Commit c6d30e83 authored by Michael Chan's avatar Michael Chan Committed by David S. Miller

bnxt_en: Add basic XDP support.

Add basic ndo_xdp support to setup and query program, configure the NIC
to run in rx page mode, and support XDP_PASS, XDP_DROP, XDP_ABORTED
actions only.

v3: Pass modified offset and length to stack for XDP_PASS.
    Remove Kconfig option.

v2: Added trace_xdp_exception()
    Added dma_syncs.
    Added XDP headroom support.
Signed-off-by: default avatarMichael Chan <michael.chan@broadcom.com>
Tested-by: default avatarAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent fa3e93e8
obj-$(CONFIG_BNXT) += bnxt_en.o obj-$(CONFIG_BNXT) += bnxt_en.o
bnxt_en-y := bnxt.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o bnxt_en-y := bnxt.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o bnxt_xdp.o
...@@ -33,6 +33,7 @@ ...@@ -33,6 +33,7 @@
#include <linux/if.h> #include <linux/if.h>
#include <linux/if_vlan.h> #include <linux/if_vlan.h>
#include <linux/rtc.h> #include <linux/rtc.h>
#include <linux/bpf.h>
#include <net/ip.h> #include <net/ip.h>
#include <net/tcp.h> #include <net/tcp.h>
#include <net/udp.h> #include <net/udp.h>
...@@ -53,6 +54,7 @@ ...@@ -53,6 +54,7 @@
#include "bnxt_sriov.h" #include "bnxt_sriov.h"
#include "bnxt_ethtool.h" #include "bnxt_ethtool.h"
#include "bnxt_dcb.h" #include "bnxt_dcb.h"
#include "bnxt_xdp.h"
#define BNXT_TX_TIMEOUT (5 * HZ) #define BNXT_TX_TIMEOUT (5 * HZ)
...@@ -642,8 +644,7 @@ static inline int bnxt_alloc_rx_data(struct bnxt *bp, ...@@ -642,8 +644,7 @@ static inline int bnxt_alloc_rx_data(struct bnxt *bp,
return 0; return 0;
} }
static void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void *data)
void *data)
{ {
u16 prod = rxr->rx_prod; u16 prod = rxr->rx_prod;
struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf; struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf;
...@@ -1480,6 +1481,11 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons, ...@@ -1480,6 +1481,11 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
len = le32_to_cpu(rxcmp->rx_cmp_len_flags_type) >> RX_CMP_LEN_SHIFT; len = le32_to_cpu(rxcmp->rx_cmp_len_flags_type) >> RX_CMP_LEN_SHIFT;
dma_addr = rx_buf->mapping; dma_addr = rx_buf->mapping;
if (bnxt_rx_xdp(bp, rxr, cons, data, &data_ptr, &len, event)) {
rc = 1;
goto next_rx;
}
if (len <= bp->rx_copy_thresh) { if (len <= bp->rx_copy_thresh) {
skb = bnxt_copy_skb(bnapi, data_ptr, len, dma_addr); skb = bnxt_copy_skb(bnapi, data_ptr, len, dma_addr);
bnxt_reuse_rx_data(rxr, cons, data); bnxt_reuse_rx_data(rxr, cons, data);
...@@ -1490,7 +1496,10 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons, ...@@ -1490,7 +1496,10 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
} else { } else {
u32 payload; u32 payload;
if (rx_buf->data_ptr == data_ptr)
payload = misc & RX_CMP_PAYLOAD_OFFSET; payload = misc & RX_CMP_PAYLOAD_OFFSET;
else
payload = 0;
skb = bp->rx_skb_func(bp, rxr, cons, data, data_ptr, dma_addr, skb = bp->rx_skb_func(bp, rxr, cons, data, data_ptr, dma_addr,
payload | len); payload | len);
if (!skb) { if (!skb) {
