* [PATCH net-next 00/13] nfp: XDP adjust head support
@ 2017-03-10 18:38 Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 01/13] nfp: separate data path information from the reset of adapter structure Jakub Kicinski
` (13 more replies)
0 siblings, 14 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Hi!
This series adds support for XDP adjust head. Bulk of the code
is actually just paying technical debt. On reconfiguration request
nfp was allocating new resources separately leaving device running
with the existing set of rings. We used to manage the new resources
in special ring set structures. This set is simply separating the
datapath part of the device structure from the control information
allowing the new datapath structure to be allocated with all new
memory and rings. The swap operation is now greatly simplified.
We also save a lot of parameter passing this way. Hopefully the churn
is worth the negative diffstat.
Support for XDP adjust head is done in a pretty standard way. NFP
is a bit special because it prepends metadata before packet data
so we have to do a bit of memcpying in case XDP will run. We also
luck out a little bit because the fact that we already have prepend
space allocated means that one byte is enough to store the extra XDP
space (256 of standard prepend space is a bit inconvenient since
it would normally require 16bits or boolean with additional shifts).
Jakub Kicinski (13):
nfp: separate data path information from the reset of adapter
structure
nfp: move control BAR pointer into data path structure
nfp: pass new data path to ring reconfig
nfp: use dp to carry number of stack tx rings and vectors
nfp: use dp to carry fl_bufsz at reconfig time
nfp: use dp to carry mtu at reconfig time
nfp: use dp to carry xdp_prog at reconfig time
nfp: switch to using data path structures for reconfiguration
nfp: store dma direction in data path structure
nfp: validate rx offset from the BAR and size down it's field
nfp: reorganize pkt_off variable
nfp: prepare metadata handling for xdp_adjust_head()
nfp: add support for xdp_adjust_head()
drivers/net/ethernet/netronome/nfp/nfp_net.h | 150 ++--
.../net/ethernet/netronome/nfp/nfp_net_common.c | 999 ++++++++++-----------
.../net/ethernet/netronome/nfp/nfp_net_debugfs.c | 4 +-
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 104 +--
drivers/net/ethernet/netronome/nfp/nfp_net_main.c | 32 +-
.../net/ethernet/netronome/nfp/nfp_net_offload.c | 30 +-
.../net/ethernet/netronome/nfp/nfp_netvf_main.c | 19 +-
7 files changed, 655 insertions(+), 683 deletions(-)
--
2.11.0
^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH net-next 01/13] nfp: separate data path information from the reset of adapter structure
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 02/13] nfp: move control BAR pointer into data path structure Jakub Kicinski
` (12 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Move all data path information into a separate structure. This way
we will be able to allocate new data path with all new rings etc.
and swap it in easily.
No functional changes.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 105 ++--
.../net/ethernet/netronome/nfp/nfp_net_common.c | 602 +++++++++++----------
.../net/ethernet/netronome/nfp/nfp_net_debugfs.c | 4 +-
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 64 +--
drivers/net/ethernet/netronome/nfp/nfp_net_main.c | 30 +-
.../net/ethernet/netronome/nfp/nfp_net_offload.c | 30 +-
.../net/ethernet/netronome/nfp/nfp_netvf_main.c | 15 +-
7 files changed, 436 insertions(+), 414 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 34f8c439f42f..7d2c38604372 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -50,14 +50,14 @@
#include "nfp_net_ctrl.h"
-#define nn_err(nn, fmt, args...) netdev_err((nn)->netdev, fmt, ## args)
-#define nn_warn(nn, fmt, args...) netdev_warn((nn)->netdev, fmt, ## args)
-#define nn_info(nn, fmt, args...) netdev_info((nn)->netdev, fmt, ## args)
-#define nn_dbg(nn, fmt, args...) netdev_dbg((nn)->netdev, fmt, ## args)
-#define nn_warn_ratelimit(nn, fmt, args...) \
+#define nn_err(nn, fmt, args...) netdev_err((nn)->dp.netdev, fmt, ## args)
+#define nn_warn(nn, fmt, args...) netdev_warn((nn)->dp.netdev, fmt, ## args)
+#define nn_info(nn, fmt, args...) netdev_info((nn)->dp.netdev, fmt, ## args)
+#define nn_dbg(nn, fmt, args...) netdev_dbg((nn)->dp.netdev, fmt, ## args)
+#define nn_dp_warn(dp, fmt, args...) \
do { \
if (unlikely(net_ratelimit())) \
- netdev_warn((nn)->netdev, fmt, ## args); \
+ netdev_warn((dp)->netdev, fmt, ## args); \
} while (0)
/* Max time to wait for NFP to respond on updates (in seconds) */
@@ -434,18 +434,62 @@ struct nfp_stat_pair {
};
/**
- * struct nfp_net - NFP network device structure
+ * struct nfp_net_dp - NFP network device datapath data structure
* @dev: Backpointer to struct device
- * @netdev: Backpointer to net_device structure
- * @is_vf: Is the driver attached to a VF?
+ * @netdev: Backpointer to net_device structure
+ * @is_vf: Is the driver attached to a VF?
* @bpf_offload_skip_sw: Offloaded BPF program will not be rerun by cls_bpf
* @bpf_offload_xdp: Offloaded BPF program is XDP
* @chained_metadata_format: Firemware will use new metadata format
- * @ctrl: Local copy of the control register/word.
- * @fl_bufsz: Currently configured size of the freelist buffers
+ * @ctrl: Local copy of the control register/word.
+ * @fl_bufsz: Currently configured size of the freelist buffers
* @rx_offset: Offset in the RX buffers where packet data starts
* @xdp_prog: Installed XDP program
- * @fw_ver: Firmware version
+ * @tx_rings: Array of pre-allocated TX ring structures
+ * @rx_rings: Array of pre-allocated RX ring structures
+ *
+ * @txd_cnt: Size of the TX ring in number of descriptors
+ * @rxd_cnt: Size of the RX ring in number of descriptors
+ * @num_r_vecs: Number of used ring vectors
+ * @num_tx_rings: Currently configured number of TX rings
+ * @num_stack_tx_rings: Number of TX rings used by the stack (not XDP)
+ * @num_rx_rings: Currently configured number of RX rings
+ */
+struct nfp_net_dp {
+ struct device *dev;
+ struct net_device *netdev;
+
+ unsigned is_vf:1;
+ unsigned bpf_offload_skip_sw:1;
+ unsigned bpf_offload_xdp:1;
+ unsigned chained_metadata_format:1;
+
+ u32 ctrl;
+ u32 fl_bufsz;
+
+ u32 rx_offset;
+
+ struct bpf_prog *xdp_prog;
+
+ struct nfp_net_tx_ring *tx_rings;
+ struct nfp_net_rx_ring *rx_rings;
+
+ /* Cold data follows */
+
+ unsigned int txd_cnt;
+ unsigned int rxd_cnt;
+
+ unsigned int num_r_vecs;
+
+ unsigned int num_tx_rings;
+ unsigned int num_stack_tx_rings;
+ unsigned int num_rx_rings;
+};
+
+/**
+ * struct nfp_net - NFP network device structure
+ * @dp: Datapath structure
+ * @fw_ver: Firmware version
* @cap: Capabilities advertised by the Firmware
* @max_mtu: Maximum support MTU advertised by the Firmware
* @rss_hfunc: RSS selected hash function
@@ -457,17 +501,9 @@ struct nfp_stat_pair {
* @rx_filter_change: Jiffies when statistics last changed
* @rx_filter_stats_timer: Timer for polling filter offload statistics
* @rx_filter_lock: Lock protecting timer state changes (teardown)
+ * @max_r_vecs: Number of allocated interrupt vectors for RX/TX
* @max_tx_rings: Maximum number of TX rings supported by the Firmware
* @max_rx_rings: Maximum number of RX rings supported by the Firmware
- * @num_tx_rings: Currently configured number of TX rings
- * @num_stack_tx_rings: Number of TX rings used by the stack (not XDP)
- * @num_rx_rings: Currently configured number of RX rings
- * @txd_cnt: Size of the TX ring in number of descriptors
- * @rxd_cnt: Size of the RX ring in number of descriptors
- * @tx_rings: Array of pre-allocated TX ring structures
- * @rx_rings: Array of pre-allocated RX ring structures
- * @max_r_vecs: Number of allocated interrupt vectors for RX/TX
- * @num_r_vecs: Number of used ring vectors
* @r_vecs: Pre-allocated array of ring vectors
* @irq_entries: Pre-allocated array of MSI-X entries
* @lsc_handler: Handler for Link State Change interrupt
@@ -502,25 +538,10 @@ struct nfp_stat_pair {
* @eth_port: Translated ETH Table port entry
*/
struct nfp_net {
- struct device *dev;
- struct net_device *netdev;
-
- unsigned is_vf:1;
- unsigned bpf_offload_skip_sw:1;
- unsigned bpf_offload_xdp:1;
- unsigned chained_metadata_format:1;
-
- u32 ctrl;
- u32 fl_bufsz;
-
- u32 rx_offset;
-
- struct bpf_prog *xdp_prog;
-
- struct nfp_net_tx_ring *tx_rings;
- struct nfp_net_rx_ring *rx_rings;
+ struct nfp_net_dp dp;
struct nfp_net_fw_version fw_ver;
+
u32 cap;
u32 max_mtu;
@@ -537,18 +558,10 @@ struct nfp_net {
unsigned int max_tx_rings;
unsigned int max_rx_rings;
- unsigned int num_tx_rings;
- unsigned int num_stack_tx_rings;
- unsigned int num_rx_rings;
-
int stride_tx;
int stride_rx;
- int txd_cnt;
- int rxd_cnt;
-
unsigned int max_r_vecs;
- unsigned int num_r_vecs;
struct nfp_net_r_vector r_vecs[NFP_NET_MAX_R_VECS];
struct msix_entry irq_entries[NFP_NET_MAX_IRQS];
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 2d964d030dbe..951d511643f1 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -86,18 +86,18 @@ void nfp_net_get_fw_version(struct nfp_net_fw_version *fw_ver,
}
static dma_addr_t
-nfp_net_dma_map_rx(struct nfp_net *nn, void *frag, unsigned int bufsz,
+nfp_net_dma_map_rx(struct nfp_net_dp *dp, void *frag, unsigned int bufsz,
int direction)
{
- return dma_map_single(nn->dev, frag + NFP_NET_RX_BUF_HEADROOM,
+ return dma_map_single(dp->dev, frag + NFP_NET_RX_BUF_HEADROOM,
bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
}
static void
-nfp_net_dma_unmap_rx(struct nfp_net *nn, dma_addr_t dma_addr,
+nfp_net_dma_unmap_rx(struct nfp_net_dp *dp, dma_addr_t dma_addr,
unsigned int bufsz, int direction)
{
- dma_unmap_single(nn->dev, dma_addr,
+ dma_unmap_single(dp->dev, dma_addr,
bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
}
@@ -329,19 +329,22 @@ void
nfp_net_irqs_assign(struct nfp_net *nn, struct msix_entry *irq_entries,
unsigned int n)
{
+ struct nfp_net_dp *dp = &nn->dp;
+
nn->max_r_vecs = n - NFP_NET_NON_Q_VECTORS;
- nn->num_r_vecs = nn->max_r_vecs;
+ dp->num_r_vecs = nn->max_r_vecs;
memcpy(nn->irq_entries, irq_entries, sizeof(*irq_entries) * n);
- if (nn->num_rx_rings > nn->num_r_vecs ||
- nn->num_tx_rings > nn->num_r_vecs)
+ if (dp->num_rx_rings > dp->num_r_vecs ||
+ dp->num_tx_rings > dp->num_r_vecs)
nn_warn(nn, "More rings (%d,%d) than vectors (%d).\n",
- nn->num_rx_rings, nn->num_tx_rings, nn->num_r_vecs);
+ dp->num_rx_rings, dp->num_tx_rings,
+ dp->num_r_vecs);
- nn->num_rx_rings = min(nn->num_r_vecs, nn->num_rx_rings);
- nn->num_tx_rings = min(nn->num_r_vecs, nn->num_tx_rings);
- nn->num_stack_tx_rings = nn->num_tx_rings;
+ dp->num_rx_rings = min(dp->num_r_vecs, dp->num_rx_rings);
+ dp->num_tx_rings = min(dp->num_r_vecs, dp->num_tx_rings);
+ dp->num_stack_tx_rings = dp->num_tx_rings;
}
/**
@@ -396,11 +399,11 @@ static void nfp_net_read_link_status(struct nfp_net *nn)
nn->link_up = link_up;
if (nn->link_up) {
- netif_carrier_on(nn->netdev);
- netdev_info(nn->netdev, "NIC Link is Up\n");
+ netif_carrier_on(nn->dp.netdev);
+ netdev_info(nn->dp.netdev, "NIC Link is Up\n");
} else {
- netif_carrier_off(nn->netdev);
- netdev_info(nn->netdev, "NIC Link is Down\n");
+ netif_carrier_off(nn->dp.netdev);
+ netdev_info(nn->dp.netdev, "NIC Link is Down\n");
}
out:
spin_unlock_irqrestore(&nn->link_status_lock, flags);
@@ -532,7 +535,7 @@ nfp_net_aux_irq_request(struct nfp_net *nn, u32 ctrl_offset,
entry = &nn->irq_entries[vector_idx];
- snprintf(name, name_sz, format, netdev_name(nn->netdev));
+ snprintf(name, name_sz, format, netdev_name(nn->dp.netdev));
err = request_irq(entry->vector, handler, 0, name, nn);
if (err) {
nn_err(nn, "Failed to request IRQ %d (err=%d).\n",
@@ -619,7 +622,6 @@ static void nfp_net_tx_ring_stop(struct netdev_queue *nd_q,
/**
* nfp_net_tx_tso() - Set up Tx descriptor for LSO
- * @nn: NFP Net device
* @r_vec: per-ring structure
* @txbuf: Pointer to driver soft TX descriptor
* @txd: Pointer to HW TX descriptor
@@ -628,7 +630,7 @@ static void nfp_net_tx_ring_stop(struct netdev_queue *nd_q,
* Set up Tx descriptor for LSO, do nothing for non-LSO skbs.
* Return error on packet header greater than maximum supported LSO header size.
*/
-static void nfp_net_tx_tso(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
+static void nfp_net_tx_tso(struct nfp_net_r_vector *r_vec,
struct nfp_net_tx_buf *txbuf,
struct nfp_net_tx_desc *txd, struct sk_buff *skb)
{
@@ -659,7 +661,7 @@ static void nfp_net_tx_tso(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
/**
* nfp_net_tx_csum() - Set TX CSUM offload flags in TX descriptor
- * @nn: NFP Net device
+ * @dp: NFP Net data path struct
* @r_vec: per-ring structure
* @txbuf: Pointer to driver soft TX descriptor
* @txd: Pointer to TX descriptor
@@ -668,7 +670,8 @@ static void nfp_net_tx_tso(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
* This function sets the TX checksum flags in the TX descriptor based
* on the configuration and the protocol of the packet to be transmitted.
*/
-static void nfp_net_tx_csum(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
+static void nfp_net_tx_csum(struct nfp_net_dp *dp,
+ struct nfp_net_r_vector *r_vec,
struct nfp_net_tx_buf *txbuf,
struct nfp_net_tx_desc *txd, struct sk_buff *skb)
{
@@ -676,7 +679,7 @@ static void nfp_net_tx_csum(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
struct iphdr *iph;
u8 l4_hdr;
- if (!(nn->ctrl & NFP_NET_CFG_CTRL_TXCSUM))
+ if (!(dp->ctrl & NFP_NET_CFG_CTRL_TXCSUM))
return;
if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -695,8 +698,7 @@ static void nfp_net_tx_csum(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
} else if (ipv6h->version == 6) {
l4_hdr = ipv6h->nexthdr;
} else {
- nn_warn_ratelimit(nn, "partial checksum but ipv=%x!\n",
- iph->version);
+ nn_dp_warn(dp, "partial checksum but ipv=%x!\n", iph->version);
return;
}
@@ -708,8 +710,7 @@ static void nfp_net_tx_csum(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
txd->flags |= PCIE_DESC_TX_UDP_CSUM;
break;
default:
- nn_warn_ratelimit(nn, "partial checksum but l4 proto=%x!\n",
- l4_hdr);
+ nn_dp_warn(dp, "partial checksum but l4 proto=%x!\n", l4_hdr);
return;
}
@@ -744,22 +745,24 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
struct nfp_net_r_vector *r_vec;
struct nfp_net_tx_buf *txbuf;
struct netdev_queue *nd_q;
+ struct nfp_net_dp *dp;
dma_addr_t dma_addr;
unsigned int fsize;
int f, nr_frags;
int wr_idx;
u16 qidx;
+ dp = &nn->dp;
qidx = skb_get_queue_mapping(skb);
- tx_ring = &nn->tx_rings[qidx];
+ tx_ring = &dp->tx_rings[qidx];
r_vec = tx_ring->r_vec;
- nd_q = netdev_get_tx_queue(nn->netdev, qidx);
+ nd_q = netdev_get_tx_queue(dp->netdev, qidx);
nr_frags = skb_shinfo(skb)->nr_frags;
if (unlikely(nfp_net_tx_full(tx_ring, nr_frags + 1))) {
- nn_warn_ratelimit(nn, "TX ring %d busy. wrp=%u rdp=%u\n",
- qidx, tx_ring->wr_p, tx_ring->rd_p);
+ nn_dp_warn(dp, "TX ring %d busy. wrp=%u rdp=%u\n",
+ qidx, tx_ring->wr_p, tx_ring->rd_p);
netif_tx_stop_queue(nd_q);
u64_stats_update_begin(&r_vec->tx_sync);
r_vec->tx_busy++;
@@ -768,9 +771,9 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
}
/* Start with the head skbuf */
- dma_addr = dma_map_single(nn->dev, skb->data, skb_headlen(skb),
+ dma_addr = dma_map_single(dp->dev, skb->data, skb_headlen(skb),
DMA_TO_DEVICE);
- if (dma_mapping_error(nn->dev, dma_addr))
+ if (dma_mapping_error(dp->dev, dma_addr))
goto err_free;
wr_idx = tx_ring->wr_p & (tx_ring->cnt - 1);
@@ -794,11 +797,11 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
txd->mss = 0;
txd->l4_offset = 0;
- nfp_net_tx_tso(nn, r_vec, txbuf, txd, skb);
+ nfp_net_tx_tso(r_vec, txbuf, txd, skb);
- nfp_net_tx_csum(nn, r_vec, txbuf, txd, skb);
+ nfp_net_tx_csum(dp, r_vec, txbuf, txd, skb);
- if (skb_vlan_tag_present(skb) && nn->ctrl & NFP_NET_CFG_CTRL_TXVLAN) {
+ if (skb_vlan_tag_present(skb) && dp->ctrl & NFP_NET_CFG_CTRL_TXVLAN) {
txd->flags |= PCIE_DESC_TX_VLAN;
txd->vlan = cpu_to_le16(skb_vlan_tag_get(skb));
}
@@ -812,9 +815,9 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
frag = &skb_shinfo(skb)->frags[f];
fsize = skb_frag_size(frag);
- dma_addr = skb_frag_dma_map(nn->dev, frag, 0,
+ dma_addr = skb_frag_dma_map(dp->dev, frag, 0,
fsize, DMA_TO_DEVICE);
- if (dma_mapping_error(nn->dev, dma_addr))
+ if (dma_mapping_error(dp->dev, dma_addr))
goto err_unmap;
wr_idx = (wr_idx + 1) & (tx_ring->cnt - 1);
@@ -853,7 +856,7 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
--f;
while (f >= 0) {
frag = &skb_shinfo(skb)->frags[f];
- dma_unmap_page(nn->dev, tx_ring->txbufs[wr_idx].dma_addr,
+ dma_unmap_page(dp->dev, tx_ring->txbufs[wr_idx].dma_addr,
skb_frag_size(frag), DMA_TO_DEVICE);
tx_ring->txbufs[wr_idx].skb = NULL;
tx_ring->txbufs[wr_idx].dma_addr = 0;
@@ -862,13 +865,13 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
if (wr_idx < 0)
wr_idx += tx_ring->cnt;
}
- dma_unmap_single(nn->dev, tx_ring->txbufs[wr_idx].dma_addr,
+ dma_unmap_single(dp->dev, tx_ring->txbufs[wr_idx].dma_addr,
skb_headlen(skb), DMA_TO_DEVICE);
tx_ring->txbufs[wr_idx].skb = NULL;
tx_ring->txbufs[wr_idx].dma_addr = 0;
tx_ring->txbufs[wr_idx].fidx = -2;
err_free:
- nn_warn_ratelimit(nn, "Failed to map DMA TX buffer\n");
+ nn_dp_warn(dp, "Failed to map DMA TX buffer\n");
u64_stats_update_begin(&r_vec->tx_sync);
r_vec->tx_errors++;
u64_stats_update_end(&r_vec->tx_sync);
@@ -885,7 +888,7 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev)
static void nfp_net_tx_complete(struct nfp_net_tx_ring *tx_ring)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
const struct skb_frag_struct *frag;
struct netdev_queue *nd_q;
u32 done_pkts = 0, done_bytes = 0;
@@ -919,7 +922,7 @@ static void nfp_net_tx_complete(struct nfp_net_tx_ring *tx_ring)
if (fidx == -1) {
/* unmap head */
- dma_unmap_single(nn->dev, tx_ring->txbufs[idx].dma_addr,
+ dma_unmap_single(dp->dev, tx_ring->txbufs[idx].dma_addr,
skb_headlen(skb), DMA_TO_DEVICE);
done_pkts += tx_ring->txbufs[idx].pkt_cnt;
@@ -927,7 +930,7 @@ static void nfp_net_tx_complete(struct nfp_net_tx_ring *tx_ring)
} else {
/* unmap fragment */
frag = &skb_shinfo(skb)->frags[fidx];
- dma_unmap_page(nn->dev, tx_ring->txbufs[idx].dma_addr,
+ dma_unmap_page(dp->dev, tx_ring->txbufs[idx].dma_addr,
skb_frag_size(frag), DMA_TO_DEVICE);
}
@@ -947,7 +950,7 @@ static void nfp_net_tx_complete(struct nfp_net_tx_ring *tx_ring)
r_vec->tx_pkts += done_pkts;
u64_stats_update_end(&r_vec->tx_sync);
- nd_q = netdev_get_tx_queue(nn->netdev, tx_ring->idx);
+ nd_q = netdev_get_tx_queue(dp->netdev, tx_ring->idx);
netdev_tx_completed_queue(nd_q, done_pkts, done_bytes);
if (nfp_net_tx_ring_should_wake(tx_ring)) {
/* Make sure TX thread will see updated tx_ring->rd_p */
@@ -965,7 +968,7 @@ static void nfp_net_tx_complete(struct nfp_net_tx_ring *tx_ring)
static void nfp_net_xdp_complete(struct nfp_net_tx_ring *tx_ring)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
u32 done_pkts = 0, done_bytes = 0;
int idx, todo;
u32 qcp_rd_p;
@@ -988,8 +991,8 @@ static void nfp_net_xdp_complete(struct nfp_net_tx_ring *tx_ring)
if (!tx_ring->txbufs[idx].frag)
continue;
- nfp_net_dma_unmap_rx(nn, tx_ring->txbufs[idx].dma_addr,
- nn->fl_bufsz, DMA_BIDIRECTIONAL);
+ nfp_net_dma_unmap_rx(dp, tx_ring->txbufs[idx].dma_addr,
+ dp->fl_bufsz, DMA_BIDIRECTIONAL);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
done_pkts++;
@@ -1014,13 +1017,13 @@ static void nfp_net_xdp_complete(struct nfp_net_tx_ring *tx_ring)
/**
* nfp_net_tx_ring_reset() - Free any untransmitted buffers and reset pointers
- * @nn: NFP Net device
+ * @dp: NFP Net data path struct
* @tx_ring: TX ring structure
*
* Assumes that the device is stopped
*/
static void
-nfp_net_tx_ring_reset(struct nfp_net *nn, struct nfp_net_tx_ring *tx_ring)
+nfp_net_tx_ring_reset(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
const struct skb_frag_struct *frag;
@@ -1034,8 +1037,8 @@ nfp_net_tx_ring_reset(struct nfp_net *nn, struct nfp_net_tx_ring *tx_ring)
tx_buf = &tx_ring->txbufs[idx];
if (tx_ring == r_vec->xdp_ring) {
- nfp_net_dma_unmap_rx(nn, tx_buf->dma_addr,
- nn->fl_bufsz, DMA_BIDIRECTIONAL);
+ nfp_net_dma_unmap_rx(dp, tx_buf->dma_addr,
+ dp->fl_bufsz, DMA_BIDIRECTIONAL);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
} else {
struct sk_buff *skb = tx_ring->txbufs[idx].skb;
@@ -1043,13 +1046,13 @@ nfp_net_tx_ring_reset(struct nfp_net *nn, struct nfp_net_tx_ring *tx_ring)
if (tx_buf->fidx == -1) {
/* unmap head */
- dma_unmap_single(nn->dev, tx_buf->dma_addr,
+ dma_unmap_single(dp->dev, tx_buf->dma_addr,
skb_headlen(skb),
DMA_TO_DEVICE);
} else {
/* unmap fragment */
frag = &skb_shinfo(skb)->frags[tx_buf->fidx];
- dma_unmap_page(nn->dev, tx_buf->dma_addr,
+ dma_unmap_page(dp->dev, tx_buf->dma_addr,
skb_frag_size(frag),
DMA_TO_DEVICE);
}
@@ -1076,7 +1079,7 @@ nfp_net_tx_ring_reset(struct nfp_net *nn, struct nfp_net_tx_ring *tx_ring)
if (tx_ring == r_vec->xdp_ring)
return;
- nd_q = netdev_get_tx_queue(nn->netdev, tx_ring->idx);
+ nd_q = netdev_get_tx_queue(dp->netdev, tx_ring->idx);
netdev_tx_reset_queue(nd_q);
}
@@ -1085,7 +1088,7 @@ static void nfp_net_tx_timeout(struct net_device *netdev)
struct nfp_net *nn = netdev_priv(netdev);
int i;
- for (i = 0; i < nn->netdev->real_num_tx_queues; i++) {
+ for (i = 0; i < nn->dp.netdev->real_num_tx_queues; i++) {
if (!netif_tx_queue_stopped(netdev_get_tx_queue(netdev, i)))
continue;
nn_warn(nn, "TX timeout on ring: %d\n", i);
@@ -1096,15 +1099,15 @@ static void nfp_net_tx_timeout(struct net_device *netdev)
/* Receive processing
*/
static unsigned int
-nfp_net_calc_fl_bufsz(struct nfp_net *nn, unsigned int mtu)
+nfp_net_calc_fl_bufsz(struct nfp_net_dp *dp, unsigned int mtu)
{
unsigned int fl_bufsz;
fl_bufsz = NFP_NET_RX_BUF_HEADROOM;
- if (nn->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
+ if (dp->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
fl_bufsz += NFP_NET_MAX_PREPEND;
else
- fl_bufsz += nn->rx_offset;
+ fl_bufsz += dp->rx_offset;
fl_bufsz += ETH_HLEN + VLAN_HLEN * 2 + mtu;
fl_bufsz = SKB_DATA_ALIGN(fl_bufsz);
@@ -1137,7 +1140,7 @@ static void *
nfp_net_rx_alloc_one(struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
unsigned int fl_bufsz, bool xdp)
{
- struct nfp_net *nn = rx_ring->r_vec->nfp_net;
+ struct nfp_net_dp *dp = &rx_ring->r_vec->nfp_net->dp;
int direction;
void *frag;
@@ -1146,16 +1149,16 @@ nfp_net_rx_alloc_one(struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
else
frag = page_address(alloc_page(GFP_KERNEL | __GFP_COLD));
if (!frag) {
- nn_warn_ratelimit(nn, "Failed to alloc receive page frag\n");
+ nn_dp_warn(dp, "Failed to alloc receive page frag\n");
return NULL;
}
direction = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
- *dma_addr = nfp_net_dma_map_rx(nn, frag, fl_bufsz, direction);
- if (dma_mapping_error(nn->dev, *dma_addr)) {
+ *dma_addr = nfp_net_dma_map_rx(dp, frag, fl_bufsz, direction);
+ if (dma_mapping_error(dp->dev, *dma_addr)) {
nfp_net_free_frag(frag, xdp);
- nn_warn_ratelimit(nn, "Failed to map DMA RX buffer\n");
+ nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
return NULL;
}
@@ -1163,23 +1166,24 @@ nfp_net_rx_alloc_one(struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
}
static void *
-nfp_net_napi_alloc_one(struct nfp_net *nn, int direction, dma_addr_t *dma_addr)
+nfp_net_napi_alloc_one(struct nfp_net_dp *dp, int direction,
+ dma_addr_t *dma_addr)
{
void *frag;
- if (!nn->xdp_prog)
- frag = napi_alloc_frag(nn->fl_bufsz);
+ if (!dp->xdp_prog)
+ frag = napi_alloc_frag(dp->fl_bufsz);
else
frag = page_address(alloc_page(GFP_ATOMIC | __GFP_COLD));
if (!frag) {
- nn_warn_ratelimit(nn, "Failed to alloc receive page frag\n");
+ nn_dp_warn(dp, "Failed to alloc receive page frag\n");
return NULL;
}
- *dma_addr = nfp_net_dma_map_rx(nn, frag, nn->fl_bufsz, direction);
- if (dma_mapping_error(nn->dev, *dma_addr)) {
- nfp_net_free_frag(frag, nn->xdp_prog);
- nn_warn_ratelimit(nn, "Failed to map DMA RX buffer\n");
+ *dma_addr = nfp_net_dma_map_rx(dp, frag, dp->fl_bufsz, direction);
+ if (dma_mapping_error(dp->dev, *dma_addr)) {
+ nfp_net_free_frag(frag, dp->xdp_prog);
+ nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
return NULL;
}
@@ -1247,7 +1251,7 @@ static void nfp_net_rx_ring_reset(struct nfp_net_rx_ring *rx_ring)
/**
* nfp_net_rx_ring_bufs_free() - Free any buffers currently on the RX ring
- * @nn: NFP Net device
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring to remove buffers from
* @xdp: Whether XDP is enabled
*
@@ -1256,8 +1260,8 @@ static void nfp_net_rx_ring_reset(struct nfp_net_rx_ring *rx_ring)
* to restore required ring geometry.
