From patchwork Tue Feb 27 18:54:53 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Justin Chen X-Patchwork-Id: 13574319 X-Patchwork-Delegate: kuba@kernel.org Received: from mail-pf1-f170.google.com (mail-pf1-f170.google.com [209.85.210.170]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 87C2814AD04 for ; Tue, 27 Feb 2024 18:55:09 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.170 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1709060111; cv=none; b=GtuP84+lRx/yAXHjaWx4iOXFCjLkeIVDYvQe3NpUIZ9tSQF6/fcHI5HT6cYd/py9T/PDNWS/g8vv/xDzV0PIsferlLWyLbGkgvwx3m8lvGq2bC1FS2v8VXIHBzTeI5m6Rwd7+XyQl7JCmXMXWQ4AvrWl08+7tlGqH2s0dFlf9Ts= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1709060111; c=relaxed/simple; bh=ToD0SmBWUrk7DVTpDD1+wlNk8RiKK71OqmhkfqYpRKM=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version:Content-Type; b=p5+ZEMzSo0mOGaE0kagNP69wwAneKB2H8eJGJVObWvGr5S1X9ga/l5Jb+tYS+Z+UBTgy6YM1OFyRFyw5CyJTgpD3COTNpte6E3c4xGSwOV4J14OF/c340X4Y8wLEokM+uDWll9lj+PoTVrawItFJPHKmjywPmkzo4uX0m8nBzDo= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=broadcom.com; spf=fail smtp.mailfrom=broadcom.com; dkim=pass (1024-bit key) header.d=broadcom.com header.i=@broadcom.com header.b=diJ54Yrw; arc=none smtp.client-ip=209.85.210.170 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=broadcom.com Authentication-Results: smtp.subspace.kernel.org; spf=fail smtp.mailfrom=broadcom.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=broadcom.com header.i=@broadcom.com header.b="diJ54Yrw" Received: by mail-pf1-f170.google.com with SMTP id d2e1a72fcca58-6e544f927d4so75082b3a.0 for ; Tue, 27 Feb 2024 10:55:09 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=broadcom.com; s=google; t=1709060109; x=1709664909; darn=vger.kernel.org; h=mime-version:references:in-reply-to:message-id:date:subject:cc:to :from:from:to:cc:subject:date:message-id:reply-to; bh=6/81acXTLD3okeoxdhRaZaWmVZzxskxx/NHE1ckLvTs=; b=diJ54YrwgiMK07fS3GTvcmUAZBm+RRgtoZ4DGSHUpQS+UEheDJVaUGF4rcxZkPCb2d scKZsbQsKcDR+hfMJoSUtAo5v8HqplhoJ127aVrvyVznWmihFL6FMlCSyAMgXYcSJNUY s+ZH10fPoytaRSqueKBuSznhTBxNMm8wSy6K0= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1709060109; x=1709664909; h=mime-version:references:in-reply-to:message-id:date:subject:cc:to :from:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=6/81acXTLD3okeoxdhRaZaWmVZzxskxx/NHE1ckLvTs=; b=MtJ8RPU9zu+1u2oR9iPu4KYI/+utvqa+nELVUUPLxnjNs7h4UnEm4rJTRlhUIrRGa1 VnYLcnTO36k3BaeMklej1iChFd7PSAA8pQ/EMzsWP9Zb1tUqg40eur/DBThAsNH5RYhT +WxNY6WF2+DKyEg8M+exR2EINh6vqncq0izKqb2FQwGzcvcIYLn1AdmUEw6jE1tRKNlE JgAZwUU/OF6UFQVU++MYuW5E4l5y39nADthpJWu+SGxLESlGRKPWLjXPTLmDaLeSe0lA l5GaDSEixxv6LbJU8hh/QV09wQak342u/aZTR5YWvRpjUd3F9MB1/5HVLBzdfsjX1Vz+ X6fQ== X-Gm-Message-State: AOJu0YyvmG2EcZJdVv63wULEIV+j/lvHli668GEt3fKSiht9xFvLmto2 swxcIg2n6n1eb3CNoeD+Wq/nZI95CLHgRmWFNodbbXbJykca2rmxZ8e5sNdVUGiavS6rsDOSQ6a Ul97B/ZAy0MDaleTh+2i2Pbx1Akzbic9/NHL0ydXlB/gTWJ0Y+oSla8POk1me20iNMB4/vYEL4i o8Vq7kYnMCluztcsLCNtKGk11duFl31VV36Ppqx0mVw4DM X-Google-Smtp-Source: AGHT+IHFml8YC5w8sVYrpw7xt2tiOtfMLaUqB/iEG7thNNsuKn7QKwHLK2kCXWJEif9XHpxMW6qMPg== X-Received: by 2002:a05:6a00:4f0a:b0:6e5:bb8:dc19 with SMTP id lb10-20020a056a004f0a00b006e50bb8dc19mr196805pfb.15.1709060108348; Tue, 27 Feb 2024 10:55:08 -0800 (PST) Received: from