Message ID | 20200423070026.26200-1-vee.khee.wong@intel.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | [net-next,1/1] net: stmmac: Add option for VLAN filter fail queue enable | expand |
From: Wong Vee Khee <vee.khee.wong@intel.com> Date: Thu, 23 Apr 2020 15:00:26 +0800 > From: "Chuah, Kim Tatt" <kim.tatt.chuah@intel.com> > > Add option in plat_stmmacenet_data struct to enable VLAN Filter Fail > Queuing. This option allows packets that fail VLAN filter to be routed > to a specific Rx queue when Receive All is also set. > > When this option is enabled: > - Enable VFFQ only when entering promiscuous mode, because Receive All > will pass up all rx packets that failed address filtering (similar to > promiscuous mode). > - VLAN-promiscuous mode is never entered to allow rx packet to fail VLAN > filters and get routed to selected VFFQ Rx queue. > > Reviewed-by: Voon Weifeng <weifeng.voon@intel.com> > Reviewed-by: Ong Boon Leong <boon.leong.ong@intel.com> > Signed-off-by: Chuah, Kim Tatt <kim.tatt.chuah@intel.com> > Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com> Why would you be setting this with a platform attribute? Even if the capability exists, wouldn't you want the user to be able to choose to opt out?
> -----Original Message----- > From: netdev-owner@vger.kernel.org <netdev-owner@vger.kernel.org> On > Behalf Of David Miller > Sent: Friday, April 24, 2020 6:53 AM > To: Wong, Vee Khee <vee.khee.wong@intel.com> > Cc: peppe.cavallaro@st.com; alexandre.torgue@st.com; > joabreu@synopsys.com; mcoquelin.stm32@gmail.com; > netdev@vger.kernel.org; linux-stm32@st-md-mailman.stormreply.com; > linux-arm-kernel@lists.infradead.org; linux-kernel@vger.kernel.org; Ong, > Boon Leong <boon.leong.ong@intel.com>; Voon, Weifeng > <weifeng.voon@intel.com> > Subject: Re: [PATCH net-next 1/1] net: stmmac: Add option for VLAN filter fail > queue enable > > From: Wong Vee Khee <vee.khee.wong@intel.com> > Date: Thu, 23 Apr 2020 15:00:26 +0800 > > > From: "Chuah, Kim Tatt" <kim.tatt.chuah@intel.com> > > > > Add option in plat_stmmacenet_data struct to enable VLAN Filter Fail > > Queuing. This option allows packets that fail VLAN filter to be routed > > to a specific Rx queue when Receive All is also set. > > > > When this option is enabled: > > - Enable VFFQ only when entering promiscuous mode, because Receive All > > will pass up all rx packets that failed address filtering (similar to > > promiscuous mode). > > - VLAN-promiscuous mode is never entered to allow rx packet to fail VLAN > > filters and get routed to selected VFFQ Rx queue. > > > > Reviewed-by: Voon Weifeng <weifeng.voon@intel.com> > > Reviewed-by: Ong Boon Leong <boon.leong.ong@intel.com> > > Signed-off-by: Chuah, Kim Tatt <kim.tatt.chuah@intel.com> > > Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com> > > Why would you be setting this with a platform attribute? Even if the > capability exists, wouldn't you want the user to be able to choose to opt out? Hi Jose/David Miller, 1/ In current implementation, TSN uses VLAN filter that can either accept/reject VLAN-tagged network packets. In some situation, we do not want to drop failed packets, but instead steer the packet to a VLAN Failed Queue Channel. 2/ VLAN Fail Queue Channel will be set to use the RxQ with higheset index as per HW IP configuration because that is the least priority channel. 3/ The way user will enable this feature is through promiscuous mode settings using ifconfig. (e.g. ifconfig enp0s30f4 promisc) 4/ VLAN Filter Fail Packets Queue feature is IP version specific (only applicable to DWMAC5). I would propose we add this under platform data (e.g. dwmac-intel), so that it can be built in according to hardware on a separate patch. Any thoughts?