*/
static void
-nfp_net_rx_ring_bufs_free(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
- bool xdp)
+nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
+ struct nfp_net_rx_ring *rx_ring, bool xdp)
{
int direction = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
unsigned int i;
@@ -1270,7 +1274,7 @@ nfp_net_rx_ring_bufs_free(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
if (!rx_ring->rxbufs[i].frag)
continue;
- nfp_net_dma_unmap_rx(nn, rx_ring->rxbufs[i].dma_addr,
+ nfp_net_dma_unmap_rx(dp, rx_ring->rxbufs[i].dma_addr,
rx_ring->bufsz, direction);
nfp_net_free_frag(rx_ring->rxbufs[i].frag, xdp);
rx_ring->rxbufs[i].dma_addr = 0;
@@ -1280,13 +1284,13 @@ nfp_net_rx_ring_bufs_free(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
/**
* nfp_net_rx_ring_bufs_alloc() - Fill RX ring with buffers (don't give to FW)
- * @nn: NFP Net device
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring to remove buffers from
* @xdp: Whether XDP is enabled
*/
static int
-nfp_net_rx_ring_bufs_alloc(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
- bool xdp)
+nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
+ struct nfp_net_rx_ring *rx_ring, bool xdp)
{
struct nfp_net_rx_buf *rxbufs;
unsigned int i;
@@ -1298,7 +1302,7 @@ nfp_net_rx_ring_bufs_alloc(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
nfp_net_rx_alloc_one(rx_ring, &rxbufs[i].dma_addr,
rx_ring->bufsz, xdp);
if (!rxbufs[i].frag) {
- nfp_net_rx_ring_bufs_free(nn, rx_ring, xdp);
+ nfp_net_rx_ring_bufs_free(dp, rx_ring, xdp);
return -ENOMEM;
}
}
@@ -1335,17 +1339,18 @@ static int nfp_net_rx_csum_has_errors(u16 flags)
/**
* nfp_net_rx_csum() - set SKB checksum field based on RX descriptor flags
- * @nn: NFP Net device
+ * @dp: NFP Net data path struct
* @r_vec: per-ring structure
* @rxd: Pointer to RX descriptor
* @skb: Pointer to SKB
*/
-static void nfp_net_rx_csum(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
+static void nfp_net_rx_csum(struct nfp_net_dp *dp,
+ struct nfp_net_r_vector *r_vec,
struct nfp_net_rx_desc *rxd, struct sk_buff *skb)
{
skb_checksum_none_assert(skb);
- if (!(nn->netdev->features & NETIF_F_RXCSUM))
+ if (!(dp->netdev->features & NETIF_F_RXCSUM))
return;
if (nfp_net_rx_csum_has_errors(le16_to_cpu(rxd->rxd.flags))) {
@@ -1462,7 +1467,7 @@ nfp_net_rx_drop(struct nfp_net_r_vector *r_vec, struct nfp_net_rx_ring *rx_ring,
}
static bool
-nfp_net_tx_xdp_buf(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
+nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
struct nfp_net_tx_ring *tx_ring,
struct nfp_net_rx_buf *rxbuf, unsigned int pkt_off,
unsigned int pkt_len)
@@ -1478,7 +1483,7 @@ nfp_net_tx_xdp_buf(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
return false;
}
- new_frag = nfp_net_napi_alloc_one(nn, DMA_BIDIRECTIONAL, &new_dma_addr);
+ new_frag = nfp_net_napi_alloc_one(dp, DMA_BIDIRECTIONAL, &new_dma_addr);
if (unlikely(!new_frag)) {
nfp_net_rx_drop(rx_ring->r_vec, rx_ring, rxbuf, NULL);
return false;
@@ -1495,7 +1500,7 @@ nfp_net_tx_xdp_buf(struct nfp_net *nn, struct nfp_net_rx_ring *rx_ring,
txbuf->pkt_cnt = 1;
txbuf->real_len = pkt_len;
- dma_sync_single_for_device(nn->dev, rxbuf->dma_addr + pkt_off,
+ dma_sync_single_for_device(dp->dev, rxbuf->dma_addr + pkt_off,
pkt_len, DMA_BIDIRECTIONAL);
/* Build TX descriptor */
@@ -1538,7 +1543,7 @@ static int nfp_net_run_xdp(struct bpf_prog *prog, void *data, unsigned int len)
static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
{
struct nfp_net_r_vector *r_vec = rx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
struct nfp_net_tx_ring *tx_ring;
struct bpf_prog *xdp_prog;
unsigned int true_bufsz;
@@ -1548,9 +1553,9 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
int idx;
rcu_read_lock();
- xdp_prog = READ_ONCE(nn->xdp_prog);
+ xdp_prog = READ_ONCE(dp->xdp_prog);
rx_dma_map_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
- true_bufsz = xdp_prog ? PAGE_SIZE : nn->fl_bufsz;
+ true_bufsz = xdp_prog ? PAGE_SIZE : dp->fl_bufsz;
tx_ring = r_vec->xdp_ring;
while (pkts_polled < budget) {
@@ -1591,10 +1596,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
data_len = le16_to_cpu(rxd->rxd.data_len);
pkt_len = data_len - meta_len;
- if (nn->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
+ if (dp->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
pkt_off = meta_len;
else
- pkt_off = nn->rx_offset;
+ pkt_off = dp->rx_offset;
data_off = NFP_NET_RX_BUF_HEADROOM + pkt_off;
/* Stats update */
@@ -1604,10 +1609,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
u64_stats_update_end(&r_vec->rx_sync);
if (xdp_prog && !(rxd->rxd.flags & PCIE_DESC_RX_BPF &&
- nn->bpf_offload_xdp)) {
+ dp->bpf_offload_xdp)) {
int act;
- dma_sync_single_for_cpu(nn->dev,
+ dma_sync_single_for_cpu(dp->dev,
rxbuf->dma_addr + pkt_off,
pkt_len, DMA_BIDIRECTIONAL);
act = nfp_net_run_xdp(xdp_prog, rxbuf->frag + data_off,
@@ -1616,15 +1621,17 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
case XDP_PASS:
break;
case XDP_TX:
- if (unlikely(!nfp_net_tx_xdp_buf(nn, rx_ring,
+ if (unlikely(!nfp_net_tx_xdp_buf(dp, rx_ring,
tx_ring, rxbuf,
- pkt_off, pkt_len)))
- trace_xdp_exception(nn->netdev, xdp_prog, act);
+ pkt_off,
+ pkt_len)))
+ trace_xdp_exception(dp->netdev,
+ xdp_prog, act);
continue;
default:
bpf_warn_invalid_xdp_action(act);
case XDP_ABORTED:
- trace_xdp_exception(nn->netdev, xdp_prog, act);
+ trace_xdp_exception(dp->netdev, xdp_prog, act);
case XDP_DROP:
nfp_net_rx_give_one(rx_ring, rxbuf->frag,
rxbuf->dma_addr);
@@ -1637,14 +1644,14 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
nfp_net_rx_drop(r_vec, rx_ring, rxbuf, NULL);
continue;
}
- new_frag = nfp_net_napi_alloc_one(nn, rx_dma_map_dir,
+ new_frag = nfp_net_napi_alloc_one(dp, rx_dma_map_dir,
&new_dma_addr);
if (unlikely(!new_frag)) {
nfp_net_rx_drop(r_vec, rx_ring, rxbuf, skb);
continue;
}
- nfp_net_dma_unmap_rx(nn, rxbuf->dma_addr, nn->fl_bufsz,
+ nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr, dp->fl_bufsz,
rx_dma_map_dir);
nfp_net_rx_give_one(rx_ring, new_frag, new_dma_addr);
@@ -1652,23 +1659,23 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
skb_reserve(skb, data_off);
skb_put(skb, pkt_len);
- if (!nn->chained_metadata_format) {
- nfp_net_set_hash_desc(nn->netdev, skb, rxd);
+ if (!dp->chained_metadata_format) {
+ nfp_net_set_hash_desc(dp->netdev, skb, rxd);
} else if (meta_len) {
void *end;
- end = nfp_net_parse_meta(nn->netdev, skb, meta_len);
+ end = nfp_net_parse_meta(dp->netdev, skb, meta_len);
if (unlikely(end != skb->data)) {
- nn_warn_ratelimit(nn, "invalid RX packet metadata\n");
+ nn_dp_warn(dp, "invalid RX packet metadata\n");
nfp_net_rx_drop(r_vec, rx_ring, NULL, skb);
continue;
}
}
skb_record_rx_queue(skb, rx_ring->idx);
- skb->protocol = eth_type_trans(skb, nn->netdev);
+ skb->protocol = eth_type_trans(skb, dp->netdev);
- nfp_net_rx_csum(nn, r_vec, rxd, skb);
+ nfp_net_rx_csum(dp, r_vec, rxd, skb);
if (rxd->rxd.flags & PCIE_DESC_RX_VLAN)
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
@@ -1722,12 +1729,12 @@ static int nfp_net_poll(struct napi_struct *napi, int budget)
static void nfp_net_tx_ring_free(struct nfp_net_tx_ring *tx_ring)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
kfree(tx_ring->txbufs);
if (tx_ring->txds)
- dma_free_coherent(nn->dev, tx_ring->size,
+ dma_free_coherent(dp->dev, tx_ring->size,
tx_ring->txds, tx_ring->dma);
tx_ring->cnt = 0;
@@ -1749,13 +1756,13 @@ static int
nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
int sz;
tx_ring->cnt = cnt;
tx_ring->size = sizeof(*tx_ring->txds) * tx_ring->cnt;
- tx_ring->txds = dma_zalloc_coherent(nn->dev, tx_ring->size,
+ tx_ring->txds = dma_zalloc_coherent(dp->dev, tx_ring->size,
&tx_ring->dma, GFP_KERNEL);
if (!tx_ring->txds)
goto err_alloc;
@@ -1766,7 +1773,7 @@ nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
goto err_alloc;
if (!is_xdp)
- netif_set_xps_queue(nn->netdev, &r_vec->affinity_mask,
+ netif_set_xps_queue(dp->netdev, &r_vec->affinity_mask,
tx_ring->idx);
return 0;
@@ -1813,17 +1820,16 @@ nfp_net_tx_ring_set_swap(struct nfp_net *nn, struct nfp_net_ring_set *s)
{
struct nfp_net_ring_set new = *s;
- s->dcnt = nn->txd_cnt;
- s->rings = nn->tx_rings;
- s->n_rings = nn->num_tx_rings;
+ s->dcnt = nn->dp.txd_cnt;
+ s->rings = nn->dp.tx_rings;
+ s->n_rings = nn->dp.num_tx_rings;
- nn->txd_cnt = new.dcnt;
- nn->tx_rings = new.rings;
- nn->num_tx_rings = new.n_rings;
+ nn->dp.txd_cnt = new.dcnt;
+ nn->dp.tx_rings = new.rings;
+ nn->dp.num_tx_rings = new.n_rings;
}
-static void
-nfp_net_tx_ring_set_free(struct nfp_net *nn, struct nfp_net_ring_set *s)
+static void nfp_net_tx_ring_set_free(struct nfp_net_ring_set *s)
{
struct nfp_net_tx_ring *rings = s->rings;
unsigned int r;
@@ -1841,12 +1847,12 @@ nfp_net_tx_ring_set_free(struct nfp_net *nn, struct nfp_net_ring_set *s)
static void nfp_net_rx_ring_free(struct nfp_net_rx_ring *rx_ring)
{
struct nfp_net_r_vector *r_vec = rx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
kfree(rx_ring->rxbufs);
if (rx_ring->rxds)
- dma_free_coherent(nn->dev, rx_ring->size,
+ dma_free_coherent(dp->dev, rx_ring->size,
rx_ring->rxds, rx_ring->dma);
rx_ring->cnt = 0;
@@ -1869,14 +1875,14 @@ nfp_net_rx_ring_alloc(struct nfp_net_rx_ring *rx_ring, unsigned int fl_bufsz,
u32 cnt)
{
struct nfp_net_r_vector *r_vec = rx_ring->r_vec;
- struct nfp_net *nn = r_vec->nfp_net;
+ struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
int sz;
rx_ring->cnt = cnt;
rx_ring->bufsz = fl_bufsz;
rx_ring->size = sizeof(*rx_ring->rxds) * rx_ring->cnt;
- rx_ring->rxds = dma_zalloc_coherent(nn->dev, rx_ring->size,
+ rx_ring->rxds = dma_zalloc_coherent(dp->dev, rx_ring->size,
&rx_ring->dma, GFP_KERNEL);
if (!rx_ring->rxds)
goto err_alloc;
@@ -1897,7 +1903,7 @@ static struct nfp_net_rx_ring *
nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
bool xdp)
{
- unsigned int fl_bufsz = nfp_net_calc_fl_bufsz(nn, s->mtu);
+ unsigned int fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, s->mtu);
struct nfp_net_rx_ring *rings;
unsigned int r;
@@ -1911,7 +1917,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
if (nfp_net_rx_ring_alloc(&rings[r], fl_bufsz, s->dcnt))
goto err_free_prev;
- if (nfp_net_rx_ring_bufs_alloc(nn, &rings[r], xdp))
+ if (nfp_net_rx_ring_bufs_alloc(&nn->dp, &rings[r], xdp))
goto err_free_ring;
}
@@ -1919,7 +1925,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
err_free_prev:
while (r--) {
- nfp_net_rx_ring_bufs_free(nn, &rings[r], xdp);
+ nfp_net_rx_ring_bufs_free(&nn->dp, &rings[r], xdp);
err_free_ring:
nfp_net_rx_ring_free(&rings[r]);
}
@@ -1932,27 +1938,27 @@ nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_ring_set *s)
{
struct nfp_net_ring_set new = *s;
- s->mtu = nn->netdev->mtu;
- s->dcnt = nn->rxd_cnt;
- s->rings = nn->rx_rings;
- s->n_rings = nn->num_rx_rings;
+ s->mtu = nn->dp.netdev->mtu;
+ s->dcnt = nn->dp.rxd_cnt;
+ s->rings = nn->dp.rx_rings;
+ s->n_rings = nn->dp.num_rx_rings;
- nn->netdev->mtu = new.mtu;
- nn->fl_bufsz = nfp_net_calc_fl_bufsz(nn, new.mtu);
- nn->rxd_cnt = new.dcnt;
- nn->rx_rings = new.rings;
- nn->num_rx_rings = new.n_rings;
+ nn->dp.netdev->mtu = new.mtu;
+ nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, new.mtu);
+ nn->dp.rxd_cnt = new.dcnt;
+ nn->dp.rx_rings = new.rings;
+ nn->dp.num_rx_rings = new.n_rings;
}
static void
-nfp_net_rx_ring_set_free(struct nfp_net *nn, struct nfp_net_ring_set *s,
+nfp_net_rx_ring_set_free(struct nfp_net_dp *dp, struct nfp_net_ring_set *s,
bool xdp)
{
struct nfp_net_rx_ring *rings = s->rings;
unsigned int r;
for (r = 0; r < s->n_rings; r++) {
- nfp_net_rx_ring_bufs_free(nn, &rings[r], xdp);
+ nfp_net_rx_ring_bufs_free(dp, &rings[r], xdp);
nfp_net_rx_ring_free(&rings[r]);
}
@@ -1960,15 +1966,15 @@ nfp_net_rx_ring_set_free(struct nfp_net *nn, struct nfp_net_ring_set *s,
}
static void
-nfp_net_vector_assign_rings(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
- int idx)
+nfp_net_vector_assign_rings(struct nfp_net_dp *dp,
+ struct nfp_net_r_vector *r_vec, int idx)
{
- r_vec->rx_ring = idx < nn->num_rx_rings ? &nn->rx_rings[idx] : NULL;
+ r_vec->rx_ring = idx < dp->num_rx_rings ? &dp->rx_rings[idx] : NULL;
r_vec->tx_ring =
- idx < nn->num_stack_tx_rings ? &nn->tx_rings[idx] : NULL;
+ idx < dp->num_stack_tx_rings ? &dp->tx_rings[idx] : NULL;
- r_vec->xdp_ring = idx < nn->num_tx_rings - nn->num_stack_tx_rings ?
- &nn->tx_rings[nn->num_stack_tx_rings + idx] : NULL;
+ r_vec->xdp_ring = idx < dp->num_tx_rings - dp->num_stack_tx_rings ?
+ &dp->tx_rings[dp->num_stack_tx_rings + idx] : NULL;
}
static int
@@ -1978,11 +1984,11 @@ nfp_net_prepare_vector(struct nfp_net *nn, struct nfp_net_r_vector *r_vec,
int err;
/* Setup NAPI */
- netif_napi_add(nn->netdev, &r_vec->napi,
+ netif_napi_add(nn->dp.netdev, &r_vec->napi,
nfp_net_poll, NAPI_POLL_WEIGHT);
snprintf(r_vec->name, sizeof(r_vec->name),
- "%s-rxtx-%d", nn->netdev->name, idx);
+ "%s-rxtx-%d", nn->dp.netdev->name, idx);
err = request_irq(r_vec->irq_vector, r_vec->handler, 0, r_vec->name,
r_vec);
if (err) {
@@ -2053,13 +2059,13 @@ void nfp_net_coalesce_write_cfg(struct nfp_net *nn)
/* copy RX interrupt coalesce parameters */
value = (nn->rx_coalesce_max_frames << 16) |
(factor * nn->rx_coalesce_usecs);
- for (i = 0; i < nn->num_rx_rings; i++)
+ for (i = 0; i < nn->dp.num_rx_rings; i++)
nn_writel(nn, NFP_NET_CFG_RXR_IRQ_MOD(i), value);
/* copy TX interrupt coalesce parameters */
value = (nn->tx_coalesce_max_frames << 16) |
(factor * nn->tx_coalesce_usecs);
- for (i = 0; i < nn->num_tx_rings; i++)
+ for (i = 0; i < nn->dp.num_tx_rings; i++)
nn_writel(nn, NFP_NET_CFG_TXR_IRQ_MOD(i), value);
}
@@ -2074,9 +2080,9 @@ void nfp_net_coalesce_write_cfg(struct nfp_net *nn)
static void nfp_net_write_mac_addr(struct nfp_net *nn)
{
nn_writel(nn, NFP_NET_CFG_MACADDR + 0,
- get_unaligned_be32(nn->netdev->dev_addr));
+ get_unaligned_be32(nn->dp.netdev->dev_addr));
nn_writew(nn, NFP_NET_CFG_MACADDR + 6,
- get_unaligned_be16(nn->netdev->dev_addr + 4));
+ get_unaligned_be16(nn->dp.netdev->dev_addr + 4));
}
static void nfp_net_vec_clear_ring_data(struct nfp_net *nn, unsigned int idx)
@@ -2100,7 +2106,7 @@ static void nfp_net_clear_config_and_disable(struct nfp_net *nn)
unsigned int r;
int err;
- new_ctrl = nn->ctrl;
+ new_ctrl = nn->dp.ctrl;
new_ctrl &= ~NFP_NET_CFG_CTRL_ENABLE;
update = NFP_NET_CFG_UPDATE_GEN;
update |= NFP_NET_CFG_UPDATE_MSIX;
@@ -2117,14 +2123,14 @@ static void nfp_net_clear_config_and_disable(struct nfp_net *nn)
if (err)
nn_err(nn, "Could not disable device: %d\n", err);
- for (r = 0; r < nn->num_rx_rings; r++)
- nfp_net_rx_ring_reset(&nn->rx_rings[r]);
- for (r = 0; r < nn->num_tx_rings; r++)
- nfp_net_tx_ring_reset(nn, &nn->tx_rings[r]);
- for (r = 0; r < nn->num_r_vecs; r++)
+ for (r = 0; r < nn->dp.num_rx_rings; r++)
+ nfp_net_rx_ring_reset(&nn->dp.rx_rings[r]);
+ for (r = 0; r < nn->dp.num_tx_rings; r++)
+ nfp_net_tx_ring_reset(&nn->dp, &nn->dp.tx_rings[r]);
+ for (r = 0; r < nn->dp.num_r_vecs; r++)
nfp_net_vec_clear_ring_data(nn, r);
- nn->ctrl = new_ctrl;
+ nn->dp.ctrl = new_ctrl;
}
static void
@@ -2152,7 +2158,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn)
unsigned int r;
int err;
- new_ctrl = nn->ctrl;
+ new_ctrl = nn->dp.ctrl;
if (nn->cap & NFP_NET_CFG_CTRL_RSS) {
nfp_net_rss_write_key(nn);
@@ -2168,22 +2174,22 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn)
update |= NFP_NET_CFG_UPDATE_IRQMOD;
}
- for (r = 0; r < nn->num_tx_rings; r++)
- nfp_net_tx_ring_hw_cfg_write(nn, &nn->tx_rings[r], r);
- for (r = 0; r < nn->num_rx_rings; r++)
- nfp_net_rx_ring_hw_cfg_write(nn, &nn->rx_rings[r], r);
+ for (r = 0; r < nn->dp.num_tx_rings; r++)
+ nfp_net_tx_ring_hw_cfg_write(nn, &nn->dp.tx_rings[r], r);
+ for (r = 0; r < nn->dp.num_rx_rings; r++)
+ nfp_net_rx_ring_hw_cfg_write(nn, &nn->dp.rx_rings[r], r);
- nn_writeq(nn, NFP_NET_CFG_TXRS_ENABLE, nn->num_tx_rings == 64 ?
- 0xffffffffffffffffULL : ((u64)1 << nn->num_tx_rings) - 1);
+ nn_writeq(nn, NFP_NET_CFG_TXRS_ENABLE, nn->dp.num_tx_rings == 64 ?
+ 0xffffffffffffffffULL : ((u64)1 << nn->dp.num_tx_rings) - 1);
- nn_writeq(nn, NFP_NET_CFG_RXRS_ENABLE, nn->num_rx_rings == 64 ?
- 0xffffffffffffffffULL : ((u64)1 << nn->num_rx_rings) - 1);
+ nn_writeq(nn, NFP_NET_CFG_RXRS_ENABLE, nn->dp.num_rx_rings == 64 ?
+ 0xffffffffffffffffULL : ((u64)1 << nn->dp.num_rx_rings) - 1);
nfp_net_write_mac_addr(nn);
- nn_writel(nn, NFP_NET_CFG_MTU, nn->netdev->mtu);
+ nn_writel(nn, NFP_NET_CFG_MTU, nn->dp.netdev->mtu);
nn_writel(nn, NFP_NET_CFG_FLBUFSZ,
- nn->fl_bufsz - NFP_NET_RX_BUF_NON_DATA);
+ nn->dp.fl_bufsz - NFP_NET_RX_BUF_NON_DATA);
/* Enable device */
new_ctrl |= NFP_NET_CFG_CTRL_ENABLE;
@@ -2196,18 +2202,18 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn)
nn_writel(nn, NFP_NET_CFG_CTRL, new_ctrl);
err = nfp_net_reconfig(nn, update);
- nn->ctrl = new_ctrl;
+ nn->dp.ctrl = new_ctrl;
- for (r = 0; r < nn->num_rx_rings; r++)
- nfp_net_rx_ring_fill_freelist(&nn->rx_rings[r]);
+ for (r = 0; r < nn->dp.num_rx_rings; r++)
+ nfp_net_rx_ring_fill_freelist(&nn->dp.rx_rings[r]);
/* Since reconfiguration requests while NFP is down are ignored we
* have to wipe the entire VXLAN configuration and reinitialize it.