stbirv-lnx-1.igp.broadcom.net ([192.19.223.252]) by smtp.gmail.com with ESMTPSA id r6-20020aa79886000000b006e466369645sm6236898pfl.132.2024.02.27.10.55.07 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 27 Feb 2024 10:55:07 -0800 (PST) From: Justin Chen To: netdev@vger.kernel.org Cc: florian.fainelli@broadcom.com, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, robh+dt@kernel.org, krzysztof.kozlowski+dt@linaro.org, conor+dt@kernel.org, opendmb@gmail.com, bcm-kernel-feedback-list@broadcom.com, andrew@lunn.ch, hkallweit1@gmail.com, linux@armlinux.org.uk, rafal@milecki.pl, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, Justin Chen Subject: [PATCH net-next v2 5/6] net: bcmasp: Keep buffers through power management Date: Tue, 27 Feb 2024 10:54:53 -0800 Message-Id: <20240227185454.2767610-6-justin.chen@broadcom.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20240227185454.2767610-1-justin.chen@broadcom.com> References: <20240227185454.2767610-1-justin.chen@broadcom.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Patchwork-Delegate: kuba@kernel.org There is no advantage of freeing and re-allocating buffers through suspend and resume. This waste cycles and makes suspend/resume time longer. We also open ourselves to failed allocations in systems with heavy memory fragmentation. Signed-off-by: Justin Chen Acked-by: Florian Fainelli --- drivers/net/ethernet/broadcom/asp2/bcmasp.h | 1 + .../net/ethernet/broadcom/asp2/bcmasp_intf.c | 181 ++++++++---------- 2 files changed, 76 insertions(+), 106 deletions(-) diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp.h b/drivers/net/ethernet/broadcom/asp2/bcmasp.h index 61598dc070b1..127a5340625e 100644 --- a/drivers/net/ethernet/broadcom/asp2/bcmasp.h +++ b/drivers/net/ethernet/broadcom/asp2/bcmasp.h @@ -315,6 +315,7 @@ struct bcmasp_intf { struct bcmasp_desc *rx_edpkt_cpu; dma_addr_t rx_edpkt_dma_addr; dma_addr_t rx_edpkt_dma_read; + dma_addr_t rx_edpkt_dma_valid; /* RX buffer prefetcher ring*/ void *rx_ring_cpu; diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c index 36e6fae937ea..0b378a6d43e7 100644 --- a/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c +++ b/drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c @@ -674,40 +674,78 @@ static void bcmasp_adj_link(struct net_device *dev) phy_print_status(phydev); } -static int bcmasp_init_rx(struct bcmasp_intf *intf) +static int bcmasp_alloc_buffers(struct bcmasp_intf *intf) { struct device *kdev = &intf->parent->pdev->dev; struct page *buffer_pg; - dma_addr_t dma; - void *p; - u32 reg; - int ret; + /* Alloc RX */ intf->rx_buf_order = get_order(RING_BUFFER_SIZE); buffer_pg = alloc_pages(GFP_KERNEL, intf->rx_buf_order); if (!buffer_pg) return -ENOMEM; - dma = dma_map_page(kdev, buffer_pg, 0, RING_BUFFER_SIZE, - DMA_FROM_DEVICE); - if (dma_mapping_error(kdev, dma)) { - __free_pages(buffer_pg, intf->rx_buf_order); - return -ENOMEM; - } intf->rx_ring_cpu = page_to_virt(buffer_pg); - intf->rx_ring_dma = dma; - intf->rx_ring_dma_valid = intf->rx_ring_dma + RING_BUFFER_SIZE - 1; + intf->rx_ring_dma = dma_map_page(kdev, buffer_pg, 0, RING_BUFFER_SIZE, + DMA_FROM_DEVICE); + if (dma_mapping_error(kdev, intf->rx_ring_dma)) + goto free_rx_buffer; + + intf->rx_edpkt_cpu = dma_alloc_coherent(kdev, DESC_RING_SIZE, + &intf->rx_edpkt_dma_addr, GFP_KERNEL); + if (!intf->rx_edpkt_cpu) + goto free_rx_buffer_dma; + + /* Alloc TX */ + intf->tx_spb_cpu = dma_alloc_coherent(kdev, DESC_RING_SIZE, + &intf->tx_spb_dma_addr, GFP_KERNEL); + if (!intf->tx_spb_cpu) + goto free_rx_edpkt_dma; - p = dma_alloc_coherent(kdev, DESC_RING_SIZE, &intf->rx_edpkt_dma_addr, + intf->tx_cbs = kcalloc(DESC_RING_COUNT, sizeof(struct bcmasp_tx_cb), GFP_KERNEL); - if (!p) { - ret = -ENOMEM; - goto free_rx_ring; - } - intf->rx_edpkt_cpu = p; + if (!intf->tx_cbs) + goto free_tx_spb_dma; - netif_napi_add(intf->ndev, &intf->rx_napi, bcmasp_rx_poll); + return 0; + +free_tx_spb_dma: + dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu, + intf->tx_spb_dma_addr); +free_rx_edpkt_dma: + dma_free_coherent(kdev, DESC_RING_SIZE, intf->rx_edpkt_cpu, + intf->rx_edpkt_dma_addr); +free_rx_buffer_dma: + dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE, + DMA_FROM_DEVICE); +free_rx_buffer: + __free_pages(buffer_pg, intf->rx_buf_order); + + return -ENOMEM; +} + +static void bcmasp_reclaim_free_buffers(struct bcmasp_intf *intf) +{ + struct device *kdev = &intf->parent->pdev->dev; + + /* RX buffers */ + dma_free_coherent(kdev, DESC_RING_SIZE, intf->rx_edpkt_cpu, + intf->rx_edpkt_dma_addr); + dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE, + DMA_FROM_DEVICE); + __free_pages(virt_to_page(intf->rx_ring_cpu), intf->rx_buf_order); + + /* TX buffers */ + dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu, + intf->tx_spb_dma_addr); + kfree(intf->tx_cbs); +} +static void bcmasp_init_rx(struct bcmasp_intf *intf) +{ + /* Restart from index 0 */ + intf->rx_ring_dma_valid = intf->rx_ring_dma + RING_BUFFER_SIZE - 1; + intf->rx_edpkt_dma_valid = intf->rx_edpkt_dma_addr + (DESC_RING_SIZE - 1); intf->rx_edpkt_dma_read = intf->rx_edpkt_dma_addr; intf->rx_edpkt_index = 0; @@ -733,64 +771,23 @@ static int bcmasp_init_rx(struct bcmasp_intf *intf) rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_WRITE); rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_READ); rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr, RX_EDPKT_DMA_BASE); - rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr + (DESC_RING_SIZE - 1), - RX_EDPKT_DMA_END); - rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_addr + (DESC_RING_SIZE - 1), - RX_EDPKT_DMA_VALID); - - reg = UMAC2FB_CFG_DEFAULT_EN | - ((intf->channel + 11) << UMAC2FB_CFG_CHID_SHIFT); - reg |= (0xd << UMAC2FB_CFG_OK_SEND_SHIFT); - umac2fb_wl(intf, reg, UMAC2FB_CFG); + rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_valid, RX_EDPKT_DMA_END); + rx_edpkt_dma_wq(intf, intf->rx_edpkt_dma_valid, RX_EDPKT_DMA_VALID); - return 0; - -free_rx_ring: - dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE, - DMA_FROM_DEVICE); - __free_pages(virt_to_page(intf->rx_ring_cpu), intf->rx_buf_order); - - return ret; + umac2fb_wl(intf, UMAC2FB_CFG_DEFAULT_EN | ((intf->channel + 11) << + UMAC2FB_CFG_CHID_SHIFT) | (0xd << UMAC2FB_CFG_OK_SEND_SHIFT), + UMAC2FB_CFG); } -static void bcmasp_reclaim_free_all_rx(struct bcmasp_intf *intf) -{ - struct device *kdev = &intf->parent->pdev->dev; - - dma_free_coherent(kdev, DESC_RING_SIZE, intf->rx_edpkt_cpu, - intf->rx_edpkt_dma_addr); - dma_unmap_page(kdev, intf->rx_ring_dma, RING_BUFFER_SIZE, - DMA_FROM_DEVICE); - __free_pages(virt_to_page(intf->rx_ring_cpu), intf->rx_buf_order); -} -static