summaryrefslogtreecommitdiffstats
path: root/drivers/net/ethernet/intel/ice/ice_txrx.h
blob: a07101b13226e170c954e6111c34fa689f45a89e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright (c) 2018, Intel Corporation. */

#ifndef _ICE_TXRX_H_
#define _ICE_TXRX_H_

#define ICE_DFLT_IRQ_WORK	256
#define ICE_RXBUF_2048		2048
#define ICE_MAX_CHAINED_RX_BUFS	5
#define ICE_MAX_BUF_TXD		8
#define ICE_MIN_TX_LEN		17

/* The size limit for a transmit buffer in a descriptor is (16K - 1).
 * In order to align with the read requests we will align the value to
 * the nearest 4K which represents our maximum read request size.
 */
#define ICE_MAX_READ_REQ_SIZE	4096
#define ICE_MAX_DATA_PER_TXD	(16 * 1024 - 1)
#define ICE_MAX_DATA_PER_TXD_ALIGNED \
	(~(ICE_MAX_READ_REQ_SIZE - 1) & ICE_MAX_DATA_PER_TXD)

#define ICE_RX_BUF_WRITE	16	/* Must be power of 2 */
#define ICE_MAX_TXQ_PER_TXQG	128

/* We are assuming that the cache line is always 64 Bytes here for ice.
 * In order to make sure that is a correct assumption there is a check in probe
 * to print a warning if the read from GLPCI_CNF2 tells us that the cache line
 * size is 128 bytes. We do it this way because we do not want to read the
 * GLPCI_CNF2 register or a variable containing the value on every pass through
 * the Tx path.
 */
#define ICE_CACHE_LINE_BYTES		64
#define ICE_DESCS_PER_CACHE_LINE	(ICE_CACHE_LINE_BYTES / \
					 sizeof(struct ice_tx_desc))
#define ICE_DESCS_FOR_CTX_DESC		1
#define ICE_DESCS_FOR_SKB_DATA_PTR	1
/* Tx descriptors needed, worst case */
#define DESC_NEEDED (MAX_SKB_FRAGS + ICE_DESCS_FOR_CTX_DESC + \
		     ICE_DESCS_PER_CACHE_LINE + ICE_DESCS_FOR_SKB_DATA_PTR)
#define ICE_DESC_UNUSED(R)	\
	((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \
	(R)->next_to_clean - (R)->next_to_use - 1)

#define ICE_TX_FLAGS_TSO	BIT(0)
#define ICE_TX_FLAGS_HW_VLAN	BIT(1)
#define ICE_TX_FLAGS_SW_VLAN	BIT(2)
#define ICE_TX_FLAGS_VLAN_M	0xffff0000
#define ICE_TX_FLAGS_VLAN_PR_M	0xe0000000
#define ICE_TX_FLAGS_VLAN_PR_S	29
#define ICE_TX_FLAGS_VLAN_S	16

#define ICE_XDP_PASS		0
#define ICE_XDP_CONSUMED	BIT(0)
#define ICE_XDP_TX		BIT(1)
#define ICE_XDP_REDIR		BIT(2)

#define ICE_RX_DMA_ATTR \
	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)

#define ICE_ETH_PKT_HDR_PAD	(ETH_HLEN + ETH_FCS_LEN + (VLAN_HLEN * 2))

#define ICE_TXD_LAST_DESC_CMD (ICE_TX_DESC_CMD_EOP | ICE_TX_DESC_CMD_RS)

struct ice_tx_buf {
	struct ice_tx_desc *next_to_watch;
	union {
		struct sk_buff *skb;
		void *raw_buf; /* used for XDP */
	};
	unsigned int bytecount;
	unsigned short gso_segs;
	u32 tx_flags;
	DEFINE_DMA_UNMAP_LEN(len);
	DEFINE_DMA_UNMAP_ADDR(dma);
};

struct ice_tx_offload_params {
	u64 cd_qw1;
	struct ice_ring *tx_ring;
	u32 td_cmd;
	u32 td_offset;
	u32 td_l2tag1;
	u32 cd_tunnel_params;
	u16 cd_l2tag2;
	u8 header_len;
};