*/
- if (nn->ctrl & NFP_NET_CFG_CTRL_VXLAN) {
+ if (nn->dp.ctrl & NFP_NET_CFG_CTRL_VXLAN) {
memset(&nn->vxlan_ports, 0, sizeof(nn->vxlan_ports));
memset(&nn->vxlan_usecnt, 0, sizeof(nn->vxlan_usecnt));
- udp_tunnel_get_rx_info(nn->netdev);
+ udp_tunnel_get_rx_info(nn->dp.netdev);
}
return err;
@@ -2236,12 +2242,12 @@ static void nfp_net_open_stack(struct nfp_net *nn)
{
unsigned int r;
- for (r = 0; r < nn->num_r_vecs; r++) {
+ for (r = 0; r < nn->dp.num_r_vecs; r++) {
napi_enable(&nn->r_vecs[r].napi);
enable_irq(nn->r_vecs[r].irq_vector);
}
- netif_tx_wake_all_queues(nn->netdev);
+ netif_tx_wake_all_queues(nn->dp.netdev);
enable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector);
nfp_net_read_link_status(nn);
@@ -2251,18 +2257,18 @@ static int nfp_net_netdev_open(struct net_device *netdev)
{
struct nfp_net *nn = netdev_priv(netdev);
struct nfp_net_ring_set rx = {
- .n_rings = nn->num_rx_rings,
- .mtu = nn->netdev->mtu,
- .dcnt = nn->rxd_cnt,
+ .n_rings = nn->dp.num_rx_rings,
+ .mtu = nn->dp.netdev->mtu,
+ .dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
- .n_rings = nn->num_tx_rings,
- .dcnt = nn->txd_cnt,
+ .n_rings = nn->dp.num_tx_rings,
+ .dcnt = nn->dp.txd_cnt,
};
int err, r;
- if (nn->ctrl & NFP_NET_CFG_CTRL_ENABLE) {
- nn_err(nn, "Dev is already enabled: 0x%08x\n", nn->ctrl);
+ if (nn->dp.ctrl & NFP_NET_CFG_CTRL_ENABLE) {
+ nn_err(nn, "Dev is already enabled: 0x%08x\n", nn->dp.ctrl);
return -EBUSY;
}
@@ -2283,33 +2289,33 @@ static int nfp_net_netdev_open(struct net_device *netdev)
goto err_free_exn;
disable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector);
- for (r = 0; r < nn->num_r_vecs; r++) {
+ for (r = 0; r < nn->dp.num_r_vecs; r++) {
err = nfp_net_prepare_vector(nn, &nn->r_vecs[r], r);
if (err)
goto err_cleanup_vec_p;
}
- nn->rx_rings = nfp_net_rx_ring_set_prepare(nn, &rx, nn->xdp_prog);
- if (!nn->rx_rings) {
+ nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &rx, nn->dp.xdp_prog);
+ if (!nn->dp.rx_rings) {
err = -ENOMEM;
goto err_cleanup_vec;
}
- nn->tx_rings = nfp_net_tx_ring_set_prepare(nn, &tx,
- nn->num_stack_tx_rings);
- if (!nn->tx_rings) {
+ nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &tx,
+ nn->dp.num_stack_tx_rings);
+ if (!nn->dp.tx_rings) {
err = -ENOMEM;
goto err_free_rx_rings;
}
for (r = 0; r < nn->max_r_vecs; r++)
- nfp_net_vector_assign_rings(nn, &nn->r_vecs[r], r);
+ nfp_net_vector_assign_rings(&nn->dp, &nn->r_vecs[r], r);
- err = netif_set_real_num_tx_queues(netdev, nn->num_stack_tx_rings);
+ err = netif_set_real_num_tx_queues(netdev, nn->dp.num_stack_tx_rings);
if (err)
goto err_free_rings;
- err = netif_set_real_num_rx_queues(netdev, nn->num_rx_rings);
+ err = netif_set_real_num_rx_queues(netdev, nn->dp.num_rx_rings);
if (err)
goto err_free_rings;
@@ -2335,11 +2341,11 @@ static int nfp_net_netdev_open(struct net_device *netdev)
return 0;
err_free_rings:
- nfp_net_tx_ring_set_free(nn, &tx);
+ nfp_net_tx_ring_set_free(&tx);
err_free_rx_rings:
- nfp_net_rx_ring_set_free(nn, &rx, nn->xdp_prog);
+ nfp_net_rx_ring_set_free(&nn->dp, &rx, nn->dp.xdp_prog);
err_cleanup_vec:
- r = nn->num_r_vecs;
+ r = nn->dp.num_r_vecs;
err_cleanup_vec_p:
while (r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
@@ -2358,15 +2364,15 @@ static void nfp_net_close_stack(struct nfp_net *nn)
unsigned int r;
disable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector);
- netif_carrier_off(nn->netdev);
+ netif_carrier_off(nn->dp.netdev);
nn->link_up = false;
- for (r = 0; r < nn->num_r_vecs; r++) {
+ for (r = 0; r < nn->dp.num_r_vecs; r++) {
disable_irq(nn->r_vecs[r].irq_vector);
napi_disable(&nn->r_vecs[r].napi);
}
- netif_tx_disable(nn->netdev);
+ netif_tx_disable(nn->dp.netdev);
}
/**
@@ -2377,17 +2383,18 @@ static void nfp_net_close_free_all(struct nfp_net *nn)
{
unsigned int r;
- for (r = 0; r < nn->num_rx_rings; r++) {
- nfp_net_rx_ring_bufs_free(nn, &nn->rx_rings[r], nn->xdp_prog);
- nfp_net_rx_ring_free(&nn->rx_rings[r]);
+ for (r = 0; r < nn->dp.num_rx_rings; r++) {
+ nfp_net_rx_ring_bufs_free(&nn->dp, &nn->dp.rx_rings[r],
+ nn->dp.xdp_prog);
+ nfp_net_rx_ring_free(&nn->dp.rx_rings[r]);
}
- for (r = 0; r < nn->num_tx_rings; r++)
- nfp_net_tx_ring_free(&nn->tx_rings[r]);
- for (r = 0; r < nn->num_r_vecs; r++)
+ for (r = 0; r < nn->dp.num_tx_rings; r++)
+ nfp_net_tx_ring_free(&nn->dp.tx_rings[r]);
+ for (r = 0; r < nn->dp.num_r_vecs; r++)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
- kfree(nn->rx_rings);
- kfree(nn->tx_rings);
+ kfree(nn->dp.rx_rings);
+ kfree(nn->dp.tx_rings);
nfp_net_aux_irq_free(nn, NFP_NET_CFG_LSC, NFP_NET_IRQ_LSC_IDX);
nfp_net_aux_irq_free(nn, NFP_NET_CFG_EXN, NFP_NET_IRQ_EXN_IDX);
@@ -2401,8 +2408,8 @@ static int nfp_net_netdev_close(struct net_device *netdev)
{
struct nfp_net *nn = netdev_priv(netdev);
- if (!(nn->ctrl & NFP_NET_CFG_CTRL_ENABLE)) {
- nn_err(nn, "Dev is not up: 0x%08x\n", nn->ctrl);
+ if (!(nn->dp.ctrl & NFP_NET_CFG_CTRL_ENABLE)) {
+ nn_err(nn, "Dev is not up: 0x%08x\n", nn->dp.ctrl);
return 0;
}
@@ -2427,7 +2434,7 @@ static void nfp_net_set_rx_mode(struct net_device *netdev)
struct nfp_net *nn = netdev_priv(netdev);
u32 new_ctrl;
- new_ctrl = nn->ctrl;
+ new_ctrl = nn->dp.ctrl;
if (netdev->flags & IFF_PROMISC) {
if (nn->cap & NFP_NET_CFG_CTRL_PROMISC)
@@ -2438,13 +2445,13 @@ static void nfp_net_set_rx_mode(struct net_device *netdev)
new_ctrl &= ~NFP_NET_CFG_CTRL_PROMISC;
}
- if (new_ctrl == nn->ctrl)
+ if (new_ctrl == nn->dp.ctrl)
return;
nn_writel(nn, NFP_NET_CFG_CTRL, new_ctrl);
nfp_net_reconfig_post(nn, NFP_NET_CFG_UPDATE_GEN);
- nn->ctrl = new_ctrl;
+ nn->dp.ctrl = new_ctrl;
}
static void nfp_net_rss_init_itbl(struct nfp_net *nn)
@@ -2453,7 +2460,7 @@ static void nfp_net_rss_init_itbl(struct nfp_net *nn)
for (i = 0; i < sizeof(nn->rss_itbl); i++)
nn->rss_itbl[i] =
- ethtool_rxfh_indir_default(i, nn->num_rx_rings);
+ ethtool_rxfh_indir_default(i, nn->dp.num_rx_rings);
}
static int
@@ -2471,24 +2478,23 @@ nfp_net_ring_swap_enable(struct nfp_net *nn, unsigned int *num_vecs,
if (tx)
nfp_net_tx_ring_set_swap(nn, tx);
- swap(*num_vecs, nn->num_r_vecs);
- swap(*stack_tx_rings, nn->num_stack_tx_rings);
- *xdp_prog = xchg(&nn->xdp_prog, *xdp_prog);
+ swap(*num_vecs, nn->dp.num_r_vecs);
+ swap(*stack_tx_rings, nn->dp.num_stack_tx_rings);
+ *xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
for (r = 0; r < nn->max_r_vecs; r++)
- nfp_net_vector_assign_rings(nn, &nn->r_vecs[r], r);
+ nfp_net_vector_assign_rings(&nn->dp, &nn->r_vecs[r], r);
- if (!netif_is_rxfh_configured(nn->netdev))
+ if (!netif_is_rxfh_configured(nn->dp.netdev))
nfp_net_rss_init_itbl(nn);
- err = netif_set_real_num_rx_queues(nn->netdev,
- nn->num_rx_rings);
+ err = netif_set_real_num_rx_queues(nn->dp.netdev, nn->dp.num_rx_rings);
if (err)
return err;
- if (nn->netdev->real_num_tx_queues != nn->num_stack_tx_rings) {
- err = netif_set_real_num_tx_queues(nn->netdev,
- nn->num_stack_tx_rings);
+ if (nn->dp.netdev->real_num_tx_queues != nn->dp.num_stack_tx_rings) {
+ err = netif_set_real_num_tx_queues(nn->dp.netdev,
+ nn->dp.num_stack_tx_rings);
if (err)
return err;
}
@@ -2503,7 +2509,7 @@ nfp_net_check_config(struct nfp_net *nn, struct bpf_prog *xdp_prog,
/* XDP-enabled tests */
if (!xdp_prog)
return 0;
- if (rx && nfp_net_calc_fl_bufsz(nn, rx->mtu) > PAGE_SIZE) {
+ if (rx && nfp_net_calc_fl_bufsz(&nn->dp, rx->mtu) > PAGE_SIZE) {
nn_warn(nn, "MTU too large w/ XDP enabled\n");
return -EINVAL;
}
@@ -2521,17 +2527,17 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *tx,
unsigned int stack_tx_rings, unsigned int num_vecs)
{
- nn->netdev->mtu = rx ? rx->mtu : nn->netdev->mtu;
- nn->fl_bufsz = nfp_net_calc_fl_bufsz(nn, nn->netdev->mtu);
- nn->rxd_cnt = rx ? rx->dcnt : nn->rxd_cnt;
- nn->txd_cnt = tx ? tx->dcnt : nn->txd_cnt;
- nn->num_rx_rings = rx ? rx->n_rings : nn->num_rx_rings;
- nn->num_tx_rings = tx ? tx->n_rings : nn->num_tx_rings;
- nn->num_stack_tx_rings = stack_tx_rings;
- nn->num_r_vecs = num_vecs;
- *xdp_prog = xchg(&nn->xdp_prog, *xdp_prog);
-
- if (!netif_is_rxfh_configured(nn->netdev))
+ nn->dp.netdev->mtu = rx ? rx->mtu : nn->dp.netdev->mtu;
+ nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, nn->dp.netdev->mtu);
+ nn->dp.rxd_cnt = rx ? rx->dcnt : nn->dp.rxd_cnt;
+ nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
+ nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
+ nn->dp.num_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
+ nn->dp.num_stack_tx_rings = stack_tx_rings;
+ nn->dp.num_r_vecs = num_vecs;
+ *xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
+
+ if (!netif_is_rxfh_configured(nn->dp.netdev))
nfp_net_rss_init_itbl(nn);
}
@@ -2542,24 +2548,24 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
unsigned int stack_tx_rings, num_vecs, r;
int err;
- stack_tx_rings = tx ? tx->n_rings : nn->num_tx_rings;
+ stack_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
if (*xdp_prog)
- stack_tx_rings -= rx ? rx->n_rings : nn->num_rx_rings;
+ stack_tx_rings -= rx ? rx->n_rings : nn->dp.num_rx_rings;
- num_vecs = max(rx ? rx->n_rings : nn->num_rx_rings, stack_tx_rings);
+ num_vecs = max(rx ? rx->n_rings : nn->dp.num_rx_rings, stack_tx_rings);
err = nfp_net_check_config(nn, *xdp_prog, rx, tx);
if (err)
return err;
- if (!netif_running(nn->netdev)) {
+ if (!netif_running(nn->dp.netdev)) {
nfp_net_ring_reconfig_down(nn, xdp_prog, rx, tx,
stack_tx_rings, num_vecs);
return 0;
}
/* Prepare new rings */
- for (r = nn->num_r_vecs; r < num_vecs; r++) {
+ for (r = nn->dp.num_r_vecs; r < num_vecs; r++) {
err = nfp_net_prepare_vector(nn, &nn->r_vecs[r], r);
if (err) {
num_vecs = r;
@@ -2597,13 +2603,13 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
nn_err(nn, "Can't restore ring config - FW communication failed (%d,%d)\n",
err, err2);
}
- for (r = num_vecs - 1; r >= nn->num_r_vecs; r--)
+ for (r = num_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
if (rx)
- nfp_net_rx_ring_set_free(nn, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(&nn->dp, rx, *xdp_prog);
if (tx)
- nfp_net_tx_ring_set_free(nn, tx);
+ nfp_net_tx_ring_set_free(tx);
nfp_net_open_stack(nn);
@@ -2611,9 +2617,9 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
err_free_rx:
if (rx)
- nfp_net_rx_ring_set_free(nn, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(&nn->dp, rx, *xdp_prog);
err_cleanup_vecs:
- for (r = num_vecs - 1; r >= nn->num_r_vecs; r--)
+ for (r = num_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
return err;
}
@@ -2622,12 +2628,12 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
{
struct nfp_net *nn = netdev_priv(netdev);
struct nfp_net_ring_set rx = {
- .n_rings = nn->num_rx_rings,
+ .n_rings = nn->dp.num_rx_rings,
.mtu = new_mtu,
- .dcnt = nn->rxd_cnt,
+ .dcnt = nn->dp.rxd_cnt,
};
- return nfp_net_ring_reconfig(nn, &nn->xdp_prog, &rx, NULL);
+ return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog, &rx, NULL);
}
static void nfp_net_stat64(struct net_device *netdev,
@@ -2636,7 +2642,7 @@ static void nfp_net_stat64(struct net_device *netdev,
struct nfp_net *nn = netdev_priv(netdev);
int r;
- for (r = 0; r < nn->num_r_vecs; r++) {
+ for (r = 0; r < nn->dp.num_r_vecs; r++) {
struct nfp_net_r_vector *r_vec = &nn->r_vecs[r];
u64 data[3];
unsigned int start;
@@ -2683,7 +2689,7 @@ nfp_net_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
return -ENOTSUPP;
if (tc->type == TC_SETUP_CLSBPF && nfp_net_ebpf_capable(nn)) {
- if (!nn->bpf_offload_xdp)
+ if (!nn->dp.bpf_offload_xdp)
return nfp_net_bpf_offload(nn, tc->cls_bpf);
else
return -EBUSY;
@@ -2702,7 +2708,7 @@ static int nfp_net_set_features(struct net_device *netdev,
/* Assume this is not called with features we have not advertised */
- new_ctrl = nn->ctrl;
+ new_ctrl = nn->dp.ctrl;
if (changed & NETIF_F_RXCSUM) {
if (features & NETIF_F_RXCSUM)
@@ -2746,7 +2752,7 @@ static int nfp_net_set_features(struct net_device *netdev,
new_ctrl &= ~NFP_NET_CFG_CTRL_GATHER;
}
- if (changed & NETIF_F_HW_TC && nn->ctrl & NFP_NET_CFG_CTRL_BPF) {
+ if (changed & NETIF_F_HW_TC && nn->dp.ctrl & NFP_NET_CFG_CTRL_BPF) {
nn_err(nn, "Cannot disable HW TC offload while in use\n");
return -EBUSY;
}
@@ -2754,16 +2760,16 @@ static int nfp_net_set_features(struct net_device *netdev,
nn_dbg(nn, "Feature change 0x%llx -> 0x%llx (changed=0x%llx)\n",
netdev->features, features, changed);
- if (new_ctrl == nn->ctrl)
+ if (new_ctrl == nn->dp.ctrl)
return 0;
- nn_dbg(nn, "NIC ctrl: 0x%x -> 0x%x\n", nn->ctrl, new_ctrl);
+ nn_dbg(nn, "NIC ctrl: 0x%x -> 0x%x\n", nn->dp.ctrl, new_ctrl);
nn_writel(nn, NFP_NET_CFG_CTRL, new_ctrl);
err = nfp_net_reconfig(nn, NFP_NET_CFG_UPDATE_GEN);
if (err)
return err;
- nn->ctrl = new_ctrl;
+ nn->dp.ctrl = new_ctrl;
return 0;
}
@@ -2846,7 +2852,7 @@ static void nfp_net_set_vxlan_port(struct nfp_net *nn, int idx, __be16 port)
nn->vxlan_ports[idx] = port;
- if (!(nn->ctrl & NFP_NET_CFG_CTRL_VXLAN))
+ if (!(nn->dp.ctrl & NFP_NET_CFG_CTRL_VXLAN))
return;
BUILD_BUG_ON(NFP_NET_N_VXLAN_PORTS & 1);
@@ -2925,8 +2931,8 @@ static int nfp_net_xdp_offload(struct nfp_net *nn, struct bpf_prog *prog)
if (!nfp_net_ebpf_capable(nn))
return -EINVAL;
- if (nn->ctrl & NFP_NET_CFG_CTRL_BPF) {
- if (!nn->bpf_offload_xdp)
+ if (nn->dp.ctrl & NFP_NET_CFG_CTRL_BPF) {
+ if (!nn->dp.bpf_offload_xdp)
return prog ? -EBUSY : 0;
cmd.command = prog ? TC_CLSBPF_REPLACE : TC_CLSBPF_DESTROY;
} else {
@@ -2939,20 +2945,20 @@ static int nfp_net_xdp_offload(struct nfp_net *nn, struct bpf_prog *prog)
/* Stop offload if replace not possible */
if (ret && cmd.command == TC_CLSBPF_REPLACE)
nfp_net_xdp_offload(nn, NULL);
- nn->bpf_offload_xdp = prog && !ret;
+ nn->dp.bpf_offload_xdp = prog && !ret;
return ret;
}
static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
{
struct nfp_net_ring_set rx = {
- .n_rings = nn->num_rx_rings,
- .mtu = nn->netdev->mtu,
- .dcnt = nn->rxd_cnt,
+ .n_rings = nn->dp.num_rx_rings,
+ .mtu = nn->dp.netdev->mtu,
+ .dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
- .n_rings = nn->num_tx_rings,
- .dcnt = nn->txd_cnt,
+ .n_rings = nn->dp.num_tx_rings,
+ .dcnt = nn->dp.txd_cnt,
};
int err;
@@ -2960,16 +2966,16 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
nn_err(nn, "Does not support bpf_xdp_adjust_head()\n");
return -EOPNOTSUPP;
}
- if (!prog && !nn->xdp_prog)
+ if (!prog && !nn->dp.xdp_prog)
return 0;
- if (prog && nn->xdp_prog) {
- prog = xchg(&nn->xdp_prog, prog);
+ if (prog && nn->dp.xdp_prog) {
+ prog = xchg(&nn->dp.xdp_prog, prog);
bpf_prog_put(prog);
- nfp_net_xdp_offload(nn, nn->xdp_prog);
+ nfp_net_xdp_offload(nn, nn->dp.xdp_prog);
return 0;
}
- tx.n_rings += prog ? nn->num_rx_rings : -nn->num_rx_rings;
+ tx.n_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
err = nfp_net_ring_reconfig(nn, &prog, &rx, &tx);
@@ -2980,7 +2986,7 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
if (prog)
bpf_prog_put(prog);
- nfp_net_xdp_offload(nn, nn->xdp_prog);
+ nfp_net_xdp_offload(nn, nn->dp.xdp_prog);
return 0;
}
@@ -2993,7 +2999,7 @@ static int nfp_net_xdp(struct net_device *netdev, struct netdev_xdp *xdp)
case XDP_SETUP_PROG:
return nfp_net_xdp_setup(nn, xdp->prog);
case XDP_QUERY_PROG:
- xdp->prog_attached = !!nn->xdp_prog;
+ xdp->prog_attached = !!nn->dp.xdp_prog;
return 0;
default:
return -EINVAL;
@@ -3025,9 +3031,9 @@ static const struct net_device_ops nfp_net_netdev_ops = {
void nfp_net_info(struct nfp_net *nn)
{
nn_info(nn, "Netronome NFP-6xxx %sNetdev: TxQs=%d/%d RxQs=%d/%d\n",
- nn->is_vf ? "VF " : "",
- nn->num_tx_rings, nn->max_tx_rings,
- nn->num_rx_rings, nn->max_rx_rings);
+ nn->dp.is_vf ? "VF " : "",
+ nn->dp.num_tx_rings, nn->max_tx_rings,
+ nn->dp.num_rx_rings, nn->max_rx_rings);
nn_info(nn, "VER: %d.%d.%d.%d, Maximum supported MTU: %d\n",
nn->fw_ver.resv, nn->fw_ver.class,
nn->fw_ver.major, nn->fw_ver.minor,
@@ -3079,22 +3085,24 @@ struct nfp_net *nfp_net_netdev_alloc(struct pci_dev *pdev,
SET_NETDEV_DEV(netdev, &pdev->dev);
nn = netdev_priv(netdev);
- nn->netdev = netdev;
- nn->dev = &pdev->dev;
+ nn->dp.netdev = netdev;
+ nn->dp.dev = &pdev->dev;
nn->pdev = pdev;
nn->max_tx_rings = max_tx_rings;
nn->max_rx_rings = max_rx_rings;
- nn->num_tx_rings = min_t(unsigned int, max_tx_rings, num_online_cpus());
- nn->num_rx_rings = min_t(unsigned int, max_rx_rings,
+ nn->dp.num_tx_rings = min_t(unsigned int,
+ max_tx_rings, num_online_cpus());
+ nn->dp.num_rx_rings = min_t(unsigned int, max_rx_rings,
netif_get_num_default_rss_queues());
- nn->num_r_vecs = max(nn->num_tx_rings, nn->num_rx_rings);
- nn->num_r_vecs = min_t(unsigned int, nn->num_r_vecs, num_online_cpus());
+ nn->dp.num_r_vecs = max(nn->dp.num_tx_rings, nn->dp.num_rx_rings);
+ nn->dp.num_r_vecs = min_t(unsigned int,
+ nn->dp.num_r_vecs, num_online_cpus());
- nn->txd_cnt = NFP_NET_TX_DESCS_DEFAULT;
- nn->rxd_cnt = NFP_NET_RX_DESCS_DEFAULT;
+ nn->dp.txd_cnt = NFP_NET_TX_DESCS_DEFAULT;
+ nn->dp.rxd_cnt = NFP_NET_RX_DESCS_DEFAULT;
spin_lock_init(&nn->reconfig_lock);
spin_lock_init(&nn->rx_filter_lock);
@@ -3114,7 +3122,7 @@ struct nfp_net *nfp_net_netdev_alloc(struct pci_dev *pdev,
*/
void nfp_net_netdev_free(struct nfp_net *nn)
{
- free_netdev(nn->netdev);
+ free_netdev(nn->dp.netdev);
}
/**
@@ -3156,7 +3164,7 @@ static void nfp_net_rss_init(struct nfp_net *nn)
func_bit = find_first_bit(&rss_cap_hfunc, NFP_NET_CFG_RSS_HFUNCS);
if (func_bit == NFP_NET_CFG_RSS_HFUNCS) {
- dev_warn(nn->dev,
+ dev_warn(nn->dp.dev,
"Bad RSS config, defaulting to Toeplitz hash\n");
func_bit = ETH_RSS_HASH_TOP_BIT;
}
@@ -3196,7 +3204,7 @@ int nfp_net_netdev_init(struct net_device *netdev)
struct nfp_net *nn = netdev_priv(netdev);
int err;