...@@ -2080,6 +2089,9 @@ static void bnxt_free_rx_rings(struct bnxt *bp) ...@@ -2080,6 +2089,9 @@ static void bnxt_free_rx_rings(struct bnxt *bp)
struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i]; struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
struct bnxt_ring_struct *ring; struct bnxt_ring_struct *ring;
if (rxr->xdp_prog)
bpf_prog_put(rxr->xdp_prog);
kfree(rxr->rx_tpa); kfree(rxr->rx_tpa);
rxr->rx_tpa = NULL; rxr->rx_tpa = NULL;
...@@ -2367,6 +2379,15 @@ static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr) ...@@ -2367,6 +2379,15 @@ static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
ring = &rxr->rx_ring_struct; ring = &rxr->rx_ring_struct;
bnxt_init_rxbd_pages(ring, type); bnxt_init_rxbd_pages(ring, type);
if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) {
rxr->xdp_prog = bpf_prog_add(bp->xdp_prog, 1);
if (IS_ERR(rxr->xdp_prog)) {
int rc = PTR_ERR(rxr->xdp_prog);
rxr->xdp_prog = NULL;
return rc;
}
}
prod = rxr->rx_prod; prod = rxr->rx_prod;
for (i = 0; i < bp->rx_ring_size; i++) { for (i = 0; i < bp->rx_ring_size; i++) {
if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL) != 0) { if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL) != 0) {
...@@ -2430,8 +2451,8 @@ static int bnxt_init_rx_rings(struct bnxt *bp) ...@@ -2430,8 +2451,8 @@ static int bnxt_init_rx_rings(struct bnxt *bp)
int i, rc = 0; int i, rc = 0;
if (BNXT_RX_PAGE_MODE(bp)) { if (BNXT_RX_PAGE_MODE(bp)) {
bp->rx_offset = NET_IP_ALIGN; bp->rx_offset = NET_IP_ALIGN + XDP_PACKET_HEADROOM;
bp->rx_dma_offset = 0; bp->rx_dma_offset = XDP_PACKET_HEADROOM;
} else { } else {
bp->rx_offset = BNXT_RX_OFFSET; bp->rx_offset = BNXT_RX_OFFSET;
bp->rx_dma_offset = BNXT_RX_DMA_OFFSET; bp->rx_dma_offset = BNXT_RX_DMA_OFFSET;
...@@ -2560,7 +2581,7 @@ static int bnxt_calc_nr_ring_pages(u32 ring_size, int desc_per_pg) ...@@ -2560,7 +2581,7 @@ static int bnxt_calc_nr_ring_pages(u32 ring_size, int desc_per_pg)
return pages; return pages;
} }
static void bnxt_set_tpa_flags(struct bnxt *bp) void bnxt_set_tpa_flags(struct bnxt *bp)
{ {
bp->flags &= ~BNXT_FLAG_TPA; bp->flags &= ~BNXT_FLAG_TPA;
if (bp->flags & BNXT_FLAG_NO_AGG_RINGS) if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
...@@ -7107,6 +7128,7 @@ static const struct net_device_ops bnxt_netdev_ops = { ...@@ -7107,6 +7128,7 @@ static const struct net_device_ops bnxt_netdev_ops = {
#endif #endif
.ndo_udp_tunnel_add = bnxt_udp_tunnel_add, .ndo_udp_tunnel_add = bnxt_udp_tunnel_add,
.ndo_udp_tunnel_del = bnxt_udp_tunnel_del, .ndo_udp_tunnel_del = bnxt_udp_tunnel_del,
.ndo_xdp = bnxt_xdp,
}; };
static void bnxt_remove_one(struct pci_dev *pdev) static void bnxt_remove_one(struct pci_dev *pdev)
...@@ -7131,6 +7153,8 @@ static void bnxt_remove_one(struct pci_dev *pdev) ...@@ -7131,6 +7153,8 @@ static void bnxt_remove_one(struct pci_dev *pdev)
pci_iounmap(pdev, bp->bar0); pci_iounmap(pdev, bp->bar0);
kfree(bp->edev); kfree(bp->edev);
bp->edev = NULL; bp->edev = NULL;
if (bp->xdp_prog)
bpf_prog_put(bp->xdp_prog);
free_netdev(dev); free_netdev(dev);
pci_release_regions(pdev); pci_release_regions(pdev);
......