struct ice_rx_buf {
	struct sk_buff *skb;
	dma_addr_t dma;
	struct page *page;
	unsigned int page_offset;
	u16 pagecnt_bias;
};

struct ice_q_stats {
	u64 pkts;
	u64 bytes;
};

struct ice_txq_stats {
	u64 restart_q;
	u64 tx_busy;
	u64 tx_linearize;
	int prev_pkt; /* negative if no pending Tx descriptors */
};

struct ice_rxq_stats {
	u64 non_eop_descs;
	u64 alloc_page_failed;
	u64 alloc_buf_failed;
	u64 page_reuse_count;
};

/* this enum matches hardware bits and is meant to be used by DYN_CTLN
 * registers and QINT registers or more generally anywhere in the manual
 * mentioning ITR_INDX, ITR_NONE cannot be used as an index 'n' into any
 * register but instead is a special value meaning "don't update" ITR0/1/2.
 */
enum ice_dyn_idx_t {
	ICE_IDX_ITR0 = 0,
	ICE_IDX_ITR1 = 1,
	ICE_IDX_ITR2 = 2,
	ICE_ITR_NONE = 3	/* ITR_NONE must not be used as an index */
};

/* Header split modes defined by DTYPE field of Rx RLAN context */
enum ice_rx_dtype {
	ICE_RX_DTYPE_NO_SPLIT		= 0,
	ICE_RX_DTYPE_HEADER_SPLIT	= 1,
	ICE_RX_DTYPE_SPLIT_ALWAYS	= 2,
};

/* indices into GLINT_ITR registers */
#define ICE_RX_ITR	ICE_IDX_ITR0
#define ICE_TX_ITR	ICE_IDX_ITR1
#define ICE_ITR_8K	124
#define ICE_ITR_20K	50
#define ICE_ITR_MAX	8160
#define ICE_DFLT_TX_ITR	(ICE_ITR_20K | ICE_ITR_DYNAMIC)
#define ICE_DFLT_RX_ITR	(ICE_ITR_20K | ICE_ITR_DYNAMIC)
#define ICE_ITR_DYNAMIC	0x8000  /* used as flag for itr_setting */
#define ITR_IS_DYNAMIC(setting) (!!((setting) & ICE_ITR_DYNAMIC))
#define ITR_TO_REG(setting)	((setting) & ~ICE_ITR_DYNAMIC)
#define ICE_ITR_GRAN_S		1	/* ITR granularity is always 2us */
#define ICE_ITR_GRAN_US		BIT(ICE_ITR_GRAN_S)
#define ICE_ITR_MASK		0x1FFE	/* ITR register value alignment mask */
#define ITR_REG_ALIGN(setting)	__ALIGN_MASK(setting, ~ICE_ITR_MASK)

#define ICE_ITR_ADAPTIVE_MIN_INC	0x0002
#define ICE_ITR_ADAPTIVE_MIN_USECS	0x0002
#define ICE_ITR_ADAPTIVE_MAX_USECS	0x00FA
#define ICE_ITR_ADAPTIVE_LATENCY	0x8000
#define ICE_ITR_ADAPTIVE_BULK		0x0000

#define ICE_DFLT_INTRL	0
#define ICE_MAX_INTRL	236

#define ICE_WB_ON_ITR_USECS	2
#define ICE_IN_WB_ON_ITR_MODE	255
/* Sets WB_ON_ITR and assumes INTENA bit is already cleared, which allows
 * setting the MSK_M bit to tell hardware to ignore the INTENA_M bit. Also,
 * set the write-back latency to the usecs passed in.
 */
#define ICE_GLINT_DYN_CTL_WB_ON_ITR(usecs, itr_idx)	\
	((((usecs) << (GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S)) & \
	  GLINT_DYN_CTL_INTERVAL_M) | \
	 (((itr_idx) << GLINT_DYN_CTL_ITR_INDX_S) & \
	  GLINT_DYN_CTL_ITR_INDX_M) | GLINT_DYN_CTL_INTENA_MSK_M | \
	 GLINT_DYN_CTL_WB_ON_ITR_M)