- nn->chained_metadata_format = nn->fw_ver.major > 3;
+ nn->dp.chained_metadata_format = nn->fw_ver.major > 3;
/* Get some of the read-only fields from the BAR */
nn->cap = nn_readl(nn, NFP_NET_CFG_CAP);
@@ -3206,16 +3214,16 @@ int nfp_net_netdev_init(struct net_device *netdev)
/* Determine RX packet/metadata boundary offset */
if (nn->fw_ver.major >= 2)
- nn->rx_offset = nn_readl(nn, NFP_NET_CFG_RX_OFFSET);
+ nn->dp.rx_offset = nn_readl(nn, NFP_NET_CFG_RX_OFFSET);
else
- nn->rx_offset = NFP_NET_RX_OFFSET;
+ nn->dp.rx_offset = NFP_NET_RX_OFFSET;
/* Set default MTU and Freelist buffer size */
if (nn->max_mtu < NFP_NET_DEFAULT_MTU)
netdev->mtu = nn->max_mtu;
else
netdev->mtu = NFP_NET_DEFAULT_MTU;
- nn->fl_bufsz = nfp_net_calc_fl_bufsz(nn, netdev->mtu);
+ nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, netdev->mtu);
/* Advertise/enable offloads based on capabilities
*
@@ -3226,31 +3234,31 @@ int nfp_net_netdev_init(struct net_device *netdev)
netdev->hw_features = NETIF_F_HIGHDMA;
if (nn->cap & NFP_NET_CFG_CTRL_RXCSUM) {
netdev->hw_features |= NETIF_F_RXCSUM;
- nn->ctrl |= NFP_NET_CFG_CTRL_RXCSUM;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_RXCSUM;
}
if (nn->cap & NFP_NET_CFG_CTRL_TXCSUM) {
netdev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
- nn->ctrl |= NFP_NET_CFG_CTRL_TXCSUM;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_TXCSUM;
}
if (nn->cap & NFP_NET_CFG_CTRL_GATHER) {
netdev->hw_features |= NETIF_F_SG;
- nn->ctrl |= NFP_NET_CFG_CTRL_GATHER;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_GATHER;
}
if ((nn->cap & NFP_NET_CFG_CTRL_LSO) && nn->fw_ver.major > 2) {
netdev->hw_features |= NETIF_F_TSO | NETIF_F_TSO6;
- nn->ctrl |= NFP_NET_CFG_CTRL_LSO;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_LSO;
}
if (nn->cap & NFP_NET_CFG_CTRL_RSS) {
netdev->hw_features |= NETIF_F_RXHASH;
nfp_net_rss_init(nn);
- nn->ctrl |= NFP_NET_CFG_CTRL_RSS;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_RSS;
}
if (nn->cap & NFP_NET_CFG_CTRL_VXLAN &&
nn->cap & NFP_NET_CFG_CTRL_NVGRE) {
if (nn->cap & NFP_NET_CFG_CTRL_LSO)
netdev->hw_features |= NETIF_F_GSO_GRE |
NETIF_F_GSO_UDP_TUNNEL;
- nn->ctrl |= NFP_NET_CFG_CTRL_VXLAN | NFP_NET_CFG_CTRL_NVGRE;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_VXLAN | NFP_NET_CFG_CTRL_NVGRE;
netdev->hw_enc_features = netdev->hw_features;
}
@@ -3259,11 +3267,11 @@ int nfp_net_netdev_init(struct net_device *netdev)
if (nn->cap & NFP_NET_CFG_CTRL_RXVLAN) {
netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX;
- nn->ctrl |= NFP_NET_CFG_CTRL_RXVLAN;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_RXVLAN;
}
if (nn->cap & NFP_NET_CFG_CTRL_TXVLAN) {
netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX;
- nn->ctrl |= NFP_NET_CFG_CTRL_TXVLAN;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_TXVLAN;
}
netdev->features = netdev->hw_features;
@@ -3276,14 +3284,14 @@ int nfp_net_netdev_init(struct net_device *netdev)
/* Allow L2 Broadcast and Multicast through by default, if supported */
if (nn->cap & NFP_NET_CFG_CTRL_L2BC)
- nn->ctrl |= NFP_NET_CFG_CTRL_L2BC;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_L2BC;
if (nn->cap & NFP_NET_CFG_CTRL_L2MC)
- nn->ctrl |= NFP_NET_CFG_CTRL_L2MC;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_L2MC;
/* Allow IRQ moderation, if supported */
if (nn->cap & NFP_NET_CFG_CTRL_IRQMOD) {
nfp_net_irqmod_init(nn);
- nn->ctrl |= NFP_NET_CFG_CTRL_IRQMOD;
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_IRQMOD;
}
/* Stash the re-configuration queue away. First odd queue in TX Bar */
@@ -3322,9 +3330,9 @@ void nfp_net_netdev_clean(struct net_device *netdev)
{
struct nfp_net *nn = netdev_priv(netdev);
- if (nn->xdp_prog)
- bpf_prog_put(nn->xdp_prog);
- if (nn->bpf_offload_xdp)
+ if (nn->dp.xdp_prog)
+ bpf_prog_put(nn->dp.xdp_prog);
+ if (nn->dp.bpf_offload_xdp)
nfp_net_xdp_offload(nn, NULL);
- unregister_netdev(nn->netdev);
+ unregister_netdev(nn->dp.netdev);
}
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_debugfs.c b/drivers/net/ethernet/netronome/nfp/nfp_net_debugfs.c
index edfa59e51fdd..74125584260b 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_debugfs.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_debugfs.c
@@ -54,7 +54,7 @@ static int nfp_net_debugfs_rx_q_read(struct seq_file *file, void *data)
goto out;
nn = r_vec->nfp_net;
rx_ring = r_vec->rx_ring;
- if (!netif_running(nn->netdev))
+ if (!netif_running(nn->dp.netdev))
goto out;
rxd_cnt = rx_ring->cnt;
@@ -145,7 +145,7 @@ static int nfp_net_debugfs_tx_q_read(struct seq_file *file, void *data)
if (!r_vec->nfp_net || !tx_ring)
goto out;
nn = r_vec->nfp_net;
- if (!netif_running(nn->netdev))
+ if (!netif_running(nn->dp.netdev))
goto out;
txd_cnt = tx_ring->cnt;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index a1bca2dca0a5..4620c1bba96e 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -127,9 +127,9 @@ static const struct _nfp_net_et_stats nfp_net_et_stats[] = {
};
#define NN_ET_GLOBAL_STATS_LEN ARRAY_SIZE(nfp_net_et_stats)
-#define NN_ET_RVEC_STATS_LEN (nn->num_r_vecs * 3)
+#define NN_ET_RVEC_STATS_LEN (nn->dp.num_r_vecs * 3)
#define NN_ET_RVEC_GATHER_STATS 7
-#define NN_ET_QUEUE_STATS_LEN ((nn->num_tx_rings + nn->num_rx_rings) * 2)
+#define NN_ET_QUEUE_STATS_LEN ((nn->dp.num_tx_rings + nn->dp.num_rx_rings) * 2)
#define NN_ET_STATS_LEN (NN_ET_GLOBAL_STATS_LEN + NN_ET_RVEC_GATHER_STATS + \
NN_ET_RVEC_STATS_LEN + NN_ET_QUEUE_STATS_LEN)
@@ -180,29 +180,29 @@ static void nfp_net_get_ringparam(struct net_device *netdev,
ring->rx_max_pending = NFP_NET_MAX_RX_DESCS;
ring->tx_max_pending = NFP_NET_MAX_TX_DESCS;
- ring->rx_pending = nn->rxd_cnt;
- ring->tx_pending = nn->txd_cnt;
+ ring->rx_pending = nn->dp.rxd_cnt;
+ ring->tx_pending = nn->dp.txd_cnt;
}
static int nfp_net_set_ring_size(struct nfp_net *nn, u32 rxd_cnt, u32 txd_cnt)
{
struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
struct nfp_net_ring_set rx = {
- .n_rings = nn->num_rx_rings,
- .mtu = nn->netdev->mtu,
+ .n_rings = nn->dp.num_rx_rings,
+ .mtu = nn->dp.netdev->mtu,
.dcnt = rxd_cnt,
};
struct nfp_net_ring_set tx = {
- .n_rings = nn->num_tx_rings,
+ .n_rings = nn->dp.num_tx_rings,
.dcnt = txd_cnt,
};
- if (nn->rxd_cnt != rxd_cnt)
+ if (nn->dp.rxd_cnt != rxd_cnt)
reconfig_rx = ℞
- if (nn->txd_cnt != txd_cnt)
+ if (nn->dp.txd_cnt != txd_cnt)
reconfig_tx = &tx;
- return nfp_net_ring_reconfig(nn, &nn->xdp_prog,
+ return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog,
reconfig_rx, reconfig_tx);
}
@@ -224,11 +224,11 @@ static int nfp_net_set_ringparam(struct net_device *netdev,
txd_cnt < NFP_NET_MIN_TX_DESCS || txd_cnt > NFP_NET_MAX_TX_DESCS)
return -EINVAL;
- if (nn->rxd_cnt == rxd_cnt && nn->txd_cnt == txd_cnt)
+ if (nn->dp.rxd_cnt == rxd_cnt && nn->dp.txd_cnt == txd_cnt)
return 0;
nn_dbg(nn, "Change ring size: RxQ %u->%u, TxQ %u->%u\n",
- nn->rxd_cnt, rxd_cnt, nn->txd_cnt, txd_cnt);
+ nn->dp.rxd_cnt, rxd_cnt, nn->dp.txd_cnt, txd_cnt);
return nfp_net_set_ring_size(nn, rxd_cnt, txd_cnt);
}
@@ -246,7 +246,7 @@ static void nfp_net_get_strings(struct net_device *netdev,
memcpy(p, nfp_net_et_stats[i].name, ETH_GSTRING_LEN);
p += ETH_GSTRING_LEN;
}
- for (i = 0; i < nn->num_r_vecs; i++) {
+ for (i = 0; i < nn->dp.num_r_vecs; i++) {
sprintf(p, "rvec_%u_rx_pkts", i);
p += ETH_GSTRING_LEN;
sprintf(p, "rvec_%u_tx_pkts", i);
@@ -268,13 +268,13 @@ static void nfp_net_get_strings(struct net_device *netdev,
p += ETH_GSTRING_LEN;
strncpy(p, "tx_lso", ETH_GSTRING_LEN);
p += ETH_GSTRING_LEN;
- for (i = 0; i < nn->num_tx_rings; i++) {
+ for (i = 0; i < nn->dp.num_tx_rings; i++) {
sprintf(p, "txq_%u_pkts", i);
p += ETH_GSTRING_LEN;
sprintf(p, "txq_%u_bytes", i);
p += ETH_GSTRING_LEN;
}
- for (i = 0; i < nn->num_rx_rings; i++) {
+ for (i = 0; i < nn->dp.num_rx_rings; i++) {
sprintf(p, "rxq_%u_pkts", i);
p += ETH_GSTRING_LEN;
sprintf(p, "rxq_%u_bytes", i);
@@ -312,7 +312,7 @@ static void nfp_net_get_stats(struct net_device *netdev,
break;
}
}
- for (j = 0; j < nn->num_r_vecs; j++) {
+ for (j = 0; j < nn->dp.num_r_vecs; j++) {
unsigned int start;
do {
@@ -338,13 +338,13 @@ static void nfp_net_get_stats(struct net_device *netdev,
}
for (j = 0; j < NN_ET_RVEC_GATHER_STATS; j++)
data[i++] = gathered_stats[j];
- for (j = 0; j < nn->num_tx_rings; j++) {
+ for (j = 0; j < nn->dp.num_tx_rings; j++) {
io_p = nn->ctrl_bar + NFP_NET_CFG_TXR_STATS(j);
data[i++] = readq(io_p);
io_p = nn->ctrl_bar + NFP_NET_CFG_TXR_STATS(j) + 8;
data[i++] = readq(io_p);
}
- for (j = 0; j < nn->num_rx_rings; j++) {
+ for (j = 0; j < nn->dp.num_rx_rings; j++) {
io_p = nn->ctrl_bar + NFP_NET_CFG_RXR_STATS(j);
data[i++] = readq(io_p);
io_p = nn->ctrl_bar + NFP_NET_CFG_RXR_STATS(j) + 8;
@@ -411,7 +411,7 @@ static int nfp_net_get_rxnfc(struct net_device *netdev,
switch (cmd->cmd) {
case ETHTOOL_GRXRINGS:
- cmd->data = nn->num_rx_rings;
+ cmd->data = nn->dp.num_rx_rings;
return 0;
case ETHTOOL_GRXFH:
return nfp_net_get_rss_hash_opts(nn, cmd);
@@ -745,16 +745,16 @@ static void nfp_net_get_channels(struct net_device *netdev,
struct nfp_net *nn = netdev_priv(netdev);
unsigned int num_tx_rings;
- num_tx_rings = nn->num_tx_rings;
- if (nn->xdp_prog)
- num_tx_rings -= nn->num_rx_rings;
+ num_tx_rings = nn->dp.num_tx_rings;
+ if (nn->dp.xdp_prog)
+ num_tx_rings -= nn->dp.num_rx_rings;
channel->max_rx = min(nn->max_rx_rings, nn->max_r_vecs);
channel->max_tx = min(nn->max_tx_rings, nn->max_r_vecs);
channel->max_combined = min(channel->max_rx, channel->max_tx);
channel->max_other = NFP_NET_NON_Q_VECTORS;
- channel->combined_count = min(nn->num_rx_rings, num_tx_rings);
- channel->rx_count = nn->num_rx_rings - channel->combined_count;
+ channel->combined_count = min(nn->dp.num_rx_rings, num_tx_rings);
+ channel->rx_count = nn->dp.num_rx_rings - channel->combined_count;
channel->tx_count = num_tx_rings - channel->combined_count;
channel->other_count = NFP_NET_NON_Q_VECTORS;
}
@@ -765,25 +765,25 @@ static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
struct nfp_net_ring_set rx = {
.n_rings = total_rx,
- .mtu = nn->netdev->mtu,
- .dcnt = nn->rxd_cnt,
+ .mtu = nn->dp.netdev->mtu,
+ .dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
.n_rings = total_tx,
- .dcnt = nn->txd_cnt,
+ .dcnt = nn->dp.txd_cnt,
};
- if (nn->num_rx_rings != total_rx)
+ if (nn->dp.num_rx_rings != total_rx)
reconfig_rx = ℞
- if (nn->num_stack_tx_rings != total_tx ||
- (nn->xdp_prog && reconfig_rx))
+ if (nn->dp.num_stack_tx_rings != total_tx ||
+ (nn->dp.xdp_prog && reconfig_rx))
reconfig_tx = &tx;
/* nfp_net_check_config() will catch tx.n_rings > nn->max_tx_rings */
- if (nn->xdp_prog)
+ if (nn->dp.xdp_prog)
tx.n_rings += total_rx;
- return nfp_net_ring_reconfig(nn, &nn->xdp_prog,
+ return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog,
reconfig_rx, reconfig_tx);
}
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
index 8a9b3f3b95a8..3935d19a273d 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
@@ -130,7 +130,7 @@ static u8 __iomem *nfp_net_map_area(struct nfp_cpp *cpp,
}
static void
-nfp_net_get_mac_addr_hwinfo(struct nfp_net *nn, struct nfp_cpp *cpp,
+nfp_net_get_mac_addr_hwinfo(struct nfp_net_dp *dp, struct nfp_cpp *cpp,
unsigned int id)
{
u8 mac_addr[ETH_ALEN];
@@ -141,22 +141,22 @@ nfp_net_get_mac_addr_hwinfo(struct nfp_net *nn, struct nfp_cpp *cpp,
mac_str = nfp_hwinfo_lookup(cpp, name);
if (!mac_str) {
- dev_warn(nn->dev, "Can't lookup MAC address. Generate\n");
- eth_hw_addr_random(nn->netdev);
+ dev_warn(dp->dev, "Can't lookup MAC address. Generate\n");
+ eth_hw_addr_random(dp->netdev);
return;
}
if (sscanf(mac_str, "%02hhx:%02hhx:%02hhx:%02hhx:%02hhx:%02hhx",
&mac_addr[0], &mac_addr[1], &mac_addr[2],
&mac_addr[3], &mac_addr[4], &mac_addr[5]) != 6) {
- dev_warn(nn->dev,
+ dev_warn(dp->dev,
"Can't parse MAC address (%s). Generate.\n", mac_str);
- eth_hw_addr_random(nn->netdev);
+ eth_hw_addr_random(dp->netdev);
return;
}
- ether_addr_copy(nn->netdev->dev_addr, mac_addr);
- ether_addr_copy(nn->netdev->perm_addr, mac_addr);
+ ether_addr_copy(dp->netdev->dev_addr, mac_addr);
+ ether_addr_copy(dp->netdev->perm_addr, mac_addr);
}
/**
@@ -179,12 +179,12 @@ nfp_net_get_mac_addr(struct nfp_net *nn, struct nfp_pf *pf, unsigned int id)
nn->eth_port = &pf->eth_tbl->ports[i];
- ether_addr_copy(nn->netdev->dev_addr, mac_addr);
- ether_addr_copy(nn->netdev->perm_addr, mac_addr);
+ ether_addr_copy(nn->dp.netdev->dev_addr, mac_addr);
+ ether_addr_copy(nn->dp.netdev->perm_addr, mac_addr);
return;
}
- nfp_net_get_mac_addr_hwinfo(nn, pf->cpp, id);
+ nfp_net_get_mac_addr_hwinfo(&nn->dp, pf->cpp, id);
}
static unsigned int nfp_net_pf_get_num_ports(struct nfp_pf *pf)
@@ -309,7 +309,7 @@ nfp_net_pf_alloc_port_netdev(struct nfp_pf *pf, void __iomem *ctrl_bar,
nn->ctrl_bar = ctrl_bar;
nn->tx_bar = tx_bar;
nn->rx_bar = rx_bar;
- nn->is_vf = 0;
+ nn->dp.is_vf = 0;
nn->stride_rx = stride;
nn->stride_tx = stride;
@@ -331,7 +331,7 @@ nfp_net_pf_init_port_netdev(struct nfp_pf *pf, struct nfp_net *nn,
*/
nn->me_freq_mhz = 1200;
- err = nfp_net_netdev_init(nn->netdev);
+ err = nfp_net_netdev_init(nn->dp.netdev);
if (err)
return err;
@@ -400,7 +400,7 @@ nfp_net_pf_spawn_netdevs(struct nfp_pf *pf,
/* Get MSI-X vectors */
wanted_irqs = 0;
list_for_each_entry(nn, &pf->ports, port_list)
- wanted_irqs += NFP_NET_NON_Q_VECTORS + nn->num_r_vecs;
+ wanted_irqs += NFP_NET_NON_Q_VECTORS + nn->dp.num_r_vecs;
pf->irq_entries = kcalloc(wanted_irqs, sizeof(*pf->irq_entries),
GFP_KERNEL);
if (!pf->irq_entries) {
@@ -445,7 +445,7 @@ nfp_net_pf_spawn_netdevs(struct nfp_pf *pf,
err_prev_deinit:
list_for_each_entry_continue_reverse(nn, &pf->ports, port_list) {
nfp_net_debugfs_dir_clean(&nn->debugfs_dir);
- nfp_net_netdev_clean(nn->netdev);
+ nfp_net_netdev_clean(nn->dp.netdev);
}
nfp_net_irqs_disable(pf->pdev);
err_vec_free:
@@ -571,7 +571,7 @@ void nfp_net_pci_remove(struct nfp_pf *pf)
list_for_each_entry(nn, &pf->ports, port_list) {
nfp_net_debugfs_dir_clean(&nn->debugfs_dir);
- nfp_net_netdev_clean(nn->netdev);
+ nfp_net_netdev_clean(nn->dp.netdev);
}
nfp_net_pf_free_netdevs(pf);
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_offload.c b/drivers/net/ethernet/netronome/nfp/nfp_net_offload.c
index f6ed1aa9d94b..b5b6f69d1e0f 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_offload.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_offload.c
@@ -58,7 +58,7 @@ void nfp_net_filter_stats_timer(unsigned long data)
spin_lock_bh(&nn->rx_filter_lock);
- if (nn->ctrl & NFP_NET_CFG_CTRL_BPF)
+ if (nn->dp.ctrl & NFP_NET_CFG_CTRL_BPF)
mod_timer(&nn->rx_filter_stats_timer,
jiffies + NFP_NET_STAT_POLL_IVL);
@@ -132,7 +132,7 @@ nfp_net_bpf_get_act(struct nfp_net *nn, struct tc_cls_bpf_offload *cls_bpf)
return NN_ACT_TC_DROP;
if (is_tcf_mirred_egress_redirect(a) &&
- tcf_mirred_ifindex(a) == nn->netdev->ifindex)
+ tcf_mirred_ifindex(a) == nn->dp.netdev->ifindex)
return NN_ACT_TC_REDIR;
}
@@ -160,7 +160,7 @@ nfp_net_bpf_offload_prepare(struct nfp_net *nn,
act = ret;
max_mtu = nn_readb(nn, NFP_NET_CFG_BPF_INL_MTU) * 64 - 32;
- if (max_mtu < nn->netdev->mtu) {
+ if (max_mtu < nn->dp.netdev->mtu) {
nn_info(nn, "BPF offload not supported with MTU larger than HW packet split boundary\n");
return -ENOTSUPP;
}
@@ -168,7 +168,7 @@ nfp_net_bpf_offload_prepare(struct nfp_net *nn,
start_off = nn_readw(nn, NFP_NET_CFG_BPF_START);
done_off = nn_readw(nn, NFP_NET_CFG_BPF_DONE);
- *code = dma_zalloc_coherent(nn->dev, code_sz, dma_addr, GFP_KERNEL);
+ *code = dma_zalloc_coherent(nn->dp.dev, code_sz, dma_addr, GFP_KERNEL);
if (!*code)
return -ENOMEM;
@@ -180,7 +180,7 @@ nfp_net_bpf_offload_prepare(struct nfp_net *nn,
return 0;
out:
- dma_free_coherent(nn->dev, code_sz, *code, *dma_addr);
+ dma_free_coherent(nn->dp.dev, code_sz, *code, *dma_addr);
return ret;
}
@@ -193,7 +193,7 @@ nfp_net_bpf_load_and_start(struct nfp_net *nn, u32 tc_flags,
u64 bpf_addr = dma_addr;
int err;
- nn->bpf_offload_skip_sw = !!(tc_flags & TCA_CLS_FLAGS_SKIP_SW);
+ nn->dp.bpf_offload_skip_sw = !!(tc_flags & TCA_CLS_FLAGS_SKIP_SW);
if (dense_mode)
bpf_addr |= NFP_NET_CFG_BPF_CFG_8CTX;
@@ -207,13 +207,13 @@ nfp_net_bpf_load_and_start(struct nfp_net *nn, u32 tc_flags,
nn_err(nn, "FW command error while loading BPF: %d\n", err);
/* Enable passing packets through BPF function */
- nn->ctrl |= NFP_NET_CFG_CTRL_BPF;
- nn_writel(nn, NFP_NET_CFG_CTRL, nn->ctrl);
+ nn->dp.ctrl |= NFP_NET_CFG_CTRL_BPF;
+ nn_writel(nn, NFP_NET_CFG_CTRL, nn->dp.ctrl);
err = nfp_net_reconfig(nn, NFP_NET_CFG_UPDATE_GEN);
if (err)
nn_err(nn, "FW command error while enabling BPF: %d\n", err);
- dma_free_coherent(nn->dev, code_sz, code, dma_addr);
+ dma_free_coherent(nn->dp.dev, code_sz, code, dma_addr);
nfp_net_bpf_stats_reset(nn);
mod_timer(&nn->rx_filter_stats_timer, jiffies + NFP_NET_STAT_POLL_IVL);
@@ -221,16 +221,16 @@ nfp_net_bpf_load_and_start(struct nfp_net *nn, u32 tc_flags,
static int nfp_net_bpf_stop(struct nfp_net *nn)
{
- if (!(nn->ctrl & NFP_NET_CFG_CTRL_BPF))
+ if (!(nn->dp.ctrl & NFP_NET_CFG_CTRL_BPF))
return 0;
spin_lock_bh(&nn->rx_filter_lock);
- nn->ctrl &= ~NFP_NET_CFG_CTRL_BPF;
+ nn->dp.ctrl &= ~NFP_NET_CFG_CTRL_BPF;
spin_unlock_bh(&nn->rx_filter_lock);
- nn_writel(nn, NFP_NET_CFG_CTRL, nn->ctrl);
+ nn_writel(nn, NFP_NET_CFG_CTRL, nn->dp.ctrl);
del_timer_sync(&nn->rx_filter_stats_timer);
- nn->bpf_offload_skip_sw = 0;
+ nn->dp.bpf_offload_skip_sw = 0;
return nfp_net_reconfig(nn, NFP_NET_CFG_UPDATE_GEN);
}
@@ -254,7 +254,7 @@ int nfp_net_bpf_offload(struct nfp_net *nn, struct tc_cls_bpf_offload *cls_bpf)
* frames which didn't have BPF applied in the hardware should
* be fine if software fallback is available, though.
