEST_CLR_YELLOW, + POL_TEST_CLR_SELECTED, + POL_TEST_ENTRY, + POL_STATUS_HIT, + POL_STATUS_HIT_RED, + POL_STATUS_HIT_YELLOW, + ALE_DEFAULT_THREAD_EN, + ALE_DEFAULT_THREAD_VAL, + ALE_THREAD_CLASS_INDEX, + ALE_THREAD_ENABLE, + ALE_THREAD_VALUE, /* terminator */ ALE_FIELDS_MAX, }; From patchwork Wed Jul 3 13:51:36 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 810618 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 432F21802CD; Wed, 3 Jul 2024 13:51:58 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014718; cv=none; b=eK7Wdkilum5ROKGeEYDssGf1FNmZdmw/lPQweL27JM16mEhhZS6cWtkb/k3Aeeqt58hj0Mo452GXsGaFPo3icoaXQM2s366QUKV0VlzmF0gd05kI1JOODEsKDadwNNdKTjy7p3tfqQ8+uzwCN+its54v3WTpJyOgmq8AoOA73JU= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014718; c=relaxed/simple; bh=mieV4TgwWxm0AsrERuGJpKmWHBNjYVtxE4w5wOIW3x0=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=Xh4LJcrkJ6VnvOs0x4X9SfdJj97iSZiLrF3VKP2j39B9VbziJP1S2UTi7JMUqp5a4X6JhbjCJVKIqgVe4/+/HlCNY1I26m/rBhVr7v5ZuX1hg5/kUfnjS4CU6laprsQko/bEav7Ytr/3UTZw8MclTb+n30f8q8g/StiE4zo/PoA= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=JRF0p4ct; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="JRF0p4ct" Received: by smtp.kernel.org (Postfix) with ESMTPSA id A8A7CC2BD10; Wed, 3 Jul 2024 13:51:54 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014718; bh=mieV4TgwWxm0AsrERuGJpKmWHBNjYVtxE4w5wOIW3x0=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=JRF0p4ctrnmG7IOaB9bmYGpJ0zNElleRI8rqIW9rGgZQd17hfJ0d2Nk/xHxAZKag2 rQjN4BJzM6Zq9j5zcdRSfW4PPekyLRIMI+I7hnoTkX3LnzqRhXWA9gjuouYFHyqGty pvraxOy/gZYi2WuCYSxxSjxBsqDfm87EPy/exRDMlpGFs9Y5qqRwFXX0FH9KAhMba3 X+XjUTeXgsnbeW/bJFeg4n4v/O2mH066N5R6A5NTdM0PM+FrFM4ousNMY0XjgEyGuX Q+WW00IrtAhxA42tFePzdJl5uGOVOTJizb9keMPe9qvfdpaqSjPCbtjSaRqAP+hbtD liJXvtkhl9KlA== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:36 +0300 Subject: [PATCH net-next v3 5/6] net: ethernet: ti: cpsw_ale: add policer/classifier helpers and setup defaults Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-5-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=6077; i=rogerq@kernel.org; h=from:subject:message-id; bh=mieV4TgwWxm0AsrERuGJpKmWHBNjYVtxE4w5wOIW3x0=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdm6QGioFiUmmdnTHRBowkKtAJyxhbbH3H6i qrmoU0VoBOJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw kw7iD/9tuZ/aKXDxk6HjMrH7TF7GOzW5smvF3Fxj58nqyHpe4rza1PnrVkuqj8HSFMK5spsydxa i5xMb7KxzfsnrpaHkUcWIEKuw3l7X1quiXqVr6D6E+XRV7QlsnLA4Mxxf4wpT5MuKk5FAsM/+VX fmzP4/bk+WwycNBvASWYqbRIe36nCVVxISfQ8HnXWSQiNFXjMZkRrROWdJE5DH2OThBXpTb9sbe AvuIvUDIiNc+z30C8gQo8z2nV4jCNx0w9qlsFIGoV9FjeQKl8O6QjnTQUv4X2iYHgMmhSk76jQx GV35/m5YzVUR9cQbmSYKaLu48sX+d4CJLnMEANCKTDrVgV8wT5En8JAAQ3TTJbB2r2dPPgOktoX OdVBbGWcnOmkyj9VoKYhwZ/7w9z7oxJaj6xLBae5u6qC2B3KoUkwhr/Ldj5nGNl2DI11Hikw7r9 YflC1CSjXXZDJjeGyH7sLOR56rseRGTCwk1DPCTno47A7RAjsucGflrq6cpN1hK14sPX5w1FgR5 eSuhc/sZRK0G4tTnlHd3CGj/Yr+hNZypd+T6r0iyToJiA2077Lw2eMJhh186yACB32tNakDhB+7 0KVAl603ahXSQ4BsA1JCALd0E5sva6yKoiZ8sMkq8W8EL3Ssz5Bp2Z+9bh5+yBhI3j6bx8l8FKm RTA9CsTO/dOe6Vw== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 The Policer registers in the ALE register space are just shadow registers and use an index field in the policer table control