/* Legacy or Advanced Mode Queue */
#define ICE_TX_ADVANCED	0
#define ICE_TX_LEGACY	1

/* descriptor ring, associated with a VSI */
struct ice_ring {
	/* CL1 - 1st cacheline starts here */
	struct ice_ring *next;		/* pointer to next ring in q_vector */
	void *desc;			/* Descriptor ring memory */
	struct device *dev;		/* Used for DMA mapping */
	struct net_device *netdev;	/* netdev ring maps to */
	struct ice_vsi *vsi;		/* Backreference to associated VSI */
	struct ice_q_vector *q_vector;	/* Backreference to associated vector */
	u8 __iomem *tail;
	union {
		struct ice_tx_buf *tx_buf;
		struct ice_rx_buf *rx_buf;
	};
	/* CL2 - 2nd cacheline starts here */
	u16 q_index;			/* Queue number of ring */
	u16 q_handle;			/* Queue handle per TC */

	u8 ring_active:1;		/* is ring online or not */

	u16 count;			/* Number of descriptors */
	u16 reg_idx;			/* HW register index of the ring */

	/* used in interrupt processing */
	u16 next_to_use;
	u16 next_to_clean;
	u16 next_to_alloc;

	/* stats structs */
	struct ice_q_stats	stats;
	struct u64_stats_sync syncp;
	union {
		struct ice_txq_stats tx_stats;
		struct ice_rxq_stats rx_stats;
	};

	struct rcu_head rcu;		/* to avoid race on free */
	struct bpf_prog *xdp_prog;
	/* CL3 - 3rd cacheline starts here */
	struct xdp_rxq_info xdp_rxq;
	/* CLX - the below items are only accessed infrequently and should be
	 * in their own cache line if possible
	 */
#define ICE_TX_FLAGS_RING_XDP		BIT(0)
	u8 flags;
	dma_addr_t dma;			/* physical address of ring */
	unsigned int size;		/* length of descriptor ring in bytes */
	u32 txq_teid;			/* Added Tx queue TEID */
	u16 rx_buf_len;
	u8 dcb_tc;			/* Traffic class of ring */
} ____cacheline_internodealigned_in_smp;

static inline bool ice_ring_is_xdp(struct ice_ring *ring)
{
	return !!(ring->flags & ICE_TX_FLAGS_RING_XDP);
}

struct ice_ring_container {
	/* head of linked-list of rings */
	struct ice_ring *ring;
	unsigned long next_update;	/* jiffies value of next queue update */
	unsigned int total_bytes;	/* total bytes processed this int */
	unsigned int total_pkts;	/* total packets processed this int */
	u16 itr_idx;		/* index in the interrupt vector */
	u16 target_itr;		/* value in usecs divided by the hw->itr_gran */
	u16 current_itr;	/* value in usecs divided by the hw->itr_gran */
	/* high bit set means dynamic ITR, rest is used to store user
	 * readable ITR value in usecs and must be converted before programming
	 * to a register.
	 */
	u16 itr_setting;
};

/* iterator for handling rings in ring container */
#define ice_for_each_ring(pos, head) \
	for (pos = (head).ring; pos; pos = pos->next)

bool ice_alloc_rx_bufs(struct ice_ring *rxr, u16 cleaned_count);
netdev_tx_t ice_start_xmit(struct sk_buff *skb, struct net_device *netdev);
void ice_clean_tx_ring(struct ice_ring *tx_ring);
void ice_clean_rx_ring(struct ice_ring *rx_ring);
int ice_setup_tx_ring(struct ice_ring *tx_ring);
int ice_setup_rx_ring(struct ice_ring *rx_ring);
void ice_free_tx_ring(struct ice_ring *tx_ring);
void ice_free_rx_ring(struct ice_ring *rx_ring);
int ice_napi_poll(struct napi_struct *napi, int budget);

#endif /* _ICE_TXRX_H_ */
OpenPOWER on IntegriCloud