]>
Commit | Line | Data |
---|---|---|
4d22de3e | 1 | /* |
a02d44a0 | 2 | * Copyright (c) 2005-2008 Chelsio, Inc. All rights reserved. |
4d22de3e | 3 | * |
1d68e93d DLR |
4 | * This software is available to you under a choice of one of two |
5 | * licenses. You may choose to be licensed under the terms of the GNU | |
6 | * General Public License (GPL) Version 2, available from the file | |
7 | * COPYING in the main directory of this source tree, or the | |
8 | * OpenIB.org BSD license below: | |
4d22de3e | 9 | * |
1d68e93d DLR |
10 | * Redistribution and use in source and binary forms, with or |
11 | * without modification, are permitted provided that the following | |
12 | * conditions are met: | |
13 | * | |
14 | * - Redistributions of source code must retain the above | |
15 | * copyright notice, this list of conditions and the following | |
16 | * disclaimer. | |
17 | * | |
18 | * - Redistributions in binary form must reproduce the above | |
19 | * copyright notice, this list of conditions and the following | |
20 | * disclaimer in the documentation and/or other materials | |
21 | * provided with the distribution. | |
22 | * | |
23 | * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, | |
24 | * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF | |
25 | * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND | |
26 | * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS | |
27 | * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN | |
28 | * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN | |
29 | * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE | |
30 | * SOFTWARE. | |
4d22de3e | 31 | */ |
4d22de3e DLR |
32 | #include <linux/skbuff.h> |
33 | #include <linux/netdevice.h> | |
34 | #include <linux/etherdevice.h> | |
35 | #include <linux/if_vlan.h> | |
36 | #include <linux/ip.h> | |
37 | #include <linux/tcp.h> | |
38 | #include <linux/dma-mapping.h> | |
a109a5b9 | 39 | #include <net/arp.h> |
4d22de3e DLR |
40 | #include "common.h" |
41 | #include "regs.h" | |
42 | #include "sge_defs.h" | |
43 | #include "t3_cpl.h" | |
44 | #include "firmware_exports.h" | |
45 | ||
46 | #define USE_GTS 0 | |
47 | ||
48 | #define SGE_RX_SM_BUF_SIZE 1536 | |
e0994eb1 | 49 | |
4d22de3e | 50 | #define SGE_RX_COPY_THRES 256 |
cf992af5 | 51 | #define SGE_RX_PULL_LEN 128 |
4d22de3e | 52 | |
5e68b772 | 53 | #define SGE_PG_RSVD SMP_CACHE_BYTES |
e0994eb1 | 54 | /* |
cf992af5 DLR |
55 | * Page chunk size for FL0 buffers if FL0 is to be populated with page chunks. |
56 | * It must be a divisor of PAGE_SIZE. If set to 0 FL0 will use sk_buffs | |
57 | * directly. | |
e0994eb1 | 58 | */ |
cf992af5 | 59 | #define FL0_PG_CHUNK_SIZE 2048 |
7385ecf3 | 60 | #define FL0_PG_ORDER 0 |
5e68b772 | 61 | #define FL0_PG_ALLOC_SIZE (PAGE_SIZE << FL0_PG_ORDER) |
7385ecf3 DLR |
62 | #define FL1_PG_CHUNK_SIZE (PAGE_SIZE > 8192 ? 16384 : 8192) |
63 | #define FL1_PG_ORDER (PAGE_SIZE > 8192 ? 0 : 1) | |
5e68b772 | 64 | #define FL1_PG_ALLOC_SIZE (PAGE_SIZE << FL1_PG_ORDER) |
cf992af5 | 65 | |
e0994eb1 | 66 | #define SGE_RX_DROP_THRES 16 |
42c8ea17 | 67 | #define RX_RECLAIM_PERIOD (HZ/4) |
4d22de3e | 68 | |
26b3871d DLR |
69 | /* |
70 | * Max number of Rx buffers we replenish at a time. | |
71 | */ | |
72 | #define MAX_RX_REFILL 16U | |
4d22de3e DLR |
73 | /* |