register to read/write to the actual Polier registers. Add helper functions to Read and Write to Policer registers. Also add a helper function to set the thread value to classifier/policer mapping. Any packet that first matches the classifier will be sent to the thread (flow) that is set in the classifier to thread mapping table. If not set then it goes to the default flow. Default behaviour is to have 8 classifiers to map 8 DSCP/PCP priorities to N receive threads (flows). N depends on number of RX channels enabled for the port. As per the standard [1] User prioritie 1 (Background) and 2 (Spare) have lower priority than the user priority 0 (default). User priority 1 being of the lowest priority. [1] IEEE802.1D-2004, IEEE Standard for Local and metropolitan area networks Table G-2 - Traffic type acronyms Table G-3 - Defining traffic types Signed-off-by: Roger Quadros --- Changelog: v3: - squashed 2 patches into 1 - added comment explaining the default thread to priority mapping table - typo fix "classifer"->"classifier" --- drivers/net/ethernet/ti/cpsw_ale.c | 94 ++++++++++++++++++++++++++++++++++++++ drivers/net/ethernet/ti/cpsw_ale.h | 1 + 2 files changed, 95 insertions(+) diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 6b64ec78d1b6..cf2958093fc8 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -1627,3 +1627,97 @@ u32 cpsw_ale_get_num_entries(struct cpsw_ale *ale) { return ale ? ale->params.ale_entries : 0; } + +/* Reads the specified policer index into ALE POLICER registers */ +static void cpsw_ale_policer_read_idx(struct cpsw_ale *ale, u32 idx) +{ + idx &= ALE_POLICER_TBL_INDEX_MASK; + writel_relaxed(idx, ale->params.ale_regs + ALE_POLICER_TBL_CTL); +} + +/* Writes the ALE POLICER registers into the specified policer index */ +static void cpsw_ale_policer_write_idx(struct cpsw_ale *ale, u32 idx) +{ + idx &= ALE_POLICER_TBL_INDEX_MASK; + idx |= ALE_POLICER_TBL_WRITE_ENABLE; + writel_relaxed(idx, ale->params.ale_regs + ALE_POLICER_TBL_CTL); +} + +/* enables/disables the custom thread value for the specified policer index */ +static void cpsw_ale_policer_thread_idx_enable(struct cpsw_ale *ale, u32 idx, + u32 thread_id, bool enable) +{ + regmap_field_write(ale->fields[ALE_THREAD_CLASS_INDEX], idx); + regmap_field_write(ale->fields[ALE_THREAD_VALUE], thread_id); + regmap_field_write(ale->fields[ALE_THREAD_ENABLE], enable ? 1 : 0); +} + +/* Disable all policer entries and thread mappings */ +static void cpsw_ale_policer_reset(struct cpsw_ale *ale) +{ + int i; + + for (i = 0; i < ale->params.num_policers ; i++) { + cpsw_ale_policer_read_idx(ale, i); + regmap_field_write(ale->fields[POL_PORT_MEN], 0); + regmap_field_write(ale->fields[POL_PRI_MEN], 0); + regmap_field_write(ale->fields[POL_OUI_MEN], 0); + regmap_field_write(ale->fields[POL_DST_MEN], 0); + regmap_field_write(ale->fields[POL_SRC_MEN], 0); + regmap_field_write(ale->fields[POL_OVLAN_MEN], 0); + regmap_field_write(ale->fields[POL_IVLAN_MEN], 0); + regmap_field_write(ale->fields[POL_ETHERTYPE_MEN], 0); + regmap_field_write(ale->fields[POL_IPSRC_MEN], 0); + regmap_field_write(ale->fields[POL_IPDST_MEN], 0); + regmap_field_write(ale->fields[POL_EN], 0); + regmap_field_write(ale->fields[POL_RED_DROP_EN], 0); + regmap_field_write(ale->fields[POL_YELLOW_DROP_EN], 0); + regmap_field_write(ale->fields[POL_PRIORITY_THREAD_EN], 0); + + cpsw_ale_policer_thread_idx_enable(ale, i, 0, 0); + } +} + +/* Default classifier is to map 8 user priorities to N receive channels */ +void cpsw_ale_classifier_setup_default(struct cpsw_ale *ale, int num_rx_ch) +{ + int pri, idx; + /* IEEE802.1D-2004, Standard for Local and metropolitan area networks + * Table G-2 - Traffic type acronyms + * Table G-3 - Defining traffic types + * User priority values 1 and 2 effectively communicate a lower + * priority than 0. In the below table 0 is assigned to higher priority + * thread than 1 and 2 wherever possible. + * The below table maps which thread the user priority needs to be + * sent to for a given number of threads (RX channels). Upper threads + * have higher priority. + * e.g. if number of threads is 8 then user priority 0 will map to + * pri_thread_map[8-1][0] i.e. thread 2 + */ + int pri_thread_map[8][8] = { { 0, 0, 0, 0, 0, 0, 0, 0, }, + { 0, 0, 0, 0, 1, 1, 1, 1, }, + { 0, 0, 0, 0, 1, 1, 2, 2, }, + { 1, 0, 0, 1, 2, 2, 3, 3, }, + { 1, 0, 0, 1, 2, 3, 4, 4, }, + { 1, 0, 0, 2, 3, 4, 5, 5, }, + { 1, 0, 0, 2, 3, 4, 5, 6, }, + { 2, 0, 1, 3, 4, 5, 6, 7, } }; + + cpsw_ale_policer_reset(ale); + + /* use first 8 classifiers to map 8 (DSCP/PCP) priorities to channels */ + for (pri = 0; pri < 8; pri++) { + idx = pri; + + /* Classifier 'idx' match on priority 'pri' */ + cpsw_ale_policer_read_idx(ale, idx); + regmap_field_write(ale->fields[POL_PRI_VAL], pri); + regmap_field_write(ale->fields[POL_PRI_MEN], 1); + cpsw_ale_policer_write_idx(ale, idx); + + /* Map Classifier 'idx' to thread provided by the map */ + cpsw_ale_policer_thread_idx_enable(ale, idx, + pri_thread_map[num_rx_ch - 1][pri], + 1); + } +} diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index 2cb76acc6d16..1e4e9a3dd234 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -193,5 +193,6 @@ int cpsw_ale_vlan_add_modify(struct cpsw_ale *ale, u16 vid, int port_mask, int cpsw_ale_vlan_del_modify(struct cpsw_ale *ale, u16 vid, int port_mask); void cpsw_ale_set_unreg_mcast(struct cpsw_ale *ale, int unreg_mcast_mask, bool add); +void cpsw_ale_classifier_setup_default(struct cpsw_ale *ale, int num_rx_ch); #endif From patchwork Wed Jul 3 13:51:37 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Roger Quadros X-Patchwork-Id: 809621 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7559117B432; Wed, 3 Jul 2024 13:52:02 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014722; cv=none; b=r2z9BYjQM2ys0eoSmw/2sIhoBmhEysgsWySs5AkXw50+FVyiq74Z3cwrcqj8dIG0Y7AZYRSnI92oPLlTg7oOZ4RIx09phty8RrBHh+mv7CheBjqEDVrWj6SV4g0X9W58yjaKwIYigyVRrefgmpTfUuB5d03YmBxniqQdjQQ6goo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1720014722; c=relaxed/simple; bh=8OS3EQTA1nsQoBdqA3J8k7Fv9Yb5rZa160z2POCWKtk=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=D9LkQ5PTGQSeeqHARcotRpLMzxYdwMCPLa8L82lBgexmCQdi88kRKaAKhzx6i77DSpgO+xQEIYpGbhQ7PyCjrLFkC3mixdJG6PY0wjijY5Rspum63fmaiTgfnpF1nCye3/XqBPyVBSSoF2cGxgu3qiA0LC00oPNuLdOCBkVGpKI= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=aZmEWP9j; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="aZmEWP9j" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 