*/
- if (nn->bpf_offload_skip_sw)
+ if (nn->dp.bpf_offload_skip_sw)
return -EBUSY;
err = nfp_net_bpf_offload_prepare(nn, cls_bpf, &res, &code,
@@ -269,7 +269,7 @@ int nfp_net_bpf_offload(struct nfp_net *nn, struct tc_cls_bpf_offload *cls_bpf)
return 0;
case TC_CLSBPF_ADD:
- if (nn->ctrl & NFP_NET_CFG_CTRL_BPF)
+ if (nn->dp.ctrl & NFP_NET_CFG_CTRL_BPF)
return -EBUSY;
err = nfp_net_bpf_offload_prepare(nn, cls_bpf, &res, &code,
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
index 39407f7cc586..bc0bdbc1c8fd 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
@@ -84,12 +84,12 @@ static void nfp_netvf_get_mac_addr(struct nfp_net *nn)
put_unaligned_be16(nn_readw(nn, NFP_NET_CFG_MACADDR + 6), &mac_addr[4]);
if (!is_valid_ether_addr(mac_addr)) {
- eth_hw_addr_random(nn->netdev);
+ eth_hw_addr_random(nn->dp.netdev);
return;
}
- ether_addr_copy(nn->netdev->dev_addr, mac_addr);
- ether_addr_copy(nn->netdev->perm_addr, mac_addr);
+ ether_addr_copy(nn->dp.netdev->dev_addr, mac_addr);
+ ether_addr_copy(nn->dp.netdev->perm_addr, mac_addr);
}
static int nfp_netvf_pci_probe(struct pci_dev *pdev,
@@ -211,7 +211,7 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev,
nn->fw_ver = fw_ver;
nn->ctrl_bar = ctrl_bar;
- nn->is_vf = 1;
+ nn->dp.is_vf = 1;
nn->stride_tx = stride;
nn->stride_rx = stride;
@@ -268,7 +268,8 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev,
num_irqs = nfp_net_irqs_alloc(pdev, vf->irq_entries,
NFP_NET_MIN_PORT_IRQS,
- NFP_NET_NON_Q_VECTORS + nn->num_r_vecs);
+ NFP_NET_NON_Q_VECTORS +
+ nn->dp.num_r_vecs);
if (!num_irqs) {
nn_warn(nn, "Unable to allocate MSI-X Vectors. Exiting\n");
err = -EIO;
@@ -282,7 +283,7 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev,
*/
nn->me_freq_mhz = 1200;
- err = nfp_net_netdev_init(nn->netdev);
+ err = nfp_net_netdev_init(nn->dp.netdev);
if (err)
goto err_irqs_disable;
@@ -327,7 +328,7 @@ static void nfp_netvf_pci_remove(struct pci_dev *pdev)
nfp_net_debugfs_dir_clean(&nn->debugfs_dir);
nfp_net_debugfs_dir_clean(&vf->ddir);
- nfp_net_netdev_clean(nn->netdev);
+ nfp_net_netdev_clean(nn->dp.netdev);
nfp_net_irqs_disable(pdev);
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 02/13] nfp: move control BAR pointer into data path structure
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 01/13] nfp: separate data path information from the reset of adapter structure Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 03/13] nfp: pass new data path to ring reconfig Jakub Kicinski
` (11 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Control BAR pointer is used to unmask interrupts so it should be
in the first cacheline of adapter structure.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 21 +++++++++++----------
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 14 +++++++-------
drivers/net/ethernet/netronome/nfp/nfp_net_main.c | 2 +-
drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c | 4 ++--
4 files changed, 21 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 7d2c38604372..90a44fad6bd5 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -447,6 +447,7 @@ struct nfp_stat_pair {
* @xdp_prog: Installed XDP program
* @tx_rings: Array of pre-allocated TX ring structures
* @rx_rings: Array of pre-allocated RX ring structures
+ * @ctrl_bar: Pointer to mapped control BAR
*
* @txd_cnt: Size of the TX ring in number of descriptors
* @rxd_cnt: Size of the RX ring in number of descriptors
@@ -474,6 +475,8 @@ struct nfp_net_dp {
struct nfp_net_tx_ring *tx_rings;
struct nfp_net_rx_ring *rx_rings;
+ u8 __iomem *ctrl_bar;
+
/* Cold data follows */
unsigned int txd_cnt;
@@ -527,7 +530,6 @@ struct nfp_net_dp {
* @vxlan_ports: VXLAN ports for RX inner csum offload communicated to HW
* @vxlan_usecnt: IPv4/IPv6 VXLAN port use counts
* @qcp_cfg: Pointer to QCP queue used for configuration notification
- * @ctrl_bar: Pointer to mapped control BAR
* @tx_bar: Pointer to mapped TX queues
* @rx_bar: Pointer to mapped FL/RX queues
* @debugfs_dir: Device directory in debugfs
@@ -595,7 +597,6 @@ struct nfp_net {
u8 __iomem *qcp_cfg;
- u8 __iomem *ctrl_bar;
u8 __iomem *tx_bar;
u8 __iomem *rx_bar;
@@ -622,42 +623,42 @@ struct nfp_net_ring_set {
*/
static inline u16 nn_readb(struct nfp_net *nn, int off)
{
- return readb(nn->ctrl_bar + off);
+ return readb(nn->dp.ctrl_bar + off);
}
static inline void nn_writeb(struct nfp_net *nn, int off, u8 val)
{
- writeb(val, nn->ctrl_bar + off);
+ writeb(val, nn->dp.ctrl_bar + off);
}
static inline u16 nn_readw(struct nfp_net *nn, int off)
{
- return readw(nn->ctrl_bar + off);
+ return readw(nn->dp.ctrl_bar + off);
}
static inline void nn_writew(struct nfp_net *nn, int off, u16 val)
{
- writew(val, nn->ctrl_bar + off);
+ writew(val, nn->dp.ctrl_bar + off);
}
static inline u32 nn_readl(struct nfp_net *nn, int off)
{
- return readl(nn->ctrl_bar + off);
+ return readl(nn->dp.ctrl_bar + off);
}
static inline void nn_writel(struct nfp_net *nn, int off, u32 val)
{
- writel(val, nn->ctrl_bar + off);
+ writel(val, nn->dp.ctrl_bar + off);
}
static inline u64 nn_readq(struct nfp_net *nn, int off)
{
- return readq(nn->ctrl_bar + off);
+ return readq(nn->dp.ctrl_bar + off);
}
static inline void nn_writeq(struct nfp_net *nn, int off, u64 val)
{
- writeq(val, nn->ctrl_bar + off);
+ writeq(val, nn->dp.ctrl_bar + off);
}
/* Flush posted PCI writes by reading something without side effects */
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 4620c1bba96e..969c30589f23 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -307,7 +307,7 @@ static void nfp_net_get_stats(struct net_device *netdev,
break;
case NFP_NET_DEV_ET_STATS:
- io_p = nn->ctrl_bar + nfp_net_et_stats[i].off;
+ io_p = nn->dp.ctrl_bar + nfp_net_et_stats[i].off;
data[i] = readq(io_p);
break;
}
@@ -339,15 +339,15 @@ static void nfp_net_get_stats(struct net_device *netdev,
for (j = 0; j < NN_ET_RVEC_GATHER_STATS; j++)
data[i++] = gathered_stats[j];
for (j = 0; j < nn->dp.num_tx_rings; j++) {
- io_p = nn->ctrl_bar + NFP_NET_CFG_TXR_STATS(j);
+ io_p = nn->dp.ctrl_bar + NFP_NET_CFG_TXR_STATS(j);
data[i++] = readq(io_p);
- io_p = nn->ctrl_bar + NFP_NET_CFG_TXR_STATS(j) + 8;
+ io_p = nn->dp.ctrl_bar + NFP_NET_CFG_TXR_STATS(j) + 8;
data[i++] = readq(io_p);
}
for (j = 0; j < nn->dp.num_rx_rings; j++) {
- io_p = nn->ctrl_bar + NFP_NET_CFG_RXR_STATS(j);
+ io_p = nn->dp.ctrl_bar + NFP_NET_CFG_RXR_STATS(j);
data[i++] = readq(io_p);
- io_p = nn->ctrl_bar + NFP_NET_CFG_RXR_STATS(j) + 8;
+ io_p = nn->dp.ctrl_bar + NFP_NET_CFG_RXR_STATS(j) + 8;
data[i++] = readq(io_p);
}
}
@@ -461,7 +461,7 @@ static int nfp_net_set_rss_hash_opt(struct nfp_net *nn,
if (new_rss_cfg == nn->rss_cfg)
return 0;
- writel(new_rss_cfg, nn->ctrl_bar + NFP_NET_CFG_RSS_CTRL);
+ writel(new_rss_cfg, nn->dp.ctrl_bar + NFP_NET_CFG_RSS_CTRL);
err = nfp_net_reconfig(nn, NFP_NET_CFG_UPDATE_RSS);
if (err)
return err;
@@ -573,7 +573,7 @@ static void nfp_net_get_regs(struct net_device *netdev,
regs->version = nn_readl(nn, NFP_NET_CFG_VERSION);
for (i = 0; i < NFP_NET_CFG_BAR_SZ / sizeof(u32); i++)
- regs_buf[i] = readl(nn->ctrl_bar + (i * sizeof(u32)));
+ regs_buf[i] = readl(nn->dp.ctrl_bar + (i * sizeof(u32)));
}
static int nfp_net_get_coalesce(struct net_device *netdev,
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
index 3935d19a273d..2025cb7c6d90 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c
@@ -306,7 +306,7 @@ nfp_net_pf_alloc_port_netdev(struct nfp_pf *pf, void __iomem *ctrl_bar,
nn->cpp = pf->cpp;
nn->fw_ver = *fw_ver;
- nn->ctrl_bar = ctrl_bar;
+ nn->dp.ctrl_bar = ctrl_bar;
nn->tx_bar = tx_bar;
nn->rx_bar = rx_bar;
nn->dp.is_vf = 0;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
index bc0bdbc1c8fd..86e61be6f35c 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c
@@ -210,7 +210,7 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev,
vf->nn = nn;
nn->fw_ver = fw_ver;
- nn->ctrl_bar = ctrl_bar;
+ nn->dp.ctrl_bar = ctrl_bar;
nn->dp.is_vf = 1;
nn->stride_tx = stride;
nn->stride_rx = stride;
@@ -338,7 +338,7 @@ static void nfp_netvf_pci_remove(struct pci_dev *pdev)
} else {
iounmap(vf->q_bar);
}
- iounmap(nn->ctrl_bar);
+ iounmap(nn->dp.ctrl_bar);
nfp_net_netdev_free(nn);
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 03/13] nfp: pass new data path to ring reconfig
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 01/13] nfp: separate data path information from the reset of adapter structure Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 02/13] nfp: move control BAR pointer into data path structure Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 04/13] nfp: use dp to carry number of stack tx rings and vectors Jakub Kicinski
` (10 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Make callers of nfp_net_ring_reconfig() pass newly allocated data
path structure. We will gradually make use of that structure
instead of passing parameters around to all the allocation functions.
This commit adds allocation and propagation of new data path struct,
no parameters are converted, yet.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 5 +-
.../net/ethernet/netronome/nfp/nfp_net_common.c | 108 ++++++++++++++-------
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 14 ++-
3 files changed, 91 insertions(+), 36 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 90a44fad6bd5..74f6d485351f 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -812,8 +812,11 @@ void nfp_net_irqs_disable(struct pci_dev *pdev);
void
nfp_net_irqs_assign(struct nfp_net *nn, struct msix_entry *irq_entries,
unsigned int n);
+
+struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn);
int
-nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
+nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *new,
+ struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx);
#ifdef CONFIG_NFP_DEBUG
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 951d511643f1..7afefb44b642 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1127,6 +1127,7 @@ nfp_net_free_frag(void *frag, bool xdp)
/**
* nfp_net_rx_alloc_one() - Allocate and map page frag for RX
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring structure of the skb
* @dma_addr: Pointer to storage for DMA address (output param)
* @fl_bufsz: size of freelist buffers
@@ -1137,10 +1138,10 @@ nfp_net_free_frag(void *frag, bool xdp)
* Return: allocated page frag or NULL on failure.
*/
static void *
-nfp_net_rx_alloc_one(struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
+nfp_net_rx_alloc_one(struct nfp_net_dp *dp,
+ struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
unsigned int fl_bufsz, bool xdp)
{
- struct nfp_net_dp *dp = &rx_ring->r_vec->nfp_net->dp;
int direction;
void *frag;
@@ -1299,7 +1300,7 @@ nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
for (i = 0; i < rx_ring->cnt - 1; i++) {
rxbufs[i].frag =
- nfp_net_rx_alloc_one(rx_ring, &rxbufs[i].dma_addr,
+ nfp_net_rx_alloc_one(dp, rx_ring, &rxbufs[i].dma_addr,
rx_ring->bufsz, xdp);
if (!rxbufs[i].frag) {
nfp_net_rx_ring_bufs_free(dp, rx_ring, xdp);
@@ -1784,7 +1785,8 @@ nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
}
static struct nfp_net_tx_ring *
-nfp_net_tx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
+nfp_net_tx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct nfp_net_ring_set *s,
unsigned int num_stack_tx_rings)
{
struct nfp_net_tx_ring *rings;
@@ -1900,10 +1902,10 @@ nfp_net_rx_ring_alloc(struct nfp_net_rx_ring *rx_ring, unsigned int fl_bufsz,
}
static struct nfp_net_rx_ring *
-nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
- bool xdp)
+nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct nfp_net_ring_set *s, bool xdp)
{
- unsigned int fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, s->mtu);
+ unsigned int fl_bufsz = nfp_net_calc_fl_bufsz(dp, s->mtu);
struct nfp_net_rx_ring *rings;
unsigned int r;
@@ -1917,7 +1919,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
if (nfp_net_rx_ring_alloc(&rings[r], fl_bufsz, s->dcnt))
goto err_free_prev;
- if (nfp_net_rx_ring_bufs_alloc(&nn->dp, &rings[r], xdp))
+ if (nfp_net_rx_ring_bufs_alloc(dp, &rings[r], xdp))
goto err_free_ring;
}
@@ -1925,7 +1927,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_ring_set *s,
err_free_prev:
while (r--) {
- nfp_net_rx_ring_bufs_free(&nn->dp, &rings[r], xdp);
+ nfp_net_rx_ring_bufs_free(dp, &rings[r], xdp);
err_free_ring:
nfp_net_rx_ring_free(&rings[r]);
}
@@ -2295,14 +2297,15 @@ static int nfp_net_netdev_open(struct net_device *netdev)
goto err_cleanup_vec_p;
}
- nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &rx, nn->dp.xdp_prog);
+ nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &nn->dp, &rx,
+ nn->dp.xdp_prog);
if (!nn->dp.rx_rings) {
err = -ENOMEM;
goto err_cleanup_vec;
}
- nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &tx,
- nn->dp.num_stack_tx_rings);
+ nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &nn->dp, &tx,
+ nn->dp.num_stack_tx_rings);
if (!nn->dp.tx_rings) {
err = -ENOMEM;
goto err_free_rx_rings;
@@ -2464,7 +2467,8 @@ static void nfp_net_rss_init_itbl(struct nfp_net *nn)
}
static int
-nfp_net_ring_swap_enable(struct nfp_net *nn, unsigned int *num_vecs,
+nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
+ unsigned int *num_vecs,
unsigned int *stack_tx_rings,
struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
@@ -2502,14 +2506,35 @@ nfp_net_ring_swap_enable(struct nfp_net *nn, unsigned int *num_vecs,
return __nfp_net_set_config_and_enable(nn);
}
+struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn)
+{
+ struct nfp_net_dp *new;
+
+ new = kmalloc(sizeof(*new), GFP_KERNEL);
+ if (!new)
+ return NULL;
+
+ *new = nn->dp;
+
+ /* Clear things which need to be recomputed */
+ new->fl_bufsz = 0;
+ new->tx_rings = NULL;
+ new->rx_rings = NULL;
+ new->num_r_vecs = 0;
+ new->num_stack_tx_rings = 0;
+
+ return new;
+}
+
static int
-nfp_net_check_config(struct nfp_net *nn, struct bpf_prog *xdp_prog,
+nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct bpf_prog *xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
{
/* XDP-enabled tests */
if (!xdp_prog)
return 0;
- if (rx && nfp_net_calc_fl_bufsz(&nn->dp, rx->mtu) > PAGE_SIZE) {
+ if (rx && nfp_net_calc_fl_bufsz(dp, rx->mtu) > PAGE_SIZE) {
nn_warn(nn, "MTU too large w/ XDP enabled\n");
return -EINVAL;
}
@@ -2522,7 +2547,8 @@ nfp_net_check_config(struct nfp_net *nn, struct bpf_prog *xdp_prog,
}
static void
-nfp_net_ring_reconfig_down(struct nfp_net *nn, struct bpf_prog **xdp_prog,
+nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
struct nfp_net_ring_set *tx,
unsigned int stack_tx_rings, unsigned int num_vecs)
@@ -2542,26 +2568,28 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct bpf_prog **xdp_prog,
}
int
-nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
+nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
{
unsigned int stack_tx_rings, num_vecs, r;
int err;
- stack_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
+ stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
if (*xdp_prog)
- stack_tx_rings -= rx ? rx->n_rings : nn->dp.num_rx_rings;
+ stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
- num_vecs = max(rx ? rx->n_rings : nn->dp.num_rx_rings, stack_tx_rings);
+ num_vecs = max(rx ? rx->n_rings : dp->num_rx_rings, stack_tx_rings);
- err = nfp_net_check_config(nn, *xdp_prog, rx, tx);
+ err = nfp_net_check_config(nn, dp, *xdp_prog, rx, tx);
if (err)
- return err;
+ goto exit_free_dp;
- if (!netif_running(nn->dp.netdev)) {
- nfp_net_ring_reconfig_down(nn, xdp_prog, rx, tx,
+ if (!netif_running(dp->netdev)) {
+ nfp_net_ring_reconfig_down(nn, dp, xdp_prog, rx, tx,
stack_tx_rings, num_vecs);
- return 0;
+ err = 0;
+ goto exit_free_dp;
}
/* Prepare new rings */
@@ -2573,13 +2601,13 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
}
}
if (rx) {
- if (!nfp_net_rx_ring_set_prepare(nn, rx, *xdp_prog)) {
+ if (!nfp_net_rx_ring_set_prepare(nn, dp, rx, *xdp_prog)) {
err = -ENOMEM;
goto err_cleanup_vecs;
}
}
if (tx) {
- if (!nfp_net_tx_ring_set_prepare(nn, tx, stack_tx_rings)) {
+ if (!nfp_net_tx_ring_set_prepare(nn, dp, tx, stack_tx_rings)) {
err = -ENOMEM;
goto err_free_rx;
}
@@ -2589,7 +2617,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
nfp_net_close_stack(nn);
nfp_net_clear_config_and_disable(nn);
- err = nfp_net_ring_swap_enable(nn, &num_vecs, &stack_tx_rings,
+ err = nfp_net_ring_swap_enable(nn, dp, &num_vecs, &stack_tx_rings,
xdp_prog, rx, tx);
if (err) {
int err2;
@@ -2597,7 +2625,8 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
nfp_net_clear_config_and_disable(nn);
/* Try with old configuration and old rings */
- err2 = nfp_net_ring_swap_enable(nn, &num_vecs, &stack_tx_rings,
+ err2 = nfp_net_ring_swap_enable(nn, dp, &num_vecs,
+ &stack_tx_rings,
xdp_prog, rx, tx);
if (err2)
nn_err(nn, "Can't restore ring config - FW communication failed (%d,%d)\n",
@@ -2607,20 +2636,23 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct bpf_prog **xdp_prog,
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
if (rx)
- nfp_net_rx_ring_set_free(&nn->dp, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(dp, rx, *xdp_prog);
if (tx)
nfp_net_tx_ring_set_free(tx);
nfp_net_open_stack(nn);
+exit_free_dp:
+ kfree(dp);
return err;
err_free_rx:
if (rx)
- nfp_net_rx_ring_set_free(&nn->dp, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(dp, rx, *xdp_prog);
err_cleanup_vecs:
for (r = num_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
+ kfree(dp);
return err;
}
@@ -2632,8 +2664,13 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
.mtu = new_mtu,
.dcnt = nn->dp.rxd_cnt,
};
+ struct nfp_net_dp *dp;
+
+ dp = nfp_net_clone_dp(nn);
+ if (!dp)
+ return -ENOMEM;
- return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog, &rx, NULL);
+ return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog, &rx, NULL);
}
static void nfp_net_stat64(struct net_device *netdev,
@@ -2960,6 +2997,7 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
.n_rings = nn->dp.num_tx_rings,
.dcnt = nn->dp.txd_cnt,
};
+ struct nfp_net_dp *dp;
int err;
if (prog && prog->xdp_adjust_head) {
@@ -2975,10 +3013,14 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
return 0;
}
+ dp = nfp_net_clone_dp(nn);
+ if (!dp)
+ return -ENOMEM;
+
tx.n_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
- err = nfp_net_ring_reconfig(nn, &prog, &rx, &tx);
+ err = nfp_net_ring_reconfig(nn, dp, &prog, &rx, &tx);
if (err)
return err;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 969c30589f23..326ccd74a4bf 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -196,13 +196,18 @@ static int nfp_net_set_ring_size(struct nfp_net *nn, u32 rxd_cnt, u32 txd_cnt)
.n_rings = nn->dp.num_tx_rings,
.dcnt = txd_cnt,
};
+ struct nfp_net_dp *dp;
if (nn->dp.rxd_cnt != rxd_cnt)
reconfig_rx = ℞
if (nn->dp.txd_cnt != txd_cnt)
reconfig_tx = &tx;
- return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog,
+ dp = nfp_net_clone_dp(nn);
+ if (!dp)
+ return -ENOMEM;
+
+ return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog,
reconfig_rx, reconfig_tx);
}
@@ -772,6 +777,7 @@ static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
.n_rings = total_tx,
.dcnt = nn->dp.txd_cnt,
};
+ struct nfp_net_dp *dp;
if (nn->dp.num_rx_rings != total_rx)
reconfig_rx = ℞
@@ -783,7 +789,11 @@ static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
if (nn->dp.xdp_prog)
tx.n_rings += total_rx;
- return nfp_net_ring_reconfig(nn, &nn->dp.xdp_prog,
+ dp = nfp_net_clone_dp(nn);
+ if (!dp)
+ return -ENOMEM;
+
+ return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog,
reconfig_rx, reconfig_tx);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 04/13] nfp: use dp to carry number of stack tx rings and vectors
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (2 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 03/13] nfp: pass new data path to ring reconfig Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 05/13] nfp: use dp to carry fl_bufsz at reconfig time Jakub Kicinski
` (9 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Instead of passing variables around use dp to store number of tx rings
for the stack and number of IRQ vectors.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
.../net/ethernet/netronome/nfp/nfp_net_common.c | 62 +++++++++++-----------
1 file changed, 31 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 7afefb44b642..52f0e9dfd15a 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1786,8 +1786,7 @@ nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
static struct nfp_net_tx_ring *
nfp_net_tx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *s,
- unsigned int num_stack_tx_rings)
+ struct nfp_net_ring_set *s)
{
struct nfp_net_tx_ring *rings;
unsigned int r;
@@ -1799,8 +1798,8 @@ nfp_net_tx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
for (r = 0; r < s->n_rings; r++) {
int bias = 0;
- if (r >= num_stack_tx_rings)
- bias = num_stack_tx_rings;
+ if (r >= dp->num_stack_tx_rings)
+ bias = dp->num_stack_tx_rings;
nfp_net_tx_ring_init(&rings[r], &nn->r_vecs[r - bias], r);
@@ -2304,8 +2303,7 @@ static int nfp_net_netdev_open(struct net_device *netdev)
goto err_cleanup_vec;
}
- nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &nn->dp, &tx,
- nn->dp.num_stack_tx_rings);
+ nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &nn->dp, &tx);
if (!nn->dp.tx_rings) {
err = -ENOMEM;
goto err_free_rx_rings;
@@ -2466,10 +2464,16 @@ static void nfp_net_rss_init_itbl(struct nfp_net *nn)
ethtool_rxfh_indir_default(i, nn->dp.num_rx_rings);
}
+static void nfp_net_dp_swap(struct nfp_net *nn, struct nfp_net_dp *dp)
+{
+ struct nfp_net_dp new_dp = *dp;
+
+ *dp = nn->dp;
+ nn->dp = new_dp;
+}
+
static int
nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
- unsigned int *num_vecs,
- unsigned int *stack_tx_rings,
struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
struct nfp_net_ring_set *tx)
@@ -2482,8 +2486,8 @@ nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
if (tx)
nfp_net_tx_ring_set_swap(nn, tx);
- swap(*num_vecs, nn->dp.num_r_vecs);
- swap(*stack_tx_rings, nn->dp.num_stack_tx_rings);
+ swap(dp->num_r_vecs, nn->dp.num_r_vecs);
+ swap(dp->num_stack_tx_rings, nn->dp.num_stack_tx_rings);
*xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
for (r = 0; r < nn->max_r_vecs; r++)
@@ -2550,17 +2554,16 @@ static void
nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
- struct nfp_net_ring_set *tx,
- unsigned int stack_tx_rings, unsigned int num_vecs)
+ struct nfp_net_ring_set *tx)
{
+ nfp_net_dp_swap(nn, dp);
+
nn->dp.netdev->mtu = rx ? rx->mtu : nn->dp.netdev->mtu;
nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, nn->dp.netdev->mtu);
nn->dp.rxd_cnt = rx ? rx->dcnt : nn->dp.rxd_cnt;
nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
nn->dp.num_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
- nn->dp.num_stack_tx_rings = stack_tx_rings;
- nn->dp.num_r_vecs = num_vecs;
*xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
if (!netif_is_rxfh_configured(nn->dp.netdev))
@@ -2572,31 +2575,31 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
{
- unsigned int stack_tx_rings, num_vecs, r;
- int err;
+ int r, err;
- stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
+ dp->num_stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
if (*xdp_prog)
- stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
+ dp->num_stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
- num_vecs = max(rx ? rx->n_rings : dp->num_rx_rings, stack_tx_rings);
+ dp->num_r_vecs = max(rx ? rx->n_rings : dp->num_rx_rings,
+ dp->num_stack_tx_rings);
err = nfp_net_check_config(nn, dp, *xdp_prog, rx, tx);
if (err)
goto exit_free_dp;
if (!netif_running(dp->netdev)) {
- nfp_net_ring_reconfig_down(nn, dp, xdp_prog, rx, tx,
- stack_tx_rings, num_vecs);
+ nfp_net_ring_reconfig_down(nn, dp, xdp_prog, rx, tx);
+
err = 0;
goto exit_free_dp;
}
/* Prepare new rings */
- for (r = nn->dp.num_r_vecs; r < num_vecs; r++) {
+ for (r = nn->dp.num_r_vecs; r < dp->num_r_vecs; r++) {
err = nfp_net_prepare_vector(nn, &nn->r_vecs[r], r);
if (err) {
- num_vecs = r;
+ dp->num_r_vecs = r;
goto err_cleanup_vecs;
}
}
@@ -2607,7 +2610,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
}
}
if (tx) {
- if (!nfp_net_tx_ring_set_prepare(nn, dp, tx, stack_tx_rings)) {
+ if (!nfp_net_tx_ring_set_prepare(nn, dp, tx)) {
err = -ENOMEM;
goto err_free_rx;
}
@@ -2617,22 +2620,19 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
nfp_net_close_stack(nn);
nfp_net_clear_config_and_disable(nn);
- err = nfp_net_ring_swap_enable(nn, dp, &num_vecs, &stack_tx_rings,
- xdp_prog, rx, tx);
+ err = nfp_net_ring_swap_enable(nn, dp, xdp_prog, rx, tx);
if (err) {
int err2;
nfp_net_clear_config_and_disable(nn);
/* Try with old configuration and old rings */
- err2 = nfp_net_ring_swap_enable(nn, dp, &num_vecs,
- &stack_tx_rings,
- xdp_prog, rx, tx);
+ err2 = nfp_net_ring_swap_enable(nn, dp, xdp_prog, rx, tx);
if (err2)
nn_err(nn, "Can't restore ring config - FW communication failed (%d,%d)\n",
err, err2);
}
- for (r = num_vecs - 1; r >= nn->dp.num_r_vecs; r--)
+ for (r = dp->num_r_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
if (rx)
@@ -2650,7 +2650,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
if (rx)
nfp_net_rx_ring_set_free(dp, rx, *xdp_prog);
err_cleanup_vecs:
- for (r = num_vecs - 1; r >= nn->dp.num_r_vecs; r--)
+ for (r = dp->num_r_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
kfree(dp);
return err;
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 05/13] nfp: use dp to carry fl_bufsz at reconfig time
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (3 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 04/13] nfp: use dp to carry number of stack tx rings and vectors Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 06/13] nfp: use dp to carry mtu " Jakub Kicinski
` (8 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Use fl_bufsz member of data path struct to carry desired size of
free list entries.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 3 --
.../net/ethernet/netronome/nfp/nfp_net_common.c | 56 +++++++++++-----------
2 files changed, 27 insertions(+), 32 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 74f6d485351f..ab5865b955dd 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -316,8 +316,6 @@ struct nfp_net_rx_buf {
* @rxds: Virtual address of FL/RX ring in host memory
* @dma: DMA address of the FL/RX ring
* @size: Size, in bytes, of the FL/RX ring (needed to free)
- * @bufsz: Buffer allocation size for convenience of management routines
- * (NOTE: this is in second cache line, do not use on fast path!)
*/
struct nfp_net_rx_ring {
struct nfp_net_r_vector *r_vec;
@@ -339,7 +337,6 @@ struct nfp_net_rx_ring {
dma_addr_t dma;
unsigned int size;
- unsigned int bufsz;
} ____cacheline_aligned;
/**
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 52f0e9dfd15a..92d4c2991a85 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -86,19 +86,19 @@ void nfp_net_get_fw_version(struct nfp_net_fw_version *fw_ver,
}
static dma_addr_t
-nfp_net_dma_map_rx(struct nfp_net_dp *dp, void *frag, unsigned int bufsz,
- int direction)
+nfp_net_dma_map_rx(struct nfp_net_dp *dp, void *frag, int direction)
{
return dma_map_single(dp->dev, frag + NFP_NET_RX_BUF_HEADROOM,
- bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
+ dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA,
+ direction);
}
static void
nfp_net_dma_unmap_rx(struct nfp_net_dp *dp, dma_addr_t dma_addr,
- unsigned int bufsz, int direction)
+ int direction)
{
dma_unmap_single(dp->dev, dma_addr,
- bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
+ dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
}
/* Firmware reconfig
@@ -992,7 +992,7 @@ static void nfp_net_xdp_complete(struct nfp_net_tx_ring *tx_ring)
continue;
nfp_net_dma_unmap_rx(dp, tx_ring->txbufs[idx].dma_addr,
- dp->fl_bufsz, DMA_BIDIRECTIONAL);
+ DMA_BIDIRECTIONAL);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
done_pkts++;
@@ -1038,7 +1038,7 @@ nfp_net_tx_ring_reset(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring)
if (tx_ring == r_vec->xdp_ring) {
nfp_net_dma_unmap_rx(dp, tx_buf->dma_addr,
- dp->fl_bufsz, DMA_BIDIRECTIONAL);
+ DMA_BIDIRECTIONAL);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
} else {
struct sk_buff *skb = tx_ring->txbufs[idx].skb;
@@ -1130,7 +1130,6 @@ nfp_net_free_frag(void *frag, bool xdp)
* @dp: NFP Net data path struct
* @rx_ring: RX ring structure of the skb
* @dma_addr: Pointer to storage for DMA address (output param)
- * @fl_bufsz: size of freelist buffers
* @xdp: Whether XDP is enabled
*
* This function will allcate a new page frag, map it for DMA.
@@ -1140,13 +1139,13 @@ nfp_net_free_frag(void *frag, bool xdp)
static void *
nfp_net_rx_alloc_one(struct nfp_net_dp *dp,
struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
- unsigned int fl_bufsz, bool xdp)
+ bool xdp)
{
int direction;
void *frag;
if (!xdp)
- frag = netdev_alloc_frag(fl_bufsz);
+ frag = netdev_alloc_frag(dp->fl_bufsz);
else
frag = page_address(alloc_page(GFP_KERNEL | __GFP_COLD));
if (!frag) {
@@ -1156,7 +1155,7 @@ nfp_net_rx_alloc_one(struct nfp_net_dp *dp,
direction = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
- *dma_addr = nfp_net_dma_map_rx(dp, frag, fl_bufsz, direction);
+ *dma_addr = nfp_net_dma_map_rx(dp, frag, direction);
if (dma_mapping_error(dp->dev, *dma_addr)) {
nfp_net_free_frag(frag, xdp);
nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
@@ -1181,7 +1180,7 @@ nfp_net_napi_alloc_one(struct nfp_net_dp *dp, int direction,
return NULL;
}
- *dma_addr = nfp_net_dma_map_rx(dp, frag, dp->fl_bufsz, direction);
+ *dma_addr = nfp_net_dma_map_rx(dp, frag, direction);
if (dma_mapping_error(dp->dev, *dma_addr)) {
nfp_net_free_frag(frag, dp->xdp_prog);
nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
@@ -1276,7 +1275,7 @@ nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
continue;
nfp_net_dma_unmap_rx(dp, rx_ring->rxbufs[i].dma_addr,
- rx_ring->bufsz, direction);
+ direction);
nfp_net_free_frag(rx_ring->rxbufs[i].frag, xdp);
rx_ring->rxbufs[i].dma_addr = 0;
rx_ring->rxbufs[i].frag = NULL;
@@ -1301,7 +1300,7 @@ nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
for (i = 0; i < rx_ring->cnt - 1; i++) {
rxbufs[i].frag =
nfp_net_rx_alloc_one(dp, rx_ring, &rxbufs[i].dma_addr,
- rx_ring->bufsz, xdp);
+ xdp);
if (!rxbufs[i].frag) {
nfp_net_rx_ring_bufs_free(dp, rx_ring, xdp);
return -ENOMEM;
@@ -1652,8 +1651,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
continue;
}
- nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr, dp->fl_bufsz,
- rx_dma_map_dir);
+ nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr, rx_dma_map_dir);
nfp_net_rx_give_one(rx_ring, new_frag, new_dma_addr);
@@ -1865,23 +1863,19 @@ static void nfp_net_rx_ring_free(struct nfp_net_rx_ring *rx_ring)
/**
* nfp_net_rx_ring_alloc() - Allocate resource for a RX ring
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring to allocate
- * @fl_bufsz: Size of buffers to allocate
* @cnt: Ring buffer count
*
* Return: 0 on success, negative errno otherwise.
*/
static int
-nfp_net_rx_ring_alloc(struct nfp_net_rx_ring *rx_ring, unsigned int fl_bufsz,
+nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
u32 cnt)
{
- struct nfp_net_r_vector *r_vec = rx_ring->r_vec;
- struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
int sz;
rx_ring->cnt = cnt;
- rx_ring->bufsz = fl_bufsz;
-
rx_ring->size = sizeof(*rx_ring->rxds) * rx_ring->cnt;
rx_ring->rxds = dma_zalloc_coherent(dp->dev, rx_ring->size,
&rx_ring->dma, GFP_KERNEL);
@@ -1904,7 +1898,6 @@ static struct nfp_net_rx_ring *
nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
struct nfp_net_ring_set *s, bool xdp)
{
- unsigned int fl_bufsz = nfp_net_calc_fl_bufsz(dp, s->mtu);
struct nfp_net_rx_ring *rings;
unsigned int r;
@@ -1915,7 +1908,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
for (r = 0; r < s->n_rings; r++) {
nfp_net_rx_ring_init(&rings[r], &nn->r_vecs[r], r);
- if (nfp_net_rx_ring_alloc(&rings[r], fl_bufsz, s->dcnt))
+ if (nfp_net_rx_ring_alloc(dp, &rings[r], s->dcnt))
goto err_free_prev;
if (nfp_net_rx_ring_bufs_alloc(dp, &rings[r], xdp))
@@ -1935,17 +1928,20 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
}
static void
-nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_ring_set *s)
+nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_dp *dp,
+ struct nfp_net_ring_set *s)
{
struct nfp_net_ring_set new = *s;
+ struct nfp_net_dp new_dp = *dp;
+ dp->fl_bufsz = nn->dp.fl_bufsz;
s->mtu = nn->dp.netdev->mtu;
s->dcnt = nn->dp.rxd_cnt;
s->rings = nn->dp.rx_rings;
s->n_rings = nn->dp.num_rx_rings;
nn->dp.netdev->mtu = new.mtu;
- nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, new.mtu);
+ nn->dp.fl_bufsz = new_dp.fl_bufsz;
nn->dp.rxd_cnt = new.dcnt;
nn->dp.rx_rings = new.rings;
nn->dp.num_rx_rings = new.n_rings;
@@ -2482,7 +2478,7 @@ nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
int err;
if (rx)
- nfp_net_rx_ring_set_swap(nn, rx);
+ nfp_net_rx_ring_set_swap(nn, dp, rx);
if (tx)
nfp_net_tx_ring_set_swap(nn, tx);
@@ -2538,7 +2534,7 @@ nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
/* XDP-enabled tests */
if (!xdp_prog)
return 0;
- if (rx && nfp_net_calc_fl_bufsz(dp, rx->mtu) > PAGE_SIZE) {
+ if (dp->fl_bufsz > PAGE_SIZE) {
nn_warn(nn, "MTU too large w/ XDP enabled\n");
return -EINVAL;
}
@@ -2559,7 +2555,6 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
nfp_net_dp_swap(nn, dp);
nn->dp.netdev->mtu = rx ? rx->mtu : nn->dp.netdev->mtu;
- nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, nn->dp.netdev->mtu);
nn->dp.rxd_cnt = rx ? rx->dcnt : nn->dp.rxd_cnt;
nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
@@ -2577,6 +2572,9 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
{
int r, err;
+ dp->fl_bufsz = nfp_net_calc_fl_bufsz(dp,
+ rx ? rx->mtu : nn->dp.netdev->mtu);
+
dp->num_stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
if (*xdp_prog)
dp->num_stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 06/13] nfp: use dp to carry mtu at reconfig time
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (4 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 05/13] nfp: use dp to carry fl_bufsz at reconfig time Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 07/13] nfp: use dp to carry xdp_prog " Jakub Kicinski
` (7 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Move the mtu member from ring set to data path struct.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 4 +++-
.../net/ethernet/netronome/nfp/nfp_net_common.c | 23 +++++++++++-----------
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 2 --
3 files changed, 15 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index ab5865b955dd..84774c281b61 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -452,6 +452,7 @@ struct nfp_stat_pair {
* @num_tx_rings: Currently configured number of TX rings
* @num_stack_tx_rings: Number of TX rings used by the stack (not XDP)
* @num_rx_rings: Currently configured number of RX rings
+ * @mtu: Device MTU
*/
struct nfp_net_dp {
struct device *dev;
@@ -484,6 +485,8 @@ struct nfp_net_dp {
unsigned int num_tx_rings;
unsigned int num_stack_tx_rings;
unsigned int num_rx_rings;
+
+ unsigned int mtu;
};
/**
@@ -610,7 +613,6 @@ struct nfp_net {
struct nfp_net_ring_set {
unsigned int n_rings;
- unsigned int mtu;
unsigned int dcnt;
void *rings;
};
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 92d4c2991a85..862e86cb5688 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1099,7 +1099,7 @@ static void nfp_net_tx_timeout(struct net_device *netdev)
/* Receive processing
*/
static unsigned int
-nfp_net_calc_fl_bufsz(struct nfp_net_dp *dp, unsigned int mtu)
+nfp_net_calc_fl_bufsz(struct nfp_net_dp *dp)
{
unsigned int fl_bufsz;
@@ -1108,7 +1108,7 @@ nfp_net_calc_fl_bufsz(struct nfp_net_dp *dp, unsigned int mtu)
fl_bufsz += NFP_NET_MAX_PREPEND;
else
fl_bufsz += dp->rx_offset;
- fl_bufsz += ETH_HLEN + VLAN_HLEN * 2 + mtu;
+ fl_bufsz += ETH_HLEN + VLAN_HLEN * 2 + dp->mtu;
fl_bufsz = SKB_DATA_ALIGN(fl_bufsz);
fl_bufsz += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
@@ -1935,12 +1935,13 @@ nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_dp *dp,
struct nfp_net_dp new_dp = *dp;
dp->fl_bufsz = nn->dp.fl_bufsz;
- s->mtu = nn->dp.netdev->mtu;
+ dp->mtu = nn->dp.netdev->mtu;
s->dcnt = nn->dp.rxd_cnt;
s->rings = nn->dp.rx_rings;
s->n_rings = nn->dp.num_rx_rings;
- nn->dp.netdev->mtu = new.mtu;
+ nn->dp.mtu = new_dp.mtu;
+ nn->dp.netdev->mtu = new_dp.mtu;
nn->dp.fl_bufsz = new_dp.fl_bufsz;
nn->dp.rxd_cnt = new.dcnt;
nn->dp.rx_rings = new.rings;
@@ -2255,7 +2256,6 @@ static int nfp_net_netdev_open(struct net_device *netdev)
struct nfp_net *nn = netdev_priv(netdev);
struct nfp_net_ring_set rx = {
.n_rings = nn->dp.num_rx_rings,
- .mtu = nn->dp.netdev->mtu,
.dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
@@ -2466,6 +2466,8 @@ static void nfp_net_dp_swap(struct nfp_net *nn, struct nfp_net_dp *dp)
*dp = nn->dp;
nn->dp = new_dp;
+
+ nn->dp.netdev->mtu = new_dp.mtu;
}
static int
@@ -2554,7 +2556,6 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
{
nfp_net_dp_swap(nn, dp);
- nn->dp.netdev->mtu = rx ? rx->mtu : nn->dp.netdev->mtu;
nn->dp.rxd_cnt = rx ? rx->dcnt : nn->dp.rxd_cnt;
nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
@@ -2572,8 +2573,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
{
int r, err;
- dp->fl_bufsz = nfp_net_calc_fl_bufsz(dp,
- rx ? rx->mtu : nn->dp.netdev->mtu);
+ dp->fl_bufsz = nfp_net_calc_fl_bufsz(dp);
dp->num_stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
if (*xdp_prog)
@@ -2659,7 +2659,6 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
struct nfp_net *nn = netdev_priv(netdev);
struct nfp_net_ring_set rx = {
.n_rings = nn->dp.num_rx_rings,
- .mtu = new_mtu,
.dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_dp *dp;
@@ -2668,6 +2667,8 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
if (!dp)
return -ENOMEM;
+ dp->mtu = new_mtu;
+
return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog, &rx, NULL);
}
@@ -2988,7 +2989,6 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
{
struct nfp_net_ring_set rx = {
.n_rings = nn->dp.num_rx_rings,
- .mtu = nn->dp.netdev->mtu,
.dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
@@ -3263,7 +3263,8 @@ int nfp_net_netdev_init(struct net_device *netdev)
netdev->mtu = nn->max_mtu;
else
netdev->mtu = NFP_NET_DEFAULT_MTU;
- nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp, netdev->mtu);
+ nn->dp.mtu = netdev->mtu;
+ nn->dp.fl_bufsz = nfp_net_calc_fl_bufsz(&nn->dp);
/* Advertise/enable offloads based on capabilities
*
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 326ccd74a4bf..eccb01f3659f 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -189,7 +189,6 @@ static int nfp_net_set_ring_size(struct nfp_net *nn, u32 rxd_cnt, u32 txd_cnt)
struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
struct nfp_net_ring_set rx = {
.n_rings = nn->dp.num_rx_rings,
- .mtu = nn->dp.netdev->mtu,
.dcnt = rxd_cnt,
};
struct nfp_net_ring_set tx = {
@@ -770,7 +769,6 @@ static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
struct nfp_net_ring_set rx = {
.n_rings = total_rx,
- .mtu = nn->dp.netdev->mtu,
.dcnt = nn->dp.rxd_cnt,
};
struct nfp_net_ring_set tx = {
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 07/13] nfp: use dp to carry xdp_prog at reconfig time
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (5 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 06/13] nfp: use dp to carry mtu " Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 08/13] nfp: switch to using data path structures for reconfiguration Jakub Kicinski
` (6 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Use xdp_prog member of data path struct to carry the xdp_prog to
alloc/free free functions.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 1 -
.../net/ethernet/netronome/nfp/nfp_net_common.c | 82 +++++++++-------------
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 6 +-
3 files changed, 37 insertions(+), 52 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 84774c281b61..19dacc3f1269 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -815,7 +815,6 @@ nfp_net_irqs_assign(struct nfp_net *nn, struct msix_entry *irq_entries,
struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn);
int
nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *new,
- struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx);
#ifdef CONFIG_NFP_DEBUG
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 862e86cb5688..6ab824a48d1d 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1130,21 +1130,19 @@ nfp_net_free_frag(void *frag, bool xdp)
* @dp: NFP Net data path struct
* @rx_ring: RX ring structure of the skb
* @dma_addr: Pointer to storage for DMA address (output param)
- * @xdp: Whether XDP is enabled
*
* This function will allcate a new page frag, map it for DMA.
*
* Return: allocated page frag or NULL on failure.
*/
static void *
-nfp_net_rx_alloc_one(struct nfp_net_dp *dp,
- struct nfp_net_rx_ring *rx_ring, dma_addr_t *dma_addr,
- bool xdp)
+nfp_net_rx_alloc_one(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
+ dma_addr_t *dma_addr)
{
int direction;
void *frag;
- if (!xdp)
+ if (!dp->xdp_prog)
frag = netdev_alloc_frag(dp->fl_bufsz);
else
frag = page_address(alloc_page(GFP_KERNEL | __GFP_COLD));
@@ -1153,11 +1151,11 @@ nfp_net_rx_alloc_one(struct nfp_net_dp *dp,
return NULL;
}
- direction = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
+ direction = dp->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
*dma_addr = nfp_net_dma_map_rx(dp, frag, direction);
if (dma_mapping_error(dp->dev, *dma_addr)) {
- nfp_net_free_frag(frag, xdp);
+ nfp_net_free_frag(frag, dp->xdp_prog);
nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
return NULL;
}
@@ -1253,7 +1251,6 @@ static void nfp_net_rx_ring_reset(struct nfp_net_rx_ring *rx_ring)
* nfp_net_rx_ring_bufs_free() - Free any buffers currently on the RX ring
* @dp: NFP Net data path struct
* @rx_ring: RX ring to remove buffers from
- * @xdp: Whether XDP is enabled
*
* Assumes that the device is stopped and buffers are in [0, ring->cnt - 1)
* entries. After device is disabled nfp_net_rx_ring_reset() must be called
@@ -1261,9 +1258,9 @@ static void nfp_net_rx_ring_reset(struct nfp_net_rx_ring *rx_ring)
*/
static void
nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
- struct nfp_net_rx_ring *rx_ring, bool xdp)
+ struct nfp_net_rx_ring *rx_ring)
{
- int direction = xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
+ int direction = dp->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
unsigned int i;
for (i = 0; i < rx_ring->cnt - 1; i++) {
@@ -1276,7 +1273,7 @@ nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
nfp_net_dma_unmap_rx(dp, rx_ring->rxbufs[i].dma_addr,
direction);
- nfp_net_free_frag(rx_ring->rxbufs[i].frag, xdp);
+ nfp_net_free_frag(rx_ring->rxbufs[i].frag, dp->xdp_prog);
rx_ring->rxbufs[i].dma_addr = 0;
rx_ring->rxbufs[i].frag = NULL;
}
@@ -1286,11 +1283,10 @@ nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
* nfp_net_rx_ring_bufs_alloc() - Fill RX ring with buffers (don't give to FW)
* @dp: NFP Net data path struct
* @rx_ring: RX ring to remove buffers from
- * @xdp: Whether XDP is enabled
*/
static int
nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
- struct nfp_net_rx_ring *rx_ring, bool xdp)
+ struct nfp_net_rx_ring *rx_ring)
{
struct nfp_net_rx_buf *rxbufs;
unsigned int i;
@@ -1299,10 +1295,9 @@ nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
for (i = 0; i < rx_ring->cnt - 1; i++) {
rxbufs[i].frag =
- nfp_net_rx_alloc_one(dp, rx_ring, &rxbufs[i].dma_addr,
- xdp);
+ nfp_net_rx_alloc_one(dp, rx_ring, &rxbufs[i].dma_addr);
if (!rxbufs[i].frag) {
- nfp_net_rx_ring_bufs_free(dp, rx_ring, xdp);
+ nfp_net_rx_ring_bufs_free(dp, rx_ring);
return -ENOMEM;
}
}
@@ -1896,7 +1891,7 @@ nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
static struct nfp_net_rx_ring *
nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *s, bool xdp)
+ struct nfp_net_ring_set *s)
{
struct nfp_net_rx_ring *rings;
unsigned int r;
@@ -1911,7 +1906,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
if (nfp_net_rx_ring_alloc(dp, &rings[r], s->dcnt))
goto err_free_prev;
- if (nfp_net_rx_ring_bufs_alloc(dp, &rings[r], xdp))
+ if (nfp_net_rx_ring_bufs_alloc(dp, &rings[r]))
goto err_free_ring;
}
@@ -1919,7 +1914,7 @@ nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
err_free_prev:
while (r--) {
- nfp_net_rx_ring_bufs_free(dp, &rings[r], xdp);
+ nfp_net_rx_ring_bufs_free(dp, &rings[r]);
err_free_ring:
nfp_net_rx_ring_free(&rings[r]);
}
@@ -1949,14 +1944,13 @@ nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_dp *dp,
}
static void
-nfp_net_rx_ring_set_free(struct nfp_net_dp *dp, struct nfp_net_ring_set *s,
- bool xdp)
+nfp_net_rx_ring_set_free(struct nfp_net_dp *dp, struct nfp_net_ring_set *s)
{
struct nfp_net_rx_ring *rings = s->rings;
unsigned int r;
for (r = 0; r < s->n_rings; r++) {
- nfp_net_rx_ring_bufs_free(dp, &rings[r], xdp);
+ nfp_net_rx_ring_bufs_free(dp, &rings[r]);
nfp_net_rx_ring_free(&rings[r]);
}
@@ -2292,8 +2286,7 @@ static int nfp_net_netdev_open(struct net_device *netdev)
goto err_cleanup_vec_p;
}
- nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &nn->dp, &rx,
- nn->dp.xdp_prog);
+ nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &nn->dp, &rx);
if (!nn->dp.rx_rings) {
err = -ENOMEM;
goto err_cleanup_vec;
@@ -2340,7 +2333,7 @@ static int nfp_net_netdev_open(struct net_device *netdev)
err_free_rings:
nfp_net_tx_ring_set_free(&tx);
err_free_rx_rings:
- nfp_net_rx_ring_set_free(&nn->dp, &rx, nn->dp.xdp_prog);
+ nfp_net_rx_ring_set_free(&nn->dp, &rx);
err_cleanup_vec:
r = nn->dp.num_r_vecs;
err_cleanup_vec_p:
@@ -2381,8 +2374,7 @@ static void nfp_net_close_free_all(struct nfp_net *nn)
unsigned int r;
for (r = 0; r < nn->dp.num_rx_rings; r++) {
- nfp_net_rx_ring_bufs_free(&nn->dp, &nn->dp.rx_rings[r],
- nn->dp.xdp_prog);
+ nfp_net_rx_ring_bufs_free(&nn->dp, &nn->dp.rx_rings[r]);
nfp_net_rx_ring_free(&nn->dp.rx_rings[r]);
}
for (r = 0; r < nn->dp.num_tx_rings; r++)
@@ -2472,7 +2464,6 @@ static void nfp_net_dp_swap(struct nfp_net *nn, struct nfp_net_dp *dp)
static int
nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
struct nfp_net_ring_set *tx)
{
@@ -2486,7 +2477,7 @@ nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
swap(dp->num_r_vecs, nn->dp.num_r_vecs);
swap(dp->num_stack_tx_rings, nn->dp.num_stack_tx_rings);
- *xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
+ dp->xdp_prog = xchg(&nn->dp.xdp_prog, dp->xdp_prog);
for (r = 0; r < nn->max_r_vecs; r++)
nfp_net_vector_assign_rings(&nn->dp, &nn->r_vecs[r], r);
@@ -2530,11 +2521,10 @@ struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn)
static int
nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct bpf_prog *xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
{
/* XDP-enabled tests */
- if (!xdp_prog)
+ if (!dp->xdp_prog)
return 0;
if (dp->fl_bufsz > PAGE_SIZE) {
nn_warn(nn, "MTU too large w/ XDP enabled\n");
@@ -2550,7 +2540,6 @@ nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
static void
nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx,
struct nfp_net_ring_set *tx)
{
@@ -2560,7 +2549,6 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
nn->dp.num_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
- *xdp_prog = xchg(&nn->dp.xdp_prog, *xdp_prog);
if (!netif_is_rxfh_configured(nn->dp.netdev))
nfp_net_rss_init_itbl(nn);
@@ -2568,7 +2556,6 @@ nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
int
nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct bpf_prog **xdp_prog,
struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
{
int r, err;
@@ -2576,18 +2563,18 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
dp->fl_bufsz = nfp_net_calc_fl_bufsz(dp);
dp->num_stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
- if (*xdp_prog)
+ if (dp->xdp_prog)
dp->num_stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
dp->num_r_vecs = max(rx ? rx->n_rings : dp->num_rx_rings,
dp->num_stack_tx_rings);
- err = nfp_net_check_config(nn, dp, *xdp_prog, rx, tx);
+ err = nfp_net_check_config(nn, dp, rx, tx);
if (err)
goto exit_free_dp;
if (!netif_running(dp->netdev)) {
- nfp_net_ring_reconfig_down(nn, dp, xdp_prog, rx, tx);
+ nfp_net_ring_reconfig_down(nn, dp, rx, tx);
err = 0;
goto exit_free_dp;
@@ -2602,7 +2589,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
}
}
if (rx) {
- if (!nfp_net_rx_ring_set_prepare(nn, dp, rx, *xdp_prog)) {
+ if (!nfp_net_rx_ring_set_prepare(nn, dp, rx)) {
err = -ENOMEM;
goto err_cleanup_vecs;
}
@@ -2618,14 +2605,14 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
nfp_net_close_stack(nn);
nfp_net_clear_config_and_disable(nn);
- err = nfp_net_ring_swap_enable(nn, dp, xdp_prog, rx, tx);
+ err = nfp_net_ring_swap_enable(nn, dp, rx, tx);
if (err) {
int err2;
nfp_net_clear_config_and_disable(nn);
/* Try with old configuration and old rings */
- err2 = nfp_net_ring_swap_enable(nn, dp, xdp_prog, rx, tx);
+ err2 = nfp_net_ring_swap_enable(nn, dp, rx, tx);
if (err2)
nn_err(nn, "Can't restore ring config - FW communication failed (%d,%d)\n",
err, err2);
@@ -2634,7 +2621,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
if (rx)
- nfp_net_rx_ring_set_free(dp, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(dp, rx);
if (tx)
nfp_net_tx_ring_set_free(tx);
@@ -2646,7 +2633,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
err_free_rx:
if (rx)
- nfp_net_rx_ring_set_free(dp, rx, *xdp_prog);
+ nfp_net_rx_ring_set_free(dp, rx);
err_cleanup_vecs:
for (r = dp->num_r_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
@@ -2669,7 +2656,7 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
dp->mtu = new_mtu;
- return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog, &rx, NULL);
+ return nfp_net_ring_reconfig(nn, dp, &rx, NULL);
}
static void nfp_net_stat64(struct net_device *netdev,
@@ -2987,6 +2974,7 @@ static int nfp_net_xdp_offload(struct nfp_net *nn, struct bpf_prog *prog)
static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
{
+ struct bpf_prog *old_prog = nn->dp.xdp_prog;
struct nfp_net_ring_set rx = {
.n_rings = nn->dp.num_rx_rings,
.dcnt = nn->dp.rxd_cnt,
@@ -3015,16 +3003,16 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
if (!dp)
return -ENOMEM;
+ dp->xdp_prog = prog;
tx.n_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
- err = nfp_net_ring_reconfig(nn, dp, &prog, &rx, &tx);
+ err = nfp_net_ring_reconfig(nn, dp, &rx, &tx);
if (err)
return err;
- /* @prog got swapped and is now the old one */
- if (prog)
- bpf_prog_put(prog);
+ if (old_prog)
+ bpf_prog_put(old_prog);
nfp_net_xdp_offload(nn, nn->dp.xdp_prog);
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index eccb01f3659f..63c1d9ab2335 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -206,8 +206,7 @@ static int nfp_net_set_ring_size(struct nfp_net *nn, u32 rxd_cnt, u32 txd_cnt)
if (!dp)
return -ENOMEM;
- return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog,
- reconfig_rx, reconfig_tx);
+ return nfp_net_ring_reconfig(nn, dp, reconfig_rx, reconfig_tx);
}
static int nfp_net_set_ringparam(struct net_device *netdev,
@@ -791,8 +790,7 @@ static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
if (!dp)
return -ENOMEM;
- return nfp_net_ring_reconfig(nn, dp, &nn->dp.xdp_prog,
- reconfig_rx, reconfig_tx);
+ return nfp_net_ring_reconfig(nn, dp, reconfig_rx, reconfig_tx);
}
static int nfp_net_set_channels(struct net_device *netdev,
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 08/13] nfp: switch to using data path structures for reconfiguration
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (6 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 07/13] nfp: use dp to carry xdp_prog " Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 09/13] nfp: store dma direction in data path structure Jakub Kicinski
` (5 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Instead of passing around sets of rings and their parameters just
store all information in the data path structure.
We will no longer user xchg() on XDP programs when we swap programs
while the traffic is guaranteed not to be flowing. This allows us
to simply assign the entire data path structures instead of copying
field by field.
The optimization to reallocate only the rings on the side (RX/TX)
which has been changed is also removed since it seems like it's not
worth the code complexity.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 10 +-
.../net/ethernet/netronome/nfp/nfp_net_common.c | 260 ++++++---------------
.../net/ethernet/netronome/nfp/nfp_net_ethtool.c | 46 +---
3 files changed, 89 insertions(+), 227 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 19dacc3f1269..5a92f6e41dae 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -611,12 +611,6 @@ struct nfp_net {
struct nfp_eth_table_port *eth_port;
};
-struct nfp_net_ring_set {
- unsigned int n_rings;
- unsigned int dcnt;
- void *rings;
-};
-
/* Functions to read/write from/to a BAR
* Performs any endian conversion necessary.
*/
@@ -813,9 +807,7 @@ nfp_net_irqs_assign(struct nfp_net *nn, struct msix_entry *irq_entries,
unsigned int n);
struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn);
-int
-nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *new,
- struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx);
+int nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *new);
#ifdef CONFIG_NFP_DEBUG
void nfp_net_debugfs_create(void);
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 6ab824a48d1d..a9359da64f80 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1740,20 +1740,20 @@ static void nfp_net_tx_ring_free(struct nfp_net_tx_ring *tx_ring)
/**
* nfp_net_tx_ring_alloc() - Allocate resource for a TX ring
+ * @dp: NFP Net data path struct
* @tx_ring: TX Ring structure to allocate
- * @cnt: Ring buffer count
* @is_xdp: True if ring will be used for XDP
*
* Return: 0 on success, negative errno otherwise.
*/
static int
-nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
+nfp_net_tx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring,
+ bool is_xdp)
{
struct nfp_net_r_vector *r_vec = tx_ring->r_vec;
- struct nfp_net_dp *dp = &r_vec->nfp_net->dp;
int sz;
- tx_ring->cnt = cnt;
+ tx_ring->cnt = dp->txd_cnt;
tx_ring->size = sizeof(*tx_ring->txds) * tx_ring->cnt;
tx_ring->txds = dma_zalloc_coherent(dp->dev, tx_ring->size,
@@ -1777,61 +1777,45 @@ nfp_net_tx_ring_alloc(struct nfp_net_tx_ring *tx_ring, u32 cnt, bool is_xdp)
return -ENOMEM;
}
-static struct nfp_net_tx_ring *
-nfp_net_tx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *s)
+static int nfp_net_tx_rings_prepare(struct nfp_net *nn, struct nfp_net_dp *dp)
{
- struct nfp_net_tx_ring *rings;
unsigned int r;
- rings = kcalloc(s->n_rings, sizeof(*rings), GFP_KERNEL);
- if (!rings)
- return NULL;
+ dp->tx_rings = kcalloc(dp->num_tx_rings, sizeof(*dp->tx_rings),
+ GFP_KERNEL);
+ if (!dp->tx_rings)
+ return -ENOMEM;
- for (r = 0; r < s->n_rings; r++) {
+ for (r = 0; r < dp->num_tx_rings; r++) {
int bias = 0;
if (r >= dp->num_stack_tx_rings)
bias = dp->num_stack_tx_rings;
- nfp_net_tx_ring_init(&rings[r], &nn->r_vecs[r - bias], r);
+ nfp_net_tx_ring_init(&dp->tx_rings[r], &nn->r_vecs[r - bias],
+ r);
- if (nfp_net_tx_ring_alloc(&rings[r], s->dcnt, bias))
+ if (nfp_net_tx_ring_alloc(dp, &dp->tx_rings[r], bias))
goto err_free_prev;
}
- return s->rings = rings;
+ return 0;
err_free_prev:
while (r--)
- nfp_net_tx_ring_free(&rings[r]);
- kfree(rings);
- return NULL;
-}
-
-static void
-nfp_net_tx_ring_set_swap(struct nfp_net *nn, struct nfp_net_ring_set *s)
-{
- struct nfp_net_ring_set new = *s;
-
- s->dcnt = nn->dp.txd_cnt;
- s->rings = nn->dp.tx_rings;
- s->n_rings = nn->dp.num_tx_rings;
-
- nn->dp.txd_cnt = new.dcnt;
- nn->dp.tx_rings = new.rings;
- nn->dp.num_tx_rings = new.n_rings;
+ nfp_net_tx_ring_free(&dp->tx_rings[r]);
+ kfree(dp->tx_rings);
+ return -ENOMEM;
}
-static void nfp_net_tx_ring_set_free(struct nfp_net_ring_set *s)
+static void nfp_net_tx_rings_free(struct nfp_net_dp *dp)
{
- struct nfp_net_tx_ring *rings = s->rings;
unsigned int r;
- for (r = 0; r < s->n_rings; r++)
- nfp_net_tx_ring_free(&rings[r]);
+ for (r = 0; r < dp->num_tx_rings; r++)
+ nfp_net_tx_ring_free(&dp->tx_rings[r]);
- kfree(rings);
+ kfree(dp->tx_rings);
}
/**
@@ -1860,17 +1844,15 @@ static void nfp_net_rx_ring_free(struct nfp_net_rx_ring *rx_ring)
* nfp_net_rx_ring_alloc() - Allocate resource for a RX ring
* @dp: NFP Net data path struct
* @rx_ring: RX ring to allocate
- * @cnt: Ring buffer count
*
* Return: 0 on success, negative errno otherwise.
*/
static int
-nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
- u32 cnt)
+nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring)
{
int sz;
- rx_ring->cnt = cnt;
+ rx_ring->cnt = dp->rxd_cnt;
rx_ring->size = sizeof(*rx_ring->rxds) * rx_ring->cnt;
rx_ring->rxds = dma_zalloc_coherent(dp->dev, rx_ring->size,
&rx_ring->dma, GFP_KERNEL);
@@ -1889,72 +1871,47 @@ nfp_net_rx_ring_alloc(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
return -ENOMEM;
}
-static struct nfp_net_rx_ring *
-nfp_net_rx_ring_set_prepare(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *s)
+static int nfp_net_rx_rings_prepare(struct nfp_net *nn, struct nfp_net_dp *dp)
{
- struct nfp_net_rx_ring *rings;
unsigned int r;
- rings = kcalloc(s->n_rings, sizeof(*rings), GFP_KERNEL);
- if (!rings)
- return NULL;
+ dp->rx_rings = kcalloc(dp->num_rx_rings, sizeof(*dp->rx_rings),
+ GFP_KERNEL);
+ if (!dp->rx_rings)
+ return -ENOMEM;
- for (r = 0; r < s->n_rings; r++) {
- nfp_net_rx_ring_init(&rings[r], &nn->r_vecs[r], r);
+ for (r = 0; r < dp->num_rx_rings; r++) {
+ nfp_net_rx_ring_init(&dp->rx_rings[r], &nn->r_vecs[r], r);
- if (nfp_net_rx_ring_alloc(dp, &rings[r], s->dcnt))
+ if (nfp_net_rx_ring_alloc(dp, &dp->rx_rings[r]))
goto err_free_prev;
- if (nfp_net_rx_ring_bufs_alloc(dp, &rings[r]))
+ if (nfp_net_rx_ring_bufs_alloc(dp, &dp->rx_rings[r]))
goto err_free_ring;
}
- return s->rings = rings;
+ return 0;
err_free_prev:
while (r--) {
- nfp_net_rx_ring_bufs_free(dp, &rings[r]);
+ nfp_net_rx_ring_bufs_free(dp, &dp->rx_rings[r]);
err_free_ring:
- nfp_net_rx_ring_free(&rings[r]);
+ nfp_net_rx_ring_free(&dp->rx_rings[r]);
}
- kfree(rings);
- return NULL;
-}
-
-static void
-nfp_net_rx_ring_set_swap(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *s)
-{
- struct nfp_net_ring_set new = *s;
- struct nfp_net_dp new_dp = *dp;
-
- dp->fl_bufsz = nn->dp.fl_bufsz;
- dp->mtu = nn->dp.netdev->mtu;
- s->dcnt = nn->dp.rxd_cnt;
- s->rings = nn->dp.rx_rings;
- s->n_rings = nn->dp.num_rx_rings;
-
- nn->dp.mtu = new_dp.mtu;
- nn->dp.netdev->mtu = new_dp.mtu;
- nn->dp.fl_bufsz = new_dp.fl_bufsz;
- nn->dp.rxd_cnt = new.dcnt;
- nn->dp.rx_rings = new.rings;
- nn->dp.num_rx_rings = new.n_rings;
+ kfree(dp->rx_rings);
+ return -ENOMEM;
}
-static void
-nfp_net_rx_ring_set_free(struct nfp_net_dp *dp, struct nfp_net_ring_set *s)
+static void nfp_net_rx_rings_free(struct nfp_net_dp *dp)
{
- struct nfp_net_rx_ring *rings = s->rings;
unsigned int r;
- for (r = 0; r < s->n_rings; r++) {
- nfp_net_rx_ring_bufs_free(dp, &rings[r]);
- nfp_net_rx_ring_free(&rings[r]);
+ for (r = 0; r < dp->num_rx_rings; r++) {
+ nfp_net_rx_ring_bufs_free(dp, &dp->rx_rings[r]);
+ nfp_net_rx_ring_free(&dp->rx_rings[r]);
}
- kfree(rings);
+ kfree(dp->rx_rings);
}
static void
@@ -2248,14 +2205,6 @@ static void nfp_net_open_stack(struct nfp_net *nn)
static int nfp_net_netdev_open(struct net_device *netdev)
{
struct nfp_net *nn = netdev_priv(netdev);
- struct nfp_net_ring_set rx = {
- .n_rings = nn->dp.num_rx_rings,
- .dcnt = nn->dp.rxd_cnt,
- };
- struct nfp_net_ring_set tx = {
- .n_rings = nn->dp.num_tx_rings,
- .dcnt = nn->dp.txd_cnt,
- };
int err, r;
if (nn->dp.ctrl & NFP_NET_CFG_CTRL_ENABLE) {
@@ -2286,17 +2235,13 @@ static int nfp_net_netdev_open(struct net_device *netdev)
goto err_cleanup_vec_p;
}
- nn->dp.rx_rings = nfp_net_rx_ring_set_prepare(nn, &nn->dp, &rx);
- if (!nn->dp.rx_rings) {
- err = -ENOMEM;
+ err = nfp_net_rx_rings_prepare(nn, &nn->dp);
+ if (err)
goto err_cleanup_vec;
- }
- nn->dp.tx_rings = nfp_net_tx_ring_set_prepare(nn, &nn->dp, &tx);
- if (!nn->dp.tx_rings) {
- err = -ENOMEM;
+ err = nfp_net_tx_rings_prepare(nn, &nn->dp);
+ if (err)
goto err_free_rx_rings;
- }
for (r = 0; r < nn->max_r_vecs; r++)
nfp_net_vector_assign_rings(&nn->dp, &nn->r_vecs[r], r);
@@ -2331,9 +2276,9 @@ static int nfp_net_netdev_open(struct net_device *netdev)
return 0;
err_free_rings:
- nfp_net_tx_ring_set_free(&tx);
+ nfp_net_tx_rings_free(&nn->dp);
err_free_rx_rings:
- nfp_net_rx_ring_set_free(&nn->dp, &rx);
+ nfp_net_rx_rings_free(&nn->dp);
err_cleanup_vec:
r = nn->dp.num_r_vecs;
err_cleanup_vec_p:
@@ -2460,31 +2405,21 @@ static void nfp_net_dp_swap(struct nfp_net *nn, struct nfp_net_dp *dp)
nn->dp = new_dp;
nn->dp.netdev->mtu = new_dp.mtu;
+
+ if (!netif_is_rxfh_configured(nn->dp.netdev))
+ nfp_net_rss_init_itbl(nn);
}
-static int
-nfp_net_ring_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *rx,
- struct nfp_net_ring_set *tx)
+static int nfp_net_dp_swap_enable(struct nfp_net *nn, struct nfp_net_dp *dp)
{
unsigned int r;
int err;
- if (rx)
- nfp_net_rx_ring_set_swap(nn, dp, rx);
- if (tx)
- nfp_net_tx_ring_set_swap(nn, tx);
-
- swap(dp->num_r_vecs, nn->dp.num_r_vecs);
- swap(dp->num_stack_tx_rings, nn->dp.num_stack_tx_rings);
- dp->xdp_prog = xchg(&nn->dp.xdp_prog, dp->xdp_prog);
+ nfp_net_dp_swap(nn, dp);
for (r = 0; r < nn->max_r_vecs; r++)
nfp_net_vector_assign_rings(&nn->dp, &nn->r_vecs[r], r);
- if (!netif_is_rxfh_configured(nn->dp.netdev))
- nfp_net_rss_init_itbl(nn);
-
err = netif_set_real_num_rx_queues(nn->dp.netdev, nn->dp.num_rx_rings);
if (err)
return err;
@@ -2519,9 +2454,7 @@ struct nfp_net_dp *nfp_net_clone_dp(struct nfp_net *nn)
return new;
}
-static int
-nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
+static int nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp)
{
/* XDP-enabled tests */
if (!dp->xdp_prog)
@@ -2530,7 +2463,7 @@ nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
nn_warn(nn, "MTU too large w/ XDP enabled\n");
return -EINVAL;
}
- if (tx && tx->n_rings > nn->max_tx_rings) {
+ if (dp->num_tx_rings > nn->max_tx_rings) {
nn_warn(nn, "Insufficient number of TX rings w/ XDP enabled\n");
return -EINVAL;
}
@@ -2538,44 +2471,24 @@ nfp_net_check_config(struct nfp_net *nn, struct nfp_net_dp *dp,
return 0;
}
-static void
-nfp_net_ring_reconfig_down(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *rx,
- struct nfp_net_ring_set *tx)
-{
- nfp_net_dp_swap(nn, dp);
-
- nn->dp.rxd_cnt = rx ? rx->dcnt : nn->dp.rxd_cnt;
- nn->dp.txd_cnt = tx ? tx->dcnt : nn->dp.txd_cnt;
- nn->dp.num_rx_rings = rx ? rx->n_rings : nn->dp.num_rx_rings;
- nn->dp.num_tx_rings = tx ? tx->n_rings : nn->dp.num_tx_rings;
-
- if (!netif_is_rxfh_configured(nn->dp.netdev))
- nfp_net_rss_init_itbl(nn);
-}
-
-int
-nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
- struct nfp_net_ring_set *rx, struct nfp_net_ring_set *tx)
+int nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp)
{
int r, err;
dp->fl_bufsz = nfp_net_calc_fl_bufsz(dp);
- dp->num_stack_tx_rings = tx ? tx->n_rings : dp->num_tx_rings;
+ dp->num_stack_tx_rings = dp->num_tx_rings;
if (dp->xdp_prog)
- dp->num_stack_tx_rings -= rx ? rx->n_rings : dp->num_rx_rings;
+ dp->num_stack_tx_rings -= dp->num_rx_rings;
- dp->num_r_vecs = max(rx ? rx->n_rings : dp->num_rx_rings,
- dp->num_stack_tx_rings);
+ dp->num_r_vecs = max(dp->num_rx_rings, dp->num_stack_tx_rings);
- err = nfp_net_check_config(nn, dp, rx, tx);
+ err = nfp_net_check_config(nn, dp);
if (err)
goto exit_free_dp;
if (!netif_running(dp->netdev)) {
- nfp_net_ring_reconfig_down(nn, dp, rx, tx);
-
+ nfp_net_dp_swap(nn, dp);
err = 0;
goto exit_free_dp;
}
@@ -2588,31 +2501,27 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
goto err_cleanup_vecs;
}
}
- if (rx) {
- if (!nfp_net_rx_ring_set_prepare(nn, dp, rx)) {
- err = -ENOMEM;
- goto err_cleanup_vecs;
- }
- }
- if (tx) {
- if (!nfp_net_tx_ring_set_prepare(nn, dp, tx)) {
- err = -ENOMEM;
- goto err_free_rx;
- }
- }
+
+ err = nfp_net_rx_rings_prepare(nn, dp);
+ if (err)
+ goto err_cleanup_vecs;
+
+ err = nfp_net_tx_rings_prepare(nn, dp);
+ if (err)
+ goto err_free_rx;
/* Stop device, swap in new rings, try to start the firmware */
nfp_net_close_stack(nn);
nfp_net_clear_config_and_disable(nn);
- err = nfp_net_ring_swap_enable(nn, dp, rx, tx);
+ err = nfp_net_dp_swap_enable(nn, dp);
if (err) {
int err2;
nfp_net_clear_config_and_disable(nn);
/* Try with old configuration and old rings */
- err2 = nfp_net_ring_swap_enable(nn, dp, rx, tx);
+ err2 = nfp_net_dp_swap_enable(nn, dp);
if (err2)
nn_err(nn, "Can't restore ring config - FW communication failed (%d,%d)\n",
err, err2);
@@ -2620,10 +2529,8 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
for (r = dp->num_r_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
- if (rx)
- nfp_net_rx_ring_set_free(dp, rx);
- if (tx)
- nfp_net_tx_ring_set_free(tx);
+ nfp_net_rx_rings_free(dp);
+ nfp_net_tx_rings_free(dp);
nfp_net_open_stack(nn);
exit_free_dp:
@@ -2632,8 +2539,7 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
return err;
err_free_rx:
- if (rx)
- nfp_net_rx_ring_set_free(dp, rx);
+ nfp_net_rx_rings_free(dp);
err_cleanup_vecs:
for (r = dp->num_r_vecs - 1; r >= nn->dp.num_r_vecs; r--)
nfp_net_cleanup_vector(nn, &nn->r_vecs[r]);
@@ -2644,10 +2550,6 @@ nfp_net_ring_reconfig(struct nfp_net *nn, struct nfp_net_dp *dp,
static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
{
struct nfp_net *nn = netdev_priv(netdev);
- struct nfp_net_ring_set rx = {
- .n_rings = nn->dp.num_rx_rings,
- .dcnt = nn->dp.rxd_cnt,
- };
struct nfp_net_dp *dp;
dp = nfp_net_clone_dp(nn);
@@ -2656,7 +2558,7 @@ static int nfp_net_change_mtu(struct net_device *netdev, int new_mtu)
dp->mtu = new_mtu;
- return nfp_net_ring_reconfig(nn, dp, &rx, NULL);
+ return nfp_net_ring_reconfig(nn, dp);
}
static void nfp_net_stat64(struct net_device *netdev,
@@ -2975,14 +2877,6 @@ static int nfp_net_xdp_offload(struct nfp_net *nn, struct bpf_prog *prog)
static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
{
struct bpf_prog *old_prog = nn->dp.xdp_prog;
- struct nfp_net_ring_set rx = {
- .n_rings = nn->dp.num_rx_rings,
- .dcnt = nn->dp.rxd_cnt,
- };
- struct nfp_net_ring_set tx = {
- .n_rings = nn->dp.num_tx_rings,
- .dcnt = nn->dp.txd_cnt,
- };
struct nfp_net_dp *dp;
int err;
@@ -3004,10 +2898,10 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
return -ENOMEM;
dp->xdp_prog = prog;
- tx.n_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
+ dp->num_tx_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
- err = nfp_net_ring_reconfig(nn, dp, &rx, &tx);
+ err = nfp_net_ring_reconfig(nn, dp);
if (err)
return err;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
index 63c1d9ab2335..ed22a813e579 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c
@@ -186,27 +186,16 @@ static void nfp_net_get_ringparam(struct net_device *netdev,
static int nfp_net_set_ring_size(struct nfp_net *nn, u32 rxd_cnt, u32 txd_cnt)
{
- struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
- struct nfp_net_ring_set rx = {
- .n_rings = nn->dp.num_rx_rings,
- .dcnt = rxd_cnt,
- };
- struct nfp_net_ring_set tx = {
- .n_rings = nn->dp.num_tx_rings,
- .dcnt = txd_cnt,
- };
struct nfp_net_dp *dp;
- if (nn->dp.rxd_cnt != rxd_cnt)
- reconfig_rx = ℞
- if (nn->dp.txd_cnt != txd_cnt)
- reconfig_tx = &tx;
-
dp = nfp_net_clone_dp(nn);
if (!dp)
return -ENOMEM;
- return nfp_net_ring_reconfig(nn, dp, reconfig_rx, reconfig_tx);
+ dp->rxd_cnt = rxd_cnt;
+ dp->txd_cnt = txd_cnt;
+
+ return nfp_net_ring_reconfig(nn, dp);
}
static int nfp_net_set_ringparam(struct net_device *netdev,
@@ -765,32 +754,19 @@ static void nfp_net_get_channels(struct net_device *netdev,
static int nfp_net_set_num_rings(struct nfp_net *nn, unsigned int total_rx,
unsigned int total_tx)
{
- struct nfp_net_ring_set *reconfig_rx = NULL, *reconfig_tx = NULL;
- struct nfp_net_ring_set rx = {
- .n_rings = total_rx,
- .dcnt = nn->dp.rxd_cnt,
- };
- struct nfp_net_ring_set tx = {
- .n_rings = total_tx,
- .dcnt = nn->dp.txd_cnt,
- };
struct nfp_net_dp *dp;
- if (nn->dp.num_rx_rings != total_rx)
- reconfig_rx = ℞
- if (nn->dp.num_stack_tx_rings != total_tx ||
- (nn->dp.xdp_prog && reconfig_rx))
- reconfig_tx = &tx;
-
- /* nfp_net_check_config() will catch tx.n_rings > nn->max_tx_rings */
- if (nn->dp.xdp_prog)
- tx.n_rings += total_rx;
-
dp = nfp_net_clone_dp(nn);
if (!dp)
return -ENOMEM;
- return nfp_net_ring_reconfig(nn, dp, reconfig_rx, reconfig_tx);
+ dp->num_rx_rings = total_rx;
+ dp->num_tx_rings = total_tx;
+ /* nfp_net_check_config() will catch num_tx_rings > nn->max_tx_rings */
+ if (dp->xdp_prog)
+ dp->num_tx_rings += total_rx;
+
+ return nfp_net_ring_reconfig(nn, dp);
}
static int nfp_net_set_channels(struct net_device *netdev,
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 09/13] nfp: store dma direction in data path structure
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (7 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 08/13] nfp: switch to using data path structures for reconfiguration Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 10/13] nfp: validate rx offset from the BAR and size down it's field Jakub Kicinski
` (4 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Instead of testing if xdp_prog is present store the dma direction
in data path structure.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 11 ++++--
.../net/ethernet/netronome/nfp/nfp_net_common.c | 45 ++++++++--------------
2 files changed, 24 insertions(+), 32 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 5a92f6e41dae..db92463da440 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -438,6 +438,7 @@ struct nfp_stat_pair {
* @bpf_offload_skip_sw: Offloaded BPF program will not be rerun by cls_bpf
* @bpf_offload_xdp: Offloaded BPF program is XDP
* @chained_metadata_format: Firemware will use new metadata format
+ * @rx_dma_dir: Mapping direction for RX buffers
* @ctrl: Local copy of the control register/word.
* @fl_bufsz: Currently configured size of the freelist buffers
* @rx_offset: Offset in the RX buffers where packet data starts
@@ -458,10 +459,12 @@ struct nfp_net_dp {
struct device *dev;
struct net_device *netdev;
- unsigned is_vf:1;
- unsigned bpf_offload_skip_sw:1;
- unsigned bpf_offload_xdp:1;
- unsigned chained_metadata_format:1;
+ u8 is_vf:1;
+ u8 bpf_offload_skip_sw:1;
+ u8 bpf_offload_xdp:1;
+ u8 chained_metadata_format:1;
+
+ u8 rx_dma_dir;
u32 ctrl;
u32 fl_bufsz;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index a9359da64f80..ab03f2f301cd 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -85,20 +85,18 @@ void nfp_net_get_fw_version(struct nfp_net_fw_version *fw_ver,
put_unaligned_le32(reg, fw_ver);
}
-static dma_addr_t
-nfp_net_dma_map_rx(struct nfp_net_dp *dp, void *frag, int direction)
+static dma_addr_t nfp_net_dma_map_rx(struct nfp_net_dp *dp, void *frag)
{
return dma_map_single(dp->dev, frag + NFP_NET_RX_BUF_HEADROOM,
dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA,
- direction);
+ dp->rx_dma_dir);
}
-static void
-nfp_net_dma_unmap_rx(struct nfp_net_dp *dp, dma_addr_t dma_addr,
- int direction)
+static void nfp_net_dma_unmap_rx(struct nfp_net_dp *dp, dma_addr_t dma_addr)
{
dma_unmap_single(dp->dev, dma_addr,
- dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA, direction);
+ dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA,
+ dp->rx_dma_dir);
}
/* Firmware reconfig
@@ -991,8 +989,7 @@ static void nfp_net_xdp_complete(struct nfp_net_tx_ring *tx_ring)
if (!tx_ring->txbufs[idx].frag)
continue;
- nfp_net_dma_unmap_rx(dp, tx_ring->txbufs[idx].dma_addr,
- DMA_BIDIRECTIONAL);
+ nfp_net_dma_unmap_rx(dp, tx_ring->txbufs[idx].dma_addr);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
done_pkts++;
@@ -1037,8 +1034,7 @@ nfp_net_tx_ring_reset(struct nfp_net_dp *dp, struct nfp_net_tx_ring *tx_ring)
tx_buf = &tx_ring->txbufs[idx];
if (tx_ring == r_vec->xdp_ring) {
- nfp_net_dma_unmap_rx(dp, tx_buf->dma_addr,
- DMA_BIDIRECTIONAL);
+ nfp_net_dma_unmap_rx(dp, tx_buf->dma_addr);
__free_page(virt_to_page(tx_ring->txbufs[idx].frag));
} else {
struct sk_buff *skb = tx_ring->txbufs[idx].skb;
@@ -1139,7 +1135,6 @@ static void *
nfp_net_rx_alloc_one(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
dma_addr_t *dma_addr)
{
- int direction;
void *frag;
if (!dp->xdp_prog)
@@ -1151,9 +1146,7 @@ nfp_net_rx_alloc_one(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
return NULL;
}
- direction = dp->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
-
- *dma_addr = nfp_net_dma_map_rx(dp, frag, direction);
+ *dma_addr = nfp_net_dma_map_rx(dp, frag);
if (dma_mapping_error(dp->dev, *dma_addr)) {
nfp_net_free_frag(frag, dp->xdp_prog);
nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
@@ -1163,9 +1156,7 @@ nfp_net_rx_alloc_one(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
return frag;
}
-static void *
-nfp_net_napi_alloc_one(struct nfp_net_dp *dp, int direction,
- dma_addr_t *dma_addr)
+static void *nfp_net_napi_alloc_one(struct nfp_net_dp *dp, dma_addr_t *dma_addr)
{
void *frag;
@@ -1178,7 +1169,7 @@ nfp_net_napi_alloc_one(struct nfp_net_dp *dp, int direction,
return NULL;
}
- *dma_addr = nfp_net_dma_map_rx(dp, frag, direction);
+ *dma_addr = nfp_net_dma_map_rx(dp, frag);
if (dma_mapping_error(dp->dev, *dma_addr)) {
nfp_net_free_frag(frag, dp->xdp_prog);
nn_dp_warn(dp, "Failed to map DMA RX buffer\n");
@@ -1260,7 +1251,6 @@ static void
nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
struct nfp_net_rx_ring *rx_ring)
{
- int direction = dp->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
unsigned int i;
for (i = 0; i < rx_ring->cnt - 1; i++) {
@@ -1271,8 +1261,7 @@ nfp_net_rx_ring_bufs_free(struct nfp_net_dp *dp,
if (!rx_ring->rxbufs[i].frag)
continue;
- nfp_net_dma_unmap_rx(dp, rx_ring->rxbufs[i].dma_addr,
- direction);
+ nfp_net_dma_unmap_rx(dp, rx_ring->rxbufs[i].dma_addr);
nfp_net_free_frag(rx_ring->rxbufs[i].frag, dp->xdp_prog);
rx_ring->rxbufs[i].dma_addr = 0;
rx_ring->rxbufs[i].frag = NULL;
@@ -1478,7 +1467,7 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
return false;
}
- new_frag = nfp_net_napi_alloc_one(dp, DMA_BIDIRECTIONAL, &new_dma_addr);
+ new_frag = nfp_net_napi_alloc_one(dp, &new_dma_addr);
if (unlikely(!new_frag)) {
nfp_net_rx_drop(rx_ring->r_vec, rx_ring, rxbuf, NULL);
return false;
@@ -1544,12 +1533,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
unsigned int true_bufsz;
struct sk_buff *skb;
int pkts_polled = 0;
- int rx_dma_map_dir;
int idx;
rcu_read_lock();
xdp_prog = READ_ONCE(dp->xdp_prog);
- rx_dma_map_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
true_bufsz = xdp_prog ? PAGE_SIZE : dp->fl_bufsz;
tx_ring = r_vec->xdp_ring;
@@ -1639,14 +1626,13 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
nfp_net_rx_drop(r_vec, rx_ring, rxbuf, NULL);
continue;
}
- new_frag = nfp_net_napi_alloc_one(dp, rx_dma_map_dir,
- &new_dma_addr);
+ new_frag = nfp_net_napi_alloc_one(dp, &new_dma_addr);
if (unlikely(!new_frag)) {
nfp_net_rx_drop(r_vec, rx_ring, rxbuf, skb);
continue;
}
- nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr, rx_dma_map_dir);
+ nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr);
nfp_net_rx_give_one(rx_ring, new_frag, new_dma_addr);
@@ -2899,6 +2885,7 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
dp->xdp_prog = prog;
dp->num_tx_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
+ dp->rx_dma_dir = prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
err = nfp_net_ring_reconfig(nn, dp);
@@ -3128,6 +3115,8 @@ int nfp_net_netdev_init(struct net_device *netdev)
nn->dp.chained_metadata_format = nn->fw_ver.major > 3;
+ nn->dp.rx_dma_dir = DMA_FROM_DEVICE;
+
/* Get some of the read-only fields from the BAR */
nn->cap = nn_readl(nn, NFP_NET_CFG_CAP);
nn->max_mtu = nn_readl(nn, NFP_NET_CFG_MAX_MTU);
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 10/13] nfp: validate rx offset from the BAR and size down it's field
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (8 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 09/13] nfp: store dma direction in data path structure Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 11/13] nfp: reorganize pkt_off variable Jakub Kicinski
` (3 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
NFP_NET_CFG_RX_OFFSET is 32bit wide, make sure what we read from
there is reasonable for packet headroom. This allows us to store
the rx_offset in a 8bit variable.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 6 +++---
drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 14 +++++++++++---
2 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index db92463da440..5f0547c6efb8 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -439,9 +439,9 @@ struct nfp_stat_pair {
* @bpf_offload_xdp: Offloaded BPF program is XDP
* @chained_metadata_format: Firemware will use new metadata format
* @rx_dma_dir: Mapping direction for RX buffers
+ * @rx_offset: Offset in the RX buffers where packet data starts
* @ctrl: Local copy of the control register/word.
* @fl_bufsz: Currently configured size of the freelist buffers
- * @rx_offset: Offset in the RX buffers where packet data starts
* @xdp_prog: Installed XDP program
* @tx_rings: Array of pre-allocated TX ring structures
* @rx_rings: Array of pre-allocated RX ring structures
@@ -466,11 +466,11 @@ struct nfp_net_dp {
u8 rx_dma_dir;
+ u8 rx_offset;
+
u32 ctrl;
u32 fl_bufsz;
- u32 rx_offset;
-
struct bpf_prog *xdp_prog;
struct nfp_net_tx_ring *tx_rings;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index ab03f2f301cd..513f55dd746b 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -3124,10 +3124,18 @@ int nfp_net_netdev_init(struct net_device *netdev)
nfp_net_write_mac_addr(nn);
/* Determine RX packet/metadata boundary offset */
- if (nn->fw_ver.major >= 2)
- nn->dp.rx_offset = nn_readl(nn, NFP_NET_CFG_RX_OFFSET);
- else
+ if (nn->fw_ver.major >= 2) {
+ u32 reg;
+
+ reg = nn_readl(nn, NFP_NET_CFG_RX_OFFSET);
+ if (reg > NFP_NET_MAX_PREPEND) {
+ nn_err(nn, "Invalid rx offset: %d\n", reg);
+ return -EINVAL;
+ }
+ nn->dp.rx_offset = reg;
+ } else {
nn->dp.rx_offset = NFP_NET_RX_OFFSET;
+ }
/* Set default MTU and Freelist buffer size */
if (nn->max_mtu < NFP_NET_DEFAULT_MTU)
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 11/13] nfp: reorganize pkt_off variable
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (9 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 10/13] nfp: validate rx offset from the BAR and size down it's field Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 12/13] nfp: prepare metadata handling for xdp_adjust_head() Jakub Kicinski
` (2 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Rename pkt_off variable to dma_off, it should hold data offset
counting from beginning of DMA mapping. Compute the value only
in XDP context.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 513f55dd746b..0e4fa6802733 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1453,7 +1453,7 @@ nfp_net_rx_drop(struct nfp_net_r_vector *r_vec, struct nfp_net_rx_ring *rx_ring,
static bool
nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
struct nfp_net_tx_ring *tx_ring,
- struct nfp_net_rx_buf *rxbuf, unsigned int pkt_off,
+ struct nfp_net_rx_buf *rxbuf, unsigned int dma_off,
unsigned int pkt_len)
{
struct nfp_net_tx_buf *txbuf;
@@ -1484,14 +1484,14 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
txbuf->pkt_cnt = 1;
txbuf->real_len = pkt_len;
- dma_sync_single_for_device(dp->dev, rxbuf->dma_addr + pkt_off,
+ dma_sync_single_for_device(dp->dev, rxbuf->dma_addr + dma_off,
pkt_len, DMA_BIDIRECTIONAL);
/* Build TX descriptor */
txd = &tx_ring->txds[wr_idx];
txd->offset_eop = PCIE_DESC_TX_EOP;
txd->dma_len = cpu_to_le16(pkt_len);
- nfp_desc_set_dma_addr(txd, rxbuf->dma_addr + pkt_off);
+ nfp_desc_set_dma_addr(txd, rxbuf->dma_addr + dma_off);
txd->data_len = cpu_to_le16(pkt_len);
txd->flags = 0;
@@ -1541,7 +1541,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
tx_ring = r_vec->xdp_ring;
while (pkts_polled < budget) {
- unsigned int meta_len, data_len, data_off, pkt_len, pkt_off;
+ unsigned int meta_len, data_len, data_off, pkt_len;
struct nfp_net_rx_buf *rxbuf;
struct nfp_net_rx_desc *rxd;
dma_addr_t new_dma_addr;
@@ -1579,10 +1579,9 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
pkt_len = data_len - meta_len;
if (dp->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
- pkt_off = meta_len;
+ data_off = NFP_NET_RX_BUF_HEADROOM + meta_len;
else
- pkt_off = dp->rx_offset;
- data_off = NFP_NET_RX_BUF_HEADROOM + pkt_off;
+ data_off = NFP_NET_RX_BUF_HEADROOM + dp->rx_offset;
/* Stats update */
u64_stats_update_begin(&r_vec->rx_sync);
@@ -1592,10 +1591,12 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
if (xdp_prog && !(rxd->rxd.flags & PCIE_DESC_RX_BPF &&
dp->bpf_offload_xdp)) {
+ unsigned int dma_off;
int act;
+ dma_off = data_off - NFP_NET_RX_BUF_HEADROOM;
dma_sync_single_for_cpu(dp->dev,
- rxbuf->dma_addr + pkt_off,
+ rxbuf->dma_addr + dma_off,
pkt_len, DMA_BIDIRECTIONAL);
act = nfp_net_run_xdp(xdp_prog, rxbuf->frag + data_off,
pkt_len);
@@ -1605,7 +1606,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
case XDP_TX:
if (unlikely(!nfp_net_tx_xdp_buf(dp, rx_ring,
tx_ring, rxbuf,
- pkt_off,
+ dma_off,
pkt_len)))
trace_xdp_exception(dp->netdev,
xdp_prog, act);
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 12/13] nfp: prepare metadata handling for xdp_adjust_head()
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (10 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 11/13] nfp: reorganize pkt_off variable Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 13/13] nfp: add support " Jakub Kicinski
2017-03-13 6:43 ` [PATCH net-next 00/13] nfp: XDP adjust head support David Miller
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
XDP may require us to move metadata to make room for pushing
headers. Track meta data location with a pointer and pass
it explicitly to functions.
While at it validate that meta_len from the descriptor is not
bogus.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
.../net/ethernet/netronome/nfp/nfp_net_common.c | 28 +++++++++++++++-------
1 file changed, 19 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 0e4fa6802733..fe7c3f6d820d 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1385,24 +1385,21 @@ static void nfp_net_set_hash(struct net_device *netdev, struct sk_buff *skb,
static void
nfp_net_set_hash_desc(struct net_device *netdev, struct sk_buff *skb,
- struct nfp_net_rx_desc *rxd)
+ void *data, struct nfp_net_rx_desc *rxd)
{
- struct nfp_net_rx_hash *rx_hash;
+ struct nfp_net_rx_hash *rx_hash = data;
if (!(rxd->rxd.flags & PCIE_DESC_RX_RSS))
return;
- rx_hash = (struct nfp_net_rx_hash *)(skb->data - sizeof(*rx_hash));
-
nfp_net_set_hash(netdev, skb, get_unaligned_be32(&rx_hash->hash_type),
&rx_hash->hash);
}
static void *
nfp_net_parse_meta(struct net_device *netdev, struct sk_buff *skb,
- int meta_len)
+ void *data, int meta_len)
{
- u8 *data = skb->data - meta_len;
u32 meta_info;
meta_info = get_unaligned_be32(data);
@@ -1546,6 +1543,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
struct nfp_net_rx_desc *rxd;
dma_addr_t new_dma_addr;
void *new_frag;
+ u8 *meta;
idx = rx_ring->rd_p & (rx_ring->cnt - 1);
@@ -1589,6 +1587,17 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
r_vec->rx_bytes += pkt_len;
u64_stats_update_end(&r_vec->rx_sync);
+ /* Pointer to start of metadata */
+ meta = rxbuf->frag + data_off - meta_len;
+
+ if (unlikely(meta_len > NFP_NET_MAX_PREPEND ||
+ (dp->rx_offset && meta_len > dp->rx_offset))) {
+ nn_dp_warn(dp, "oversized RX packet metadata %u\n",
+ meta_len);
+ nfp_net_rx_drop(r_vec, rx_ring, rxbuf, NULL);
+ continue;
+ }
+
if (xdp_prog && !(rxd->rxd.flags & PCIE_DESC_RX_BPF &&
dp->bpf_offload_xdp)) {
unsigned int dma_off;
@@ -1641,12 +1650,13 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
skb_put(skb, pkt_len);
if (!dp->chained_metadata_format) {
- nfp_net_set_hash_desc(dp->netdev, skb, rxd);
+ nfp_net_set_hash_desc(dp->netdev, skb, meta, rxd);
} else if (meta_len) {
void *end;
- end = nfp_net_parse_meta(dp->netdev, skb, meta_len);
- if (unlikely(end != skb->data)) {
+ end = nfp_net_parse_meta(dp->netdev, skb, meta,
+ meta_len);
+ if (unlikely(end != meta + meta_len)) {
nn_dp_warn(dp, "invalid RX packet metadata\n");
nfp_net_rx_drop(r_vec, rx_ring, NULL, skb);
continue;
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH net-next 13/13] nfp: add support for xdp_adjust_head()
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (11 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 12/13] nfp: prepare metadata handling for xdp_adjust_head() Jakub Kicinski
@ 2017-03-10 18:38 ` Jakub Kicinski
2017-03-13 6:43 ` [PATCH net-next 00/13] nfp: XDP adjust head support David Miller
13 siblings, 0 replies; 15+ messages in thread
From: Jakub Kicinski @ 2017-03-10 18:38 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, kubakici, Jakub Kicinski
Support prepending data from XDP. We are already always allocating
some headroom because FW may prepend metadata to packets.
xdp_adjust_head() can be supported by making sure that headroom is
big enough for XDP. In case FW had prepended metadata to the packet,
however, we have to move it out of the way before we call XDP.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
drivers/net/ethernet/netronome/nfp/nfp_net.h | 2 +
.../net/ethernet/netronome/nfp/nfp_net_common.c | 98 +++++++++++++++-------
2 files changed, 70 insertions(+), 30 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h
index 5f0547c6efb8..4d45f4573b57 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h
@@ -439,6 +439,7 @@ struct nfp_stat_pair {
* @bpf_offload_xdp: Offloaded BPF program is XDP
* @chained_metadata_format: Firemware will use new metadata format
* @rx_dma_dir: Mapping direction for RX buffers
+ * @rx_dma_off: Offset at which DMA packets (for XDP headroom)
* @rx_offset: Offset in the RX buffers where packet data starts
* @ctrl: Local copy of the control register/word.
* @fl_bufsz: Currently configured size of the freelist buffers
@@ -465,6 +466,7 @@ struct nfp_net_dp {
u8 chained_metadata_format:1;
u8 rx_dma_dir;
+ u8 rx_dma_off;
u8 rx_offset;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index fe7c3f6d820d..f134f1808b9a 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -1100,6 +1100,7 @@ nfp_net_calc_fl_bufsz(struct nfp_net_dp *dp)
unsigned int fl_bufsz;
fl_bufsz = NFP_NET_RX_BUF_HEADROOM;
+ fl_bufsz += dp->rx_dma_off;
if (dp->rx_offset == NFP_NET_CFG_RX_OFFSET_DYNAMIC)
fl_bufsz += NFP_NET_MAX_PREPEND;
else
@@ -1181,11 +1182,13 @@ static void *nfp_net_napi_alloc_one(struct nfp_net_dp *dp, dma_addr_t *dma_addr)
/**
* nfp_net_rx_give_one() - Put mapped skb on the software and hardware rings
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring structure
* @frag: page fragment buffer
* @dma_addr: DMA address of skb mapping
*/
-static void nfp_net_rx_give_one(struct nfp_net_rx_ring *rx_ring,
+static void nfp_net_rx_give_one(const struct nfp_net_dp *dp,
+ struct nfp_net_rx_ring *rx_ring,
void *frag, dma_addr_t dma_addr)
{
unsigned int wr_idx;
@@ -1199,7 +1202,8 @@ static void nfp_net_rx_give_one(struct nfp_net_rx_ring *rx_ring,
/* Fill freelist descriptor */
rx_ring->rxds[wr_idx].fld.reserved = 0;
rx_ring->rxds[wr_idx].fld.meta_len_dd = 0;
- nfp_desc_set_dma_addr(&rx_ring->rxds[wr_idx].fld, dma_addr);
+ nfp_desc_set_dma_addr(&rx_ring->rxds[wr_idx].fld,
+ dma_addr + dp->rx_dma_off);
rx_ring->wr_p++;
rx_ring->wr_ptr_add++;
@@ -1296,14 +1300,17 @@ nfp_net_rx_ring_bufs_alloc(struct nfp_net_dp *dp,
/**
* nfp_net_rx_ring_fill_freelist() - Give buffers from the ring to FW
+ * @dp: NFP Net data path struct
* @rx_ring: RX ring to fill
*/
-static void nfp_net_rx_ring_fill_freelist(struct nfp_net_rx_ring *rx_ring)
+static void
+nfp_net_rx_ring_fill_freelist(struct nfp_net_dp *dp,
+ struct nfp_net_rx_ring *rx_ring)
{
unsigned int i;
for (i = 0; i < rx_ring->cnt - 1; i++)
- nfp_net_rx_give_one(rx_ring, rx_ring->rxbufs[i].frag,
+ nfp_net_rx_give_one(dp, rx_ring, rx_ring->rxbufs[i].frag,
rx_ring->rxbufs[i].dma_addr);
}
@@ -1429,8 +1436,9 @@ nfp_net_parse_meta(struct net_device *netdev, struct sk_buff *skb,
}
static void
-nfp_net_rx_drop(struct nfp_net_r_vector *r_vec, struct nfp_net_rx_ring *rx_ring,
- struct nfp_net_rx_buf *rxbuf, struct sk_buff *skb)
+nfp_net_rx_drop(const struct nfp_net_dp *dp, struct nfp_net_r_vector *r_vec,
+ struct nfp_net_rx_ring *rx_ring, struct nfp_net_rx_buf *rxbuf,
+ struct sk_buff *skb)
{
u64_stats_update_begin(&r_vec->rx_sync);
r_vec->rx_drops++;
@@ -1442,7 +1450,7 @@ nfp_net_rx_drop(struct nfp_net_r_vector *r_vec, struct nfp_net_rx_ring *rx_ring,
if (skb && rxbuf && skb->head == rxbuf->frag)
page_ref_inc(virt_to_head_page(rxbuf->frag));
if (rxbuf)
- nfp_net_rx_give_one(rx_ring, rxbuf->frag, rxbuf->dma_addr);
+ nfp_net_rx_give_one(dp, rx_ring, rxbuf->frag, rxbuf->dma_addr);
if (skb)
dev_kfree_skb_any(skb);
}
@@ -1460,16 +1468,16 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
int wr_idx;
if (unlikely(nfp_net_tx_full(tx_ring, 1))) {
- nfp_net_rx_drop(rx_ring->r_vec, rx_ring, rxbuf, NULL);
+ nfp_net_rx_drop(dp, rx_ring->r_vec, rx_ring, rxbuf, NULL);
return false;
}
new_frag = nfp_net_napi_alloc_one(dp, &new_dma_addr);
if (unlikely(!new_frag)) {
- nfp_net_rx_drop(rx_ring->r_vec, rx_ring, rxbuf, NULL);
+ nfp_net_rx_drop(dp, rx_ring->r_vec, rx_ring, rxbuf, NULL);
return false;
}
- nfp_net_rx_give_one(rx_ring, new_frag, new_dma_addr);
+ nfp_net_rx_give_one(dp, rx_ring, new_frag, new_dma_addr);
wr_idx = tx_ring->wr_p & (tx_ring->cnt - 1);
@@ -1500,14 +1508,24 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring,
return true;
}
-static int nfp_net_run_xdp(struct bpf_prog *prog, void *data, unsigned int len)
+static int nfp_net_run_xdp(struct bpf_prog *prog, void *data, void *hard_start,
+ unsigned int *off, unsigned int *len)
{
struct xdp_buff xdp;
+ void *orig_data;
+ int ret;
+
+ xdp.data_hard_start = hard_start;
+ xdp.data = data + *off;
+ xdp.data_end = data + *off + *len;
- xdp.data = data;
- xdp.data_end = data + len;
+ orig_data = xdp.data;
+ ret = bpf_prog_run_xdp(prog, &xdp);
- return bpf_prog_run_xdp(prog, &xdp);
+ *len -= xdp.data - orig_data;
+ *off += xdp.data - orig_data;
+
+ return ret;
}
/**
@@ -1539,6 +1557,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
while (pkts_polled < budget) {
unsigned int meta_len, data_len, data_off, pkt_len;
+ u8 meta_prepend[NFP_NET_MAX_PREPEND];
struct nfp_net_rx_buf *rxbuf;
struct nfp_net_rx_desc *rxd;
dma_addr_t new_dma_addr;
@@ -1580,6 +1599,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
data_off = NFP_NET_RX_BUF_HEADROOM + meta_len;
else
data_off = NFP_NET_RX_BUF_HEADROOM + dp->rx_offset;
+ data_off += dp->rx_dma_off;
/* Stats update */
u64_stats_update_begin(&r_vec->rx_sync);
@@ -1594,25 +1614,35 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
(dp->rx_offset && meta_len > dp->rx_offset))) {
nn_dp_warn(dp, "oversized RX packet metadata %u\n",
meta_len);
- nfp_net_rx_drop(r_vec, rx_ring, rxbuf, NULL);
+ nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf, NULL);
continue;
}
if (xdp_prog && !(rxd->rxd.flags & PCIE_DESC_RX_BPF &&
dp->bpf_offload_xdp)) {
unsigned int dma_off;
+ void *hard_start;
int act;
+ hard_start = rxbuf->frag + NFP_NET_RX_BUF_HEADROOM;
dma_off = data_off - NFP_NET_RX_BUF_HEADROOM;
- dma_sync_single_for_cpu(dp->dev,
- rxbuf->dma_addr + dma_off,
- pkt_len, DMA_BIDIRECTIONAL);
- act = nfp_net_run_xdp(xdp_prog, rxbuf->frag + data_off,
- pkt_len);
+ dma_sync_single_for_cpu(dp->dev, rxbuf->dma_addr,
+ dma_off + pkt_len,
+ DMA_BIDIRECTIONAL);
+
+ /* Move prepend out of the way */
+ if (xdp_prog->xdp_adjust_head) {
+ memcpy(meta_prepend, meta, meta_len);
+ meta = meta_prepend;
+ }
+
+ act = nfp_net_run_xdp(xdp_prog, rxbuf->frag, hard_start,
+ &data_off, &pkt_len);
switch (act) {
case XDP_PASS:
break;
case XDP_TX:
+ dma_off = data_off - NFP_NET_RX_BUF_HEADROOM;
if (unlikely(!nfp_net_tx_xdp_buf(dp, rx_ring,
tx_ring, rxbuf,
dma_off,
@@ -1625,7 +1655,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
case XDP_ABORTED:
trace_xdp_exception(dp->netdev, xdp_prog, act);
case XDP_DROP:
- nfp_net_rx_give_one(rx_ring, rxbuf->frag,
+ nfp_net_rx_give_one(dp, rx_ring, rxbuf->frag,
rxbuf->dma_addr);
continue;
}
@@ -1633,18 +1663,18 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
skb = build_skb(rxbuf->frag, true_bufsz);
if (unlikely(!skb)) {
- nfp_net_rx_drop(r_vec, rx_ring, rxbuf, NULL);
+ nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf, NULL);
continue;
}
new_frag = nfp_net_napi_alloc_one(dp, &new_dma_addr);
if (unlikely(!new_frag)) {
- nfp_net_rx_drop(r_vec, rx_ring, rxbuf, skb);
+ nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf, skb);
continue;
}
nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr);
- nfp_net_rx_give_one(rx_ring, new_frag, new_dma_addr);
+ nfp_net_rx_give_one(dp, rx_ring, new_frag, new_dma_addr);
skb_reserve(skb, data_off);
skb_put(skb, pkt_len);
@@ -1658,7 +1688,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget)
meta_len);
if (unlikely(end != meta + meta_len)) {
nn_dp_warn(dp, "invalid RX packet metadata\n");
- nfp_net_rx_drop(r_vec, rx_ring, NULL, skb);
+ nfp_net_rx_drop(dp, r_vec, rx_ring, NULL, skb);
continue;
}
}
@@ -2151,7 +2181,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn)
nn->dp.ctrl = new_ctrl;
for (r = 0; r < nn->dp.num_rx_rings; r++)
- nfp_net_rx_ring_fill_freelist(&nn->dp.rx_rings[r]);
+ nfp_net_rx_ring_fill_freelist(&nn->dp, &nn->dp.rx_rings[r]);
/* Since reconfiguration requests while NFP is down are ignored we
* have to wipe the entire VXLAN configuration and reinitialize it.
@@ -2877,10 +2907,6 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
struct nfp_net_dp *dp;
int err;
- if (prog && prog->xdp_adjust_head) {
- nn_err(nn, "Does not support bpf_xdp_adjust_head()\n");
- return -EOPNOTSUPP;
- }
if (!prog && !nn->dp.xdp_prog)
return 0;
if (prog && nn->dp.xdp_prog) {
@@ -2897,6 +2923,11 @@ static int nfp_net_xdp_setup(struct nfp_net *nn, struct bpf_prog *prog)
dp->xdp_prog = prog;
dp->num_tx_rings += prog ? nn->dp.num_rx_rings : -nn->dp.num_rx_rings;
dp->rx_dma_dir = prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
+ if (prog)
+ dp->rx_dma_off = XDP_PACKET_HEADROOM -
+ (nn->dp.rx_offset ?: NFP_NET_MAX_PREPEND);
+ else
+ dp->rx_dma_off = 0;
/* We need RX reconfig to remap the buffers (BIDIR vs FROM_DEV) */
err = nfp_net_ring_reconfig(nn, dp);
@@ -3124,6 +3155,13 @@ int nfp_net_netdev_init(struct net_device *netdev)
struct nfp_net *nn = netdev_priv(netdev);
int err;
+ /* XDP calls for 256 byte packet headroom which wouldn't fit in a u8.
+ * We, however, reuse the metadata prepend space for XDP buffers which
+ * is at least 1 byte long and as long as XDP headroom doesn't increase
+ * above 256 the *extra* XDP headroom will fit on 8 bits.
+ */
+ BUILD_BUG_ON(XDP_PACKET_HEADROOM > 256);
+
nn->dp.chained_metadata_format = nn->fw_ver.major > 3;
nn->dp.rx_dma_dir = DMA_FROM_DEVICE;
--
2.11.0
^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [PATCH net-next 00/13] nfp: XDP adjust head support
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
` (12 preceding siblings ...)
2017-03-10 18:38 ` [PATCH net-next 13/13] nfp: add support " Jakub Kicinski
@ 2017-03-13 6:43 ` David Miller
13 siblings, 0 replies; 15+ messages in thread
From: David Miller @ 2017-03-13 6:43 UTC (permalink / raw)
To: jakub.kicinski; +Cc: netdev, oss-drivers, kubakici
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Fri, 10 Mar 2017 10:38:26 -0800
> This series adds support for XDP adjust head. Bulk of the code
> is actually just paying technical debt. On reconfiguration request
> nfp was allocating new resources separately leaving device running
> with the existing set of rings. We used to manage the new resources
> in special ring set structures. This set is simply separating the
> datapath part of the device structure from the control information
> allowing the new datapath structure to be allocated with all new
> memory and rings. The swap operation is now greatly simplified.
> We also save a lot of parameter passing this way. Hopefully the churn
> is worth the negative diffstat.
>
> Support for XDP adjust head is done in a pretty standard way. NFP
> is a bit special because it prepends metadata before packet data
> so we have to do a bit of memcpying in case XDP will run. We also
> luck out a little bit because the fact that we already have prepend
> space allocated means that one byte is enough to store the extra XDP
> space (256 of standard prepend space is a bit inconvenient since
> it would normally require 16bits or boolean with additional shifts).
Series applied, thanks.
^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2017-03-13 6:43 UTC | newest]
Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-03-10 18:38 [PATCH net-next 00/13] nfp: XDP adjust head support Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 01/13] nfp: separate data path information from the reset of adapter structure Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 02/13] nfp: move control BAR pointer into data path structure Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 03/13] nfp: pass new data path to ring reconfig Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 04/13] nfp: use dp to carry number of stack tx rings and vectors Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 05/13] nfp: use dp to carry fl_bufsz at reconfig time Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 06/13] nfp: use dp to carry mtu " Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 07/13] nfp: use dp to carry xdp_prog " Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 08/13] nfp: switch to using data path structures for reconfiguration Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 09/13] nfp: store dma direction in data path structure Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 10/13] nfp: validate rx offset from the BAR and size down it's field Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 11/13] nfp: reorganize pkt_off variable Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 12/13] nfp: prepare metadata handling for xdp_adjust_head() Jakub Kicinski
2017-03-10 18:38 ` [PATCH net-next 13/13] nfp: add support " Jakub Kicinski
2017-03-13 6:43 ` [PATCH net-next 00/13] nfp: XDP adjust head support David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).