74 | * Period of the Tx buffer reclaim timer. This timer does not need to run | |
75 | * frequently as Tx buffers are usually reclaimed by new Tx packets. | |
76 | */ | |
77 | #define TX_RECLAIM_PERIOD (HZ / 4) | |
42c8ea17 DLR |
78 | #define TX_RECLAIM_TIMER_CHUNK 64U |
79 | #define TX_RECLAIM_CHUNK 16U | |
4d22de3e DLR |
80 | |
81 | /* WR size in bytes */ | |
82 | #define WR_LEN (WR_FLITS * 8) | |
83 | ||
84 | /* | |
85 | * Types of Tx queues in each queue set. Order here matters, do not change. | |
86 | */ | |
87 | enum { TXQ_ETH, TXQ_OFLD, TXQ_CTRL }; | |
88 | ||
89 | /* Values for sge_txq.flags */ | |
90 | enum { | |
91 | TXQ_RUNNING = 1 << 0, /* fetch engine is running */ | |
92 | TXQ_LAST_PKT_DB = 1 << 1, /* last packet rang the doorbell */ | |
93 | }; | |
94 | ||
95 | struct tx_desc { | |
fb8e4444 | 96 | __be64 flit[TX_DESC_FLITS]; |
4d22de3e DLR |
97 | }; |
98 | ||
99 | struct rx_desc { | |
100 | __be32 addr_lo; | |
101 | __be32 len_gen; | |
102 | __be32 gen2; | |
103 | __be32 addr_hi; | |
104 | }; | |
105 | ||
106 | struct tx_sw_desc { /* SW state per Tx descriptor */ | |
107 | struct sk_buff *skb; | |
23561c94 DLR |
108 | u8 eop; /* set if last descriptor for packet */ |
109 | u8 addr_idx; /* buffer index of first SGL entry in descriptor */ | |
110 | u8 fragidx; /* first page fragment associated with descriptor */ | |
111 | s8 sflit; /* start flit of first SGL entry in descriptor */ | |
4d22de3e DLR |
112 | }; |
113 | ||
cf992af5 | 114 | struct rx_sw_desc { /* SW state per Rx descriptor */ |
e0994eb1 DLR |
115 | union { |
116 | struct sk_buff *skb; | |
cf992af5 DLR |
117 | struct fl_pg_chunk pg_chunk; |
118 | }; | |
119 | DECLARE_PCI_UNMAP_ADDR(dma_addr); | |
4d22de3e DLR |
120 | }; |
121 | ||
122 | struct rsp_desc { /* response queue descriptor */ | |
123 | struct rss_header rss_hdr; | |
124 | __be32 flags; | |
125 | __be32 len_cq; | |
126 | u8 imm_data[47]; | |
127 | u8 intr_gen; | |
128 | }; | |
129 | ||
99d7cf30 DLR |
130 | /* |
131 | * Holds unmapping information for Tx packets that need deferred unmapping. | |
132 | * This structure lives at skb->head and must be allocated by callers. | |
133 | */ | |
134 | struct deferred_unmap_info { | |
135 | struct pci_dev *pdev; | |
136 | dma_addr_t addr[MAX_SKB_FRAGS + 1]; | |
137 | }; | |
138 | ||
4d22de3e DLR |
139 | /* |
140 | * Maps a number of flits to the number of Tx descriptors that can hold them. | |
141 | * The formula is | |
142 | * | |
143 | * desc = 1 + (flits - 2) / (WR_FLITS - 1). | |
144 | * | |
145 | * HW allows up to 4 descriptors to be combined into a WR. | |
146 | */ | |
147 | static u8 flit_desc_map[] = { | |
148 | 0, | |
149 | #if SGE_NUM_GENBITS == 1 | |
150 | 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, | |
151 | 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, | |
152 | 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, | |
153 | 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 | |
154 | #elif SGE_NUM_GENBITS == 2 | |
155 | 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, | |
156 | 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, | |
157 | 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, | |
158 | 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, | |
159 | #else | |
160 | # error "SGE_NUM_GENBITS must be 1 or 2" | |
161 | #endif | |
162 | }; | |