...@@ -418,7 +418,8 @@ struct rx_tpa_end_cmp_ext { ...@@ -418,7 +418,8 @@ struct rx_tpa_end_cmp_ext {
#define BNXT_MAX_MTU 9500 #define BNXT_MAX_MTU 9500
#define BNXT_MAX_PAGE_MODE_MTU \ #define BNXT_MAX_PAGE_MODE_MTU \
((unsigned int)PAGE_SIZE - VLAN_ETH_HLEN - NET_IP_ALIGN) ((unsigned int)PAGE_SIZE - VLAN_ETH_HLEN - NET_IP_ALIGN - \
XDP_PACKET_HEADROOM)
#define BNXT_MIN_PKT_SIZE 52 #define BNXT_MIN_PKT_SIZE 52
...@@ -618,6 +619,8 @@ struct bnxt_rx_ring_info { ...@@ -618,6 +619,8 @@ struct bnxt_rx_ring_info {
void __iomem *rx_doorbell; void __iomem *rx_doorbell;
void __iomem *rx_agg_doorbell; void __iomem *rx_agg_doorbell;
struct bpf_prog *xdp_prog;
struct rx_bd *rx_desc_ring[MAX_RX_PAGES]; struct rx_bd *rx_desc_ring[MAX_RX_PAGES];
struct bnxt_sw_rx_bd *rx_buf_ring; struct bnxt_sw_rx_bd *rx_buf_ring;
...@@ -1167,6 +1170,8 @@ struct bnxt { ...@@ -1167,6 +1170,8 @@ struct bnxt {
u8 num_leds; u8 num_leds;
struct bnxt_led_info leds[BNXT_MAX_LED]; struct bnxt_led_info leds[BNXT_MAX_LED];
struct bpf_prog *xdp_prog;
}; };
#define BNXT_RX_STATS_OFFSET(counter) \ #define BNXT_RX_STATS_OFFSET(counter) \
...@@ -1186,6 +1191,8 @@ struct bnxt { ...@@ -1186,6 +1191,8 @@ struct bnxt {
#define SFF_MODULE_ID_QSFP28 0x11 #define SFF_MODULE_ID_QSFP28 0x11
#define BNXT_MAX_PHY_I2C_RESP_SIZE 64 #define BNXT_MAX_PHY_I2C_RESP_SIZE 64
void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons, void *data);
void bnxt_set_tpa_flags(struct bnxt *bp);
void bnxt_set_ring_params(struct bnxt *); void bnxt_set_ring_params(struct bnxt *);
int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode); int bnxt_set_rx_skb_mode(struct bnxt *bp, bool page_mode);
void bnxt_hwrm_cmd_hdr_init(struct bnxt *, void *, u16, u16, u16); void bnxt_hwrm_cmd_hdr_init(struct bnxt *, void *, u16, u16, u16);
......
/* Broadcom NetXtreme-C/E network driver.
*
* Copyright (c) 2016-2017 Broadcom Limited
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation.
*/
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/if_vlan.h>
#include <linux/bpf.h>
#include <linux/bpf_trace.h>
#include <linux/filter.h>
#include "bnxt_hsi.h"
#include "bnxt.h"
#include "bnxt_xdp.h"
/* returns the following:
* true - packet consumed by XDP and new buffer is allocated.
* false - packet should be passed to the stack.