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 127f75862962..52e6d1ef7176 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -474,6 +474,8 @@ struct mac_device_info { unsigned int num_vlan; u32 vlan_filter[32]; unsigned int promisc; + bool vlan_fail_q_en; + u8 vlan_fail_q; }; struct stmmac_rx_routing { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 61f3249bd724..592b043f9676 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -76,6 +76,7 @@ #define GMAC_PACKET_FILTER_HPF BIT(10) #define GMAC_PACKET_FILTER_VTFE BIT(16) #define GMAC_PACKET_FILTER_IPFE BIT(20) +#define GMAC_PACKET_FILTER_RA BIT(31) #define GMAC_MAX_PERFECT_ADDRESSES 128 diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index ecd834e0e121..002791b77356 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -618,7 +618,18 @@ static void dwmac4_set_filter(struct mac_device_info *hw, value &= ~GMAC_PACKET_FILTER_PM; value &= ~GMAC_PACKET_FILTER_PR; if (dev->flags & IFF_PROMISC) { - value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_PCF; + /* VLAN Tag Filter Fail Packets Queuing */ + if (hw->vlan_fail_q_en) { + value = readl(ioaddr + GMAC_RXQ_CTRL4); + value &= ~GMAC_RXQCTRL_VFFQ_MASK; + value |= GMAC_RXQCTRL_VFFQE | + (hw->vlan_fail_q << GMAC_RXQCTRL_VFFQ_SHIFT); + writel(value, ioaddr + GMAC_RXQ_CTRL4); + value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_RA; + } else { + value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_PCF; + } + } else if ((dev->flags & IFF_ALLMULTI) || (netdev_mc_count(dev) > hw->multicast_filter_bins)) { /* Pass all multi */ @@ -680,7 +691,7 @@ static void dwmac4_set_filter(struct mac_device_info *hw, writel(value, ioaddr + GMAC_PACKET_FILTER); - if (dev->flags & IFF_PROMISC) { + if (dev->flags & IFF_PROMISC && !hw->vlan_fail_q_en) { if (!hw->promisc) { hw->promisc = 1; dwmac4_vlan_promisc_enable(dev, hw); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h index 3e8faa96b4d4..64530deb55ba 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h @@ -92,6 +92,12 @@ #define TCEIE BIT(0) #define DMA_ECC_INT_STATUS 0x00001088 +/* EQoS version 5.xx VLAN Tag Filter Fail Packets Queuing */ +#define GMAC_RXQ_CTRL4 0x00000094 +#define GMAC_RXQCTRL_VFFQ_MASK GENMASK(19, 17) +#define GMAC_RXQCTRL_VFFQ_SHIFT 17 +#define GMAC_RXQCTRL_VFFQE BIT(16) + int dwmac5_safety_feat_config(void __iomem *ioaddr, unsigned int asp); int dwmac5_safety_feat_irq_status(struct net_device *ndev, void __iomem *ioaddr, unsigned int asp, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index e6898fd5223f..27d9bf98c908 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4723,6 +4723,9 @@ static int stmmac_hw_init(struct stmmac_priv *priv) if (priv->dma_cap.tsoen) dev_info(priv->device, "TSO supported\n"); + priv->hw->vlan_fail_q_en = priv->plat->vlan_fail_q_en; + priv->hw->vlan_fail_q = priv->plat->vlan_fail_q; + /* Run HW quirks, if any */ if (priv->hwif_quirks) { ret = priv->hwif_quirks(priv); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index fbafb353e9be..f4287f363031 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -196,5 +196,7 @@ struct plat_stmmacenet_data { int mac_port_sel_speed; bool en_tx_lpi_clockgating; int has_xgmac; + bool vlan_fail_q_en; + u8 vlan_fail_q; }; #endif