163 | ||
164 | static inline struct sge_qset *fl_to_qset(const struct sge_fl *q, int qidx) | |
165 | { | |
166 | return container_of(q, struct sge_qset, fl[qidx]); | |
167 | } | |
168 | ||
169 | static inline struct sge_qset *rspq_to_qset(const struct sge_rspq *q) | |
170 | { | |
171 | return container_of(q, struct sge_qset, rspq); | |
172 | } | |
173 | ||
174 | static inline struct sge_qset *txq_to_qset(const struct sge_txq *q, int qidx) | |
175 | { | |
176 | return container_of(q, struct sge_qset, txq[qidx]); | |
177 | } | |
178 | ||
179 | /** | |
180 | * refill_rspq - replenish an SGE response queue | |
181 | * @adapter: the adapter | |
182 | * @q: the response queue to replenish | |
183 | * @credits: how many new responses to make available | |
184 | * | |
185 | * Replenishes a response queue by making the supplied number of responses | |
186 | * available to HW. | |
187 | */ | |
188 | static inline void refill_rspq(struct adapter *adapter, | |
189 | const struct sge_rspq *q, unsigned int credits) | |
190 | { | |
afefce66 | 191 | rmb(); |
4d22de3e DLR |
192 | t3_write_reg(adapter, A_SG_RSPQ_CREDIT_RETURN, |
193 | V_RSPQ(q->cntxt_id) | V_CREDITS(credits)); | |
194 | } | |
195 | ||
196 | /** | |
197 | * need_skb_unmap - does the platform need unmapping of sk_buffs? | |
198 | * | |
199 | * Returns true if the platfrom needs sk_buff unmapping. The compiler | |
200 | * optimizes away unecessary code if this returns true. | |
201 | */ | |
202 | static inline int need_skb_unmap(void) | |
203 | { | |
204 | /* | |
205 | * This structure is used to tell if the platfrom needs buffer | |
206 | * unmapping by checking if DECLARE_PCI_UNMAP_ADDR defines anything. | |
207 | */ | |
208 | struct dummy { | |
209 | DECLARE_PCI_UNMAP_ADDR(addr); | |
210 | }; | |
211 | ||
212 | return sizeof(struct dummy) != 0; | |
213 | } | |
214 | ||
215 | /** | |
216 | * unmap_skb - unmap a packet main body and its page fragments | |
217 | * @skb: the packet | |
218 | * @q: the Tx queue containing Tx descriptors for the packet | |
219 | * @cidx: index of Tx descriptor | |
220 | * @pdev: the PCI device | |
221 | * | |
222 | * Unmap the main body of an sk_buff and its page fragments, if any. | |
223 | * Because of the fairly complicated structure of our SGLs and the desire | |
23561c94 DLR |
224 | * to conserve space for metadata, the information necessary to unmap an |
225 | * sk_buff is spread across the sk_buff itself (buffer lengths), the HW Tx | |
226 | * descriptors (the physical addresses of the various data buffers), and | |
227 | * the SW descriptor state (assorted indices). The send functions | |
228 | * initialize the indices for the first packet descriptor so we can unmap | |
229 | * the buffers held in the first Tx descriptor here, and we have enough | |
230 | * information at this point to set the state for the next Tx descriptor. | |
231 | * | |
232 | * Note that it is possible to clean up the first descriptor of a packet | |
233 | * before the send routines have written the next descriptors, but this | |
234 | * race does not cause any problem. We just end up writing the unmapping | |
235 | * info for the descriptor first. | |
4d22de3e DLR |
236 | */ |
237 | static inline void unmap_skb(struct sk_buff *skb, struct sge_txq *q, | |
238 | unsigned int cidx, struct pci_dev *pdev) | |
239 | { | |
240 | const struct sg_ent *sgp; | |
23561c94 DLR |
241 | struct tx_sw_desc *d = &q->sdesc[cidx]; |
242 | int nfrags, frag_idx, curflit, j = d->addr_idx; | |
4d22de3e | 243 | |
23561c94 DLR |
244 | sgp = (struct sg_ent *)&q->desc[cidx].flit[d->sflit]; |
245 | frag_idx = d->fragidx; | |
4d22de3e | 246 | |
23561c94 DLR |
247 | if (frag_idx == 0 && skb_headlen(skb)) { |
248 | pci_unmap_single(pdev, be64_to_cpu(sgp->addr[0]), | |
249 | skb_headlen(skb), PCI_DMA_TODEVICE); | |
4d22de3e DLR |
250 | j = 1; |
251 | } | |
252 | ||
23561c94 | 253 | curflit = d->sflit + 1 + j; |
4d22de3e DLR |
254 | nfrags = skb_shinfo(skb)->nr_frags; |
255 | ||
256 | while (frag_idx < nfrags && curflit < WR_FLITS) { | |
257 | pci_unmap_page(pdev, be64_to_cpu(sgp->addr[j]), | |
258 | skb_shinfo(skb)->frags[frag_idx].size, | |
259 | PCI_DMA_TODEVICE); | |
260 | j ^= 1; | |
261 | if (j == 0) { | |
262 | sgp++; | |
263 | curflit++; | |
264 | } | |
265 | curflit++; | |
266 | frag_idx++; | |
267 | } | |
268 | ||
23561c94 DLR |
269 | if (frag_idx < nfrags) { /* SGL continues into next Tx descriptor */ |
270 | d = cidx + 1 == q->size ? q->sdesc : d + 1; | |
271 | d->fragidx = frag_idx; | |
272 | d->addr_idx = j; | |
273 | d->sflit = curflit - WR_FLITS - j; /* sflit can be -1 */ | |
4d22de3e DLR |
274 | } |
275 | } | |
276 | ||
277 | /** | |
278 | * free_tx_desc - reclaims Tx descriptors and their buffers | |
279 | * @adapter: the adapter | |
280 | * @q: the Tx queue to reclaim descriptors from | |
281 | * @n: the number of descriptors to reclaim | |
282 | * | |
283 | * Reclaims Tx descriptors from an SGE Tx queue and frees the associated | |
284 | * Tx buffers. Called with the Tx queue lock held. | |
285 | */ | |
286 | static void free_tx_desc(struct adapter *adapter, struct sge_txq *q, | |
287 | unsigned int n) | |
288 | { | |
289 | struct tx_sw_desc *d; | |
290 | struct pci_dev *pdev = adapter->pdev; | |
291 | unsigned int cidx = q->cidx; | |
292 | ||
99d7cf30 DLR |
293 | const int need_unmap = need_skb_unmap() && |
294 | q->cntxt_id >= FW_TUNNEL_SGEEC_START; | |
295 | ||
4d22de3e DLR |
296 | d = &q->sdesc[cidx]; |
297 | while (n--) { | |
298 | if (d->skb) { /* an SGL is present */ | |
99d7cf30 | 299 | if (need_unmap) |
4d22de3e | 300 | unmap_skb(d->skb, q, cidx, pdev); |
23561c94 | 301 | if (d->eop) |
4d22de3e DLR |
302 | kfree_skb(d->skb); |
303 | } | |
304 | ++d; | |
305 | if (++cidx == q->size) { | |
306 | cidx = 0; | |
307 | d = q->sdesc; | |
308 | } | |
309 | } | |
310 | q->cidx = cidx; | |
311 | } | |
312 | ||
313 | /** | |
314 | * reclaim_completed_tx - reclaims completed Tx descriptors | |
315 | * @adapter: the adapter | |
316 | * @q: the Tx queue to reclaim completed descriptors from | |
42c8ea17 | 317 | * @chunk: maximum number of descriptors to reclaim |
4d22de3e DLR |
318 | * |
319 | * Reclaims Tx descriptors that the SGE has indicated it has processed, | |
320 | * and frees the associated buffers if possible. Called with the Tx | |
321 | * queue's lock held. | |
322 | */ | |
42c8ea17 DLR |
323 | static inline unsigned int reclaim_completed_tx(struct adapter *adapter, |
324 | struct sge_txq *q, | |
325 | unsigned int chunk) | |
4d22de3e DLR |
326 | { |
327 | unsigned int reclaim = q->processed - q->cleaned; | |
328 | ||
42c8ea17 | 329 | reclaim = min(chunk, reclaim); |
4d22de3e DLR |
330 | if (reclaim) { |
331 | free_tx_desc(adapter, q, reclaim); | |
332 | q->cleaned += reclaim; | |
333 | q->in_use -= reclaim; | |
334 | } | |
42c8ea17 | 335 | return q->processed - q->cleaned; |
4d22de3e DLR |
336 | } |
337 | ||
338 | /** | |
339 | * should_restart_tx - are there enough resources to restart a Tx queue? | |
340 | * @q: the Tx queue | |
341 | * | |
342 | * Checks if there are enough descriptors to restart a suspended Tx queue. | |
343 | */ | |
344 | static inline int should_restart_tx(const struct sge_txq *q) | |
345 | { | |
346 | unsigned int r = q->processed - q->cleaned; | |
347 | ||
348 | return q->in_use - r < (q->size >> 1); | |
349 | } | |
350 | ||
5e68b772 DLR |
351 | static void clear_rx_desc(struct pci_dev *pdev, const struct sge_fl *q, |
352 | struct rx_sw_desc *d) | |
9bb2b31e | 353 | { |
5e68b772 DLR |
354 | if (q->use_pages && d->pg_chunk.page) { |
355 | (*d->pg_chunk.p_cnt)--; | |
356 | if (!*d->pg_chunk.p_cnt) | |
357 | pci_unmap_page(pdev, | |
10b6d956 | 358 | d->pg_chunk.mapping, |
5e68b772 DLR |
359 | q->alloc_size, PCI_DMA_FROMDEVICE); |
360 | ||
361 | put_page(d->pg_chunk.page); | |
9bb2b31e DLR |
362 | d->pg_chunk.page = NULL; |
363 | } else { | |
5e68b772 DLR |
364 | pci_unmap_single(pdev, pci_unmap_addr(d, dma_addr), |
365 | q->buf_size, PCI_DMA_FROMDEVICE); | |
9bb2b31e DLR |
366 | kfree_skb(d->skb); |
367 | d->skb = NULL; | |
368 | } | |
369 | } | |
370 | ||
4d22de3e DLR |
371 | /** |
372 | * free_rx_bufs - free the Rx buffers on an SGE free list | |
373 | * @pdev: the PCI device associated with the adapter | |
374 | * @rxq: the SGE free list to clean up | |
375 | * | |
376 | * Release the buffers on an SGE free-buffer Rx queue. HW fetching from | |
377 | * this queue should be stopped before calling this function. | |
378 | */ | |
379 | static void free_rx_bufs(struct pci_dev *pdev, struct sge_fl *q) | |
380 | { | |
381 | unsigned int cidx = q->cidx; | |
382 | ||
383 | while (q->credits--) { | |
384 | struct rx_sw_desc *d = &q->sdesc[cidx]; | |
385 | ||
5e68b772 DLR |
386 | |
387 | clear_rx_desc(pdev, q, d); | |
4d22de3e DLR |
388 | if (++cidx == q->size) |
389 | cidx = 0; | |
390 | } | |
e0994eb1 | 391 | |
cf992af5 | 392 | if (q->pg_chunk.page) { |
7385ecf3 | 393 | __free_pages(q->pg_chunk.page, q->order); |
cf992af5 DLR |
394 | q->pg_chunk.page = NULL; |
395 | } | |
4d22de3e DLR |
396 | } |
397 | ||
398 | /** | |
399 | * add_one_rx_buf - add a packet buffer to a free-buffer list | |
cf992af5 | 400 | * @va: buffer start VA |
4d22de3e DLR |
401 | * @len: the buffer length |
402 | * @d: the HW Rx descriptor to write | |
403 | * @sd: the SW Rx descriptor to write | |
404 | * @gen: the generation bit value | |
405 | * @pdev: the PCI device associated with the adapter | |
406 | * | |
407 | * Add a buffer of the given length to the supplied HW and SW Rx | |
408 | * descriptors. | |
409 | */ | |
b1fb1f28 DLR |
410 | static inline int add_one_rx_buf(void *va, unsigned int len, |
411 | struct rx_desc *d, struct rx_sw_desc *sd, | |
412 | unsigned int gen, struct pci_dev *pdev) | |
4d22de3e DLR |
413 | { |
414 | dma_addr_t mapping; | |
415 | ||
e0994eb1 | 416 | mapping = pci_map_single(pdev, va, len, PCI_DMA_FROMDEVICE); |
8d8bb39b | 417 | if (unlikely(pci_dma_mapping_error(pdev