*/
bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
struct page *page, u8 **data_ptr, unsigned int *len, u8 *event)
{
struct bpf_prog *xdp_prog = READ_ONCE(rxr->xdp_prog);
struct bnxt_sw_rx_bd *rx_buf;
struct pci_dev *pdev;
struct xdp_buff xdp;
dma_addr_t mapping;
void *orig_data;
u32 offset;
u32 act;
if (!xdp_prog)
return false;
pdev = bp->pdev;
rx_buf = &rxr->rx_buf_ring[cons];
offset = bp->rx_offset;
xdp.data_hard_start = *data_ptr - offset;
xdp.data = *data_ptr;
xdp.data_end = *data_ptr + *len;
orig_data = xdp.data;
mapping = rx_buf->mapping - bp->rx_dma_offset;
dma_sync_single_for_cpu(&pdev->dev, mapping + offset, *len, bp->rx_dir);
rcu_read_lock();
act = bpf_prog_run_xdp(xdp_prog, &xdp);
rcu_read_unlock();
if (orig_data != xdp.data) {
offset = xdp.data - xdp.data_hard_start;
*data_ptr = xdp.data_hard_start + offset;
*len = xdp.data_end - xdp.data;
}
switch (act) {
case XDP_PASS:
return false;
default:
bpf_warn_invalid_xdp_action(act);
/* Fall thru */
case XDP_ABORTED:
trace_xdp_exception(bp->dev, xdp_prog, act);
/* Fall thru */
case XDP_DROP:
bnxt_reuse_rx_data(rxr, cons, page);
break;
}
return true;
}
/* Under rtnl_lock */
static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
{
struct net_device *dev = bp->dev;
int tx_xdp = 0, rc, tc;
struct bpf_prog *old;
if (prog && bp->dev->mtu > BNXT_MAX_PAGE_MODE_MTU) {
netdev_warn(dev, "MTU %d larger than largest XDP supported MTU %d.\n",
bp->dev->mtu, BNXT_MAX_PAGE_MODE_MTU);
return -EOPNOTSUPP;
}
if (!(bp->flags & BNXT_FLAG_SHARED_RINGS)) {
netdev_warn(dev, "ethtool rx/tx channels must be combined to support XDP.\n");
return -EOPNOTSUPP;
}
if (prog)
tx_xdp = bp->rx_nr_rings;
tc = netdev_get_num_tc(dev);
if (!tc)
tc = 1;
rc = bnxt_reserve_rings(bp, bp->tx_nr_rings_per_tc, bp->rx_nr_rings,
tc, tx_xdp);
if (rc) {
netdev_warn(dev, "Unable to reserve enough TX rings to support XDP.\n");
return rc;
}
if (netif_running(dev))
bnxt_close_nic(bp, true, false);
old = xchg(&bp->xdp_prog, prog);
if (old)
bpf_prog_put(old);
if (prog) {
bnxt_set_rx_skb_mode(bp, true);
} else {
int rx, tx;
bnxt_set_rx_skb_mode(bp, false);
bnxt_get_max_rings(bp, &rx, &tx, true);
if (rx > 1) {
bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS;
bp->dev->hw_features |= NETIF_F_LRO;
}
}
bp->tx_nr_rings_xdp = tx_xdp;
bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tc + tx_xdp;
bp->cp_nr_rings = max_t(int, bp->tx_nr_rings, bp->rx_nr_rings);
bp->num_stat_ctxs = bp->cp_nr_rings;
bnxt_set_tpa_flags(bp);
bnxt_set_ring_params(bp);
if (netif_running(dev))
return bnxt_open_nic(bp, true, false);
return 0;
}
int bnxt_xdp(struct net_device *dev, struct netdev_xdp *xdp)
{
struct bnxt *bp = netdev_priv(dev);
int rc;
switch (xdp->command) {
case XDP_SETUP_PROG:
rc = bnxt_xdp_set(bp, xdp->prog);
break;
case XDP_QUERY_PROG:
xdp->prog_attached = !!bp->xdp_prog;
rc = 0;
break;
default:
rc = -EINVAL;
break;
}
return rc;
}
/* Broadcom NetXtreme-C/E network driver.
*
* Copyright (c) 2016-2017 Broadcom Limited
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation.
*/
#ifndef BNXT_XDP_H
#define BNXT_XDP_H
bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons,
struct page *page, u8 **data_ptr, unsigned int *len,
u8 *event);
int bnxt_xdp(struct net_device *dev, struct netdev_xdp *xdp);
#endif
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment