From: Felix Fietkau Date: Wed, 26 Aug 2020 17:02:30 +0200 Subject: [PATCH] net: ethernet: mtk_eth_soc: implement dynamic interrupt moderation Reduces the number of interrupts under load Signed-off-by: Felix Fietkau --- --- a/drivers/net/ethernet/mediatek/Kconfig +++ b/drivers/net/ethernet/mediatek/Kconfig @@ -10,6 +10,7 @@ if NET_VENDOR_MEDIATEK config NET_MEDIATEK_SOC tristate "MediaTek SoC Gigabit Ethernet support" select PHYLINK + select DIMLIB ---help--- This driver supports the gigabit ethernet MACs in the MediaTek SoC family. --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1249,12 +1249,13 @@ static void mtk_update_rx_cpu_idx(struct static int mtk_poll_rx(struct napi_struct *napi, int budget, struct mtk_eth *eth) { + struct dim_sample dim_sample = {}; struct mtk_rx_ring *ring; int idx; struct sk_buff *skb; u8 *data, *new_data; struct mtk_rx_dma *rxd, trxd; - int done = 0; + int done = 0, bytes = 0; while (done < budget) { struct net_device *netdev; @@ -1328,6 +1329,7 @@ static int mtk_poll_rx(struct napi_struc else skb_checksum_none_assert(skb); skb->protocol = eth_type_trans(skb, netdev); + bytes += pktlen; if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX && (trxd.rxd2 & RX_DMA_VTAG)) @@ -1359,6 +1361,12 @@ rx_done: mtk_update_rx_cpu_idx(eth); } + eth->rx_packets += done; + eth->rx_bytes += bytes; + dim_update_sample(eth->rx_events, eth->rx_packets, eth->rx_bytes, + &dim_sample); + net_dim(ð->rx_dim, dim_sample); + return done; } @@ -1451,6 +1459,7 @@ static int mtk_poll_tx_pdma(struct mtk_e static int mtk_poll_tx(struct mtk_eth *eth, int budget) { struct mtk_tx_ring *ring = ð->tx_ring; + struct dim_sample dim_sample = {}; unsigned int done[MTK_MAX_DEVS]; unsigned int bytes[MTK_MAX_DEVS]; int total = 0, i; @@ -1468,8 +1477,14 @@ static int mtk_poll_tx(struct mtk_eth *e continue; netdev_completed_queue(eth->netdev[i], done[i], bytes[i]); total += done[i]; + eth->tx_packets += done[i]; + eth->tx_bytes += bytes[i]; } + dim_update_sample(eth->tx_events, eth->tx_packets, eth->tx_bytes, + &dim_sample); + net_dim(ð->tx_dim, dim_sample); + if (mtk_queue_stopped(eth) && (atomic_read(&ring->free_count) > ring->thresh)) mtk_wake_queue(eth); @@ -2144,6 +2159,7 @@ static irqreturn_t mtk_handle_irq_rx(int { struct mtk_eth *eth = _eth; + eth->rx_events++; if (likely(napi_schedule_prep(ð->rx_napi))) { __napi_schedule(ð->rx_napi); mtk_rx_irq_disable(eth, MTK_RX_DONE_INT); @@ -2156,6 +2172,7 @@ static irqreturn_t mtk_handle_irq_tx(int { struct mtk_eth *eth = _eth; + eth->tx_events++; if (likely(napi_schedule_prep(ð->tx_napi))) { __napi_schedule(ð->tx_napi); mtk_tx_irq_disable(eth, MTK_TX_DONE_INT); @@ -2332,6 +2349,9 @@ static int mtk_stop(struct net_device *d napi_disable(ð->tx_napi); napi_disable(ð->rx_napi); + cancel_work_sync(ð->rx_dim.work); + cancel_work_sync(ð->tx_dim.work); + if (MTK_HAS_CAPS(eth->soc->caps, MTK_QDMA)) mtk_stop_dma(eth, MTK_QDMA_GLO_CFG); mtk_stop_dma(eth, MTK_PDMA_GLO_CFG); @@ -2381,6 +2401,64 @@ err_disable_clks: return ret; } +static void mtk_dim_rx(struct work_struct *work) +{ + struct dim *dim = container_of(work, struct dim, work); + struct mtk_eth *eth = container_of(dim, struct mtk_eth, rx_dim); + struct dim_cq_moder cur_profile; + u32 val, cur; + + cur_profile = net_dim_get_rx_moderation(eth->rx_dim.mode, + dim->profile_ix); + spin_lock_bh(ð->dim_lock); + + val = mtk_r32(eth, MTK_PDMA_DELAY_INT); + val &= MTK_PDMA_DELAY_TX_MASK; + val |= MTK_PDMA_DELAY_RX_EN; + + cur = min_t(u32, DIV_ROUND_UP(cur_profile.usec, 20), MTK_PDMA_DELAY_PTIME_MASK); + val |= cur << MTK_PDMA_DELAY_RX_PTIME_SHIFT; + + cur = min_t(u32, cur_profile.pkts, MTK_PDMA_DELAY_PINT_MASK); + val |= cur << MTK_PDMA_DELAY_RX_PINT_SHIFT; + + mtk_w32(eth, val, MTK_PDMA_DELAY_INT); + mtk_w32(eth, val, MTK_QDMA_DELAY_INT); + + spin_unlock_bh(ð->dim_lock); + + dim->state = DIM_START_MEASURE; +} + +static void mtk_dim_tx(struct work_struct *work) +{ + struct dim *dim = container_of(work, struct dim, work); + struct mtk_eth *eth = container_of(dim, struct mtk_eth, tx_dim); + struct dim_cq_moder cur_profile; + u32 val, cur; + + cur_profile = net_dim_get_tx_moderation(eth->tx_dim.mode, + dim->profile_ix); + spin_lock_bh(ð->dim_lock); + + val = mtk_r32(eth, MTK_PDMA_DELAY_INT); + val &= MTK_PDMA_DELAY_RX_MASK; + val |= MTK_PDMA_DELAY_TX_EN; + + cur = min_t(u32, DIV_ROUND_UP(cur_profile.usec, 20), MTK_PDMA_DELAY_PTIME_MASK); + val |= cur << MTK_PDMA_DELAY_TX_PTIME_SHIFT; + + cur = min_t(u32, cur_profile.pkts, MTK_PDMA_DELAY_PINT_MASK); + val |= cur << MTK_PDMA_DELAY_TX_PINT_SHIFT; + + mtk_w32(eth, val, MTK_PDMA_DELAY_INT); + mtk_w32(eth, val, MTK_QDMA_DELAY_INT); + + spin_unlock_bh(ð->dim_lock); + + dim->state = DIM_START_MEASURE; +} + static int mtk_hw_init(struct mtk_eth *eth) { int i, val, ret; @@ -2402,9 +2480,6 @@ static int mtk_hw_init(struct mtk_eth *e goto err_disable_pm; } - /* enable interrupt delay for RX */ - mtk_w32(eth, MTK_PDMA_DELAY_RX_DELAY, MTK_PDMA_DELAY_INT); - /* disable delay and normal interrupt */ mtk_tx_irq_disable(eth, ~0); mtk_rx_irq_disable(eth, ~0); @@ -2443,11 +2518,10 @@ static int mtk_hw_init(struct mtk_eth *e /* Enable RX VLan Offloading */ mtk_w32(eth, 1, MTK_CDMP_EG_CTRL); - /* enable interrupt delay for RX */ - mtk_w32(eth, MTK_PDMA_DELAY_RX_DELAY, MTK_PDMA_DELAY_INT); + mtk_dim_rx(ð->rx_dim.work); + mtk_dim_tx(ð->tx_dim.work); /* disable delay and normal interrupt */ - mtk_w32(eth, 0, MTK_QDMA_DELAY_INT); mtk_tx_irq_disable(eth, ~0); mtk_rx_irq_disable(eth, ~0); @@ -2948,6 +3022,13 @@ static int mtk_probe(struct platform_dev spin_lock_init(ð->page_lock); spin_lock_init(ð->tx_irq_lock); spin_lock_init(ð->rx_irq_lock); + spin_lock_init(ð->dim_lock); + + eth->rx_dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + INIT_WORK(ð->rx_dim.work, mtk_dim_rx); + + eth->tx_dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + INIT_WORK(ð->tx_dim.work, mtk_dim_tx); if (!MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628)) { eth->ethsys = syscon_regmap_lookup_by_phandle(pdev->dev.of_node, --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -15,6 +15,7 @@ #include #include #include +#include #define MTK_QDMA_PAGE_SIZE 2048 #define MTK_MAX_RX_LENGTH 1536 @@ -131,13 +132,18 @@ /* PDMA Delay Interrupt Register */ #define MTK_PDMA_DELAY_INT 0xa0c +#define MTK_PDMA_DELAY_RX_MASK GENMASK(15, 0) #define MTK_PDMA_DELAY_RX_EN BIT(15) -#define MTK_PDMA_DELAY_RX_PINT 4 #define MTK_PDMA_DELAY_RX_PINT_SHIFT 8 -#define MTK_PDMA_DELAY_RX_PTIME 4 -#define MTK_PDMA_DELAY_RX_DELAY \ - (MTK_PDMA_DELAY_RX_EN | MTK_PDMA_DELAY_RX_PTIME | \ - (MTK_PDMA_DELAY_RX_PINT << MTK_PDMA_DELAY_RX_PINT_SHIFT)) +#define MTK_PDMA_DELAY_RX_PTIME_SHIFT 0 + +#define MTK_PDMA_DELAY_TX_MASK GENMASK(31, 16) +#define MTK_PDMA_DELAY_TX_EN BIT(31) +#define MTK_PDMA_DELAY_TX_PINT_SHIFT 24 +#define MTK_PDMA_DELAY_TX_PTIME_SHIFT 16 + +#define MTK_PDMA_DELAY_PINT_MASK 0x7f +#define MTK_PDMA_DELAY_PTIME_MASK 0xff /* PDMA Interrupt Status Register */ #define MTK_PDMA_INT_STATUS 0xa20 @@ -219,6 +225,7 @@ /* QDMA Interrupt Status Register */ #define MTK_QDMA_INT_STATUS 0x1A18 #define MTK_RX_DONE_DLY BIT(30) +#define MTK_TX_DONE_DLY BIT(28) #define MTK_RX_DONE_INT3 BIT(19) #define MTK_RX_DONE_INT2 BIT(18) #define MTK_RX_DONE_INT1 BIT(17) @@ -228,8 +235,7 @@ #define MTK_TX_DONE_INT1 BIT(1) #define MTK_TX_DONE_INT0 BIT(0) #define MTK_RX_DONE_INT MTK_RX_DONE_DLY -#define MTK_TX_DONE_INT (MTK_TX_DONE_INT0 | MTK_TX_DONE_INT1 | \ - MTK_TX_DONE_INT2 | MTK_TX_DONE_INT3) +#define MTK_TX_DONE_INT MTK_TX_DONE_DLY /* QDMA Interrupt grouping registers */ #define MTK_QDMA_INT_GRP1 0x1a20 @@ -912,6 +918,18 @@ struct mtk_eth { const struct mtk_soc_data *soc; + spinlock_t dim_lock; + + u32 rx_events; + u32 rx_packets; + u32 rx_bytes; + struct dim rx_dim; + + u32 tx_events; + u32 tx_packets; + u32 tx_bytes; + struct dim tx_dim; + u32 tx_int_mask_reg; u32 tx_int_status_reg; u32 rx_dma_l4_valid;