91CEDC3277B; Wed, 3 Jul 2024 13:51:58 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1720014722; bh=8OS3EQTA1nsQoBdqA3J8k7Fv9Yb5rZa160z2POCWKtk=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=aZmEWP9jyupuCkzheaevVIXB2nm4yGoDXZwPomP9xyF9P8+YIfWrq+ozDe4NRjBL2 Pib2KO37Qv1EcHUS9dOkLI6+mgkUx69AHIluXHeFXQxs825uzsGr/YrWFDiV4J8bmg qekDTuhS8VPDFOexJ+cZF8xQ6dXeLLh6AjEBhKpUwWieHDQmFRtbJ7jJJN5ddZLO5y wtfM/GrYA3jpIrdf5zGsKZvgY4ZXpQRQlfGU1xnbaqruIQO9Hxe2VqnJn/9ZOTa+Ja yh6g20GRkBEGHODExhGFz+dUF7/LpMeFpO+4wmDVlVXpcAX7kX1WKRoI3QlZro78OC mDKHteIm2Bj6w== From: Roger Quadros Date: Wed, 03 Jul 2024 16:51:37 +0300 Subject: [PATCH net-next v3 6/6] net: ethernet: ti: am65-cpsw: setup priority to flow mapping Precedence: bulk X-Mailing-List: linux-omap@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-Id: <20240703-am65-cpsw-multi-rx-v3-6-f11cd860fd72@kernel.org> References: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> In-Reply-To: <20240703-am65-cpsw-multi-rx-v3-0-f11cd860fd72@kernel.org> To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Siddharth Vadapalli , Julien Panis Cc: Simon Horman , Andrew Lunn , srk@ti.com, vigneshr@ti.com, danishanwar@ti.com, pekka Varis , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Roger Quadros X-Mailer: b4 0.12.4 X-Developer-Signature: v=1; a=openpgp-sha256; l=1075; i=rogerq@kernel.org; h=from:subject:message-id; bh=8OS3EQTA1nsQoBdqA3J8k7Fv9Yb5rZa160z2POCWKtk=; b=owEBbQKS/ZANAwAIAdJaa9O+djCTAcsmYgBmhVdm+OKRJdx9lr33jldo0NJ4kivOse5Ni9dtr cCQFZvK+tqJAjMEAAEIAB0WIQRBIWXUTJ9SeA+rEFjSWmvTvnYwkwUCZoVXZgAKCRDSWmvTvnYw k7tsEAC9Haw1DbVbwvt7Bfx397ddS5mrCZVd/3k0XTwOxBmM4cpkitCv1f/iuqL4rDDHlI+wPaT F8Do25w8YK7xBZDErXxmWTwDVM3CEv1Xv5sPlW4El+UPrWKma32XJEMAmVJ9cgwAgy0iB6+mBLG qfqMc0sRwwV5vhR++Vw8n1fP9H4QL4uZkjvVUMAgRg2wFIYjh7FGn0dK8ENwfSl+MFrexALhX3N lvatdFMb3kUfJOZ+S0g6JXHfPGHDFI3o4AwCc0dIFHaSVDC/+7ryaCV9n69OuJYkcdOxcXhf9+E x9V3IDX8m54rk4uo5kkV4wUFISD8lkLdgsMAwnTaFbDpTtZggcsH4/RSC1ElQw/5ToyGFGaVrfx 9+Gnuy6XDpoeaTqe1FlmfNpDIXIsMUtOznW9hWnUTr+cVHT04vsD2d6Qw1vd8qAL0cCES+dlK76 h8/aD60HO7gxI/m/4moZpZACzn0uTRerW1V7Mi8Tugesh1IxiHLNFs2ZXJPTtkfKVH95M2WaSXP O1GCr2Man+LQ+a3rVg682qoX4sYvAUpPah0liiNksmDPZtUAV3evrHkmHRthMqS08s7civtVsYH J3hKTEGrCPKU9S/QqgqAR9LA+93eg8vrOhM4OPoaoAv944KNXVb7n+SBw0DAn/0mExtaRNAkos9 qTBmB3J2I2IklAA== X-Developer-Key: i=rogerq@kernel.org; a=openpgp; fpr=412165D44C9F52780FAB1058D25A6BD3BE763093 Now that we support multiple RX queues, enable default priority to flow mapping so that higher priority packets come on higher channels (flows). The Classifier checks for PCP/DSCP priority in the packet and routes them to the appropriate flow. Signed-off-by: Roger Quadros Reviewed-by: Simon Horman --- Changelog: v3: - added Reviewed-by Simon Horman --- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index deb8e39e4648..4db03010d234 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -2497,6 +2497,9 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common) } } + /* setup classifier to route priorities to flows */ + cpsw_ale_classifier_setup_default(common->ale, common->rx_ch_num_flows); + err: i = devm_add_action(dev, am65_cpsw_nuss_free_rx_chns, common); if (i) {