int bcmasp_init_tx(struct bcmasp_intf *intf) +static void bcmasp_init_tx(struct bcmasp_intf *intf) { - struct device *kdev = &intf->parent->pdev->dev; - void *p; - int ret; - - p = dma_alloc_coherent(kdev, DESC_RING_SIZE, &intf->tx_spb_dma_addr, - GFP_KERNEL); - if (!p) - return -ENOMEM; - - intf->tx_spb_cpu = p; + /* Restart from index 0 */ intf->tx_spb_dma_valid = intf->tx_spb_dma_addr + DESC_RING_SIZE - 1; intf->tx_spb_dma_read = intf->tx_spb_dma_addr; - - intf->tx_cbs = kcalloc(DESC_RING_COUNT, sizeof(struct bcmasp_tx_cb), - GFP_KERNEL); - if (!intf->tx_cbs) { - ret = -ENOMEM; - goto free_tx_spb; - } - intf->tx_spb_index = 0; intf->tx_spb_clean_index = 0; - netif_napi_add_tx(intf->ndev, &intf->tx_napi, bcmasp_tx_poll); - /* Make sure channels are disabled */ tx_spb_ctrl_wl(intf, 0x0, TX_SPB_CTRL_ENABLE); tx_epkt_core_wl(intf, 0x0, TX_EPKT_C_CFG_MISC); @@ -806,26 +803,6 @@ static int bcmasp_init_tx(struct bcmasp_intf *intf) tx_spb_dma_wq(intf, intf->tx_spb_dma_addr, TX_SPB_DMA_BASE); tx_spb_dma_wq(intf, intf->tx_spb_dma_valid, TX_SPB_DMA_END); tx_spb_dma_wq(intf, intf->tx_spb_dma_valid, TX_SPB_DMA_VALID); - - return 0; - -free_tx_spb: - dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu, - intf->tx_spb_dma_addr); - - return ret; -} - -static void bcmasp_reclaim_free_all_tx(struct bcmasp_intf *intf) -{ - struct device *kdev = &intf->parent->pdev->dev; - - /* Free descriptors */ - dma_free_coherent(kdev, DESC_RING_SIZE, intf->tx_spb_cpu, - intf->tx_spb_dma_addr); - - /* Free cbs */ - kfree(intf->tx_cbs); } static void bcmasp_ephy_enable_set(struct bcmasp_intf *intf, bool enable) @@ -915,10 +892,7 @@ static void bcmasp_netif_deinit(struct net_device *dev) bcmasp_enable_rx_irq(intf, 0); netif_napi_del(&intf->tx_napi); - bcmasp_reclaim_free_all_tx(intf); - netif_napi_del(&intf->rx_napi); - bcmasp_reclaim_free_all_rx(intf); } static int bcmasp_stop(struct net_device *dev) @@ -932,6 +906,8 @@ static int bcmasp_stop(struct net_device *dev) bcmasp_netif_deinit(dev); + bcmasp_reclaim_free_buffers(intf); + phy_disconnect(dev->phydev); /* Disable internal EPHY or external PHY */ @@ -1073,17 +1049,12 @@ static int bcmasp_netif_init(struct net_device *dev, bool phy_connect) intf->old_link = -1; intf->old_pause = -1; - ret = bcmasp_init_tx(intf); - if (ret) - goto err_phy_disconnect; - - /* Turn on asp */ + bcmasp_init_tx(intf); + netif_napi_add_tx(intf->ndev, &intf->tx_napi, bcmasp_tx_poll); bcmasp_enable_tx(intf, 1); - ret = bcmasp_init_rx(intf); - if (ret) - goto err_reclaim_tx; - + bcmasp_init_rx(intf); + netif_napi_add(intf->ndev, &intf->rx_napi, bcmasp_rx_poll); bcmasp_enable_rx(intf, 1); /* Turn on UniMAC TX/RX */ @@ -1097,12 +1068,6 @@ static int bcmasp_netif_init(struct net_device *dev, bool phy_connect) return 0; -err_reclaim_tx: - netif_napi_del(&intf->tx_napi); - bcmasp_reclaim_free_all_tx(intf); -err_phy_disconnect: - if (phydev) - phy_disconnect(phydev); err_phy_disable: if (intf->internal_phy) bcmasp_ephy_enable_set(intf, false); @@ -1118,6 +1083,10 @@ static int bcmasp_open(struct net_device *dev) netif_dbg(intf, ifup, dev, "bcmasp open\n"); + ret = bcmasp_alloc_buffers(intf); + if (ret) + return ret; + ret = clk_prepare_enable(intf->parent->clk); if (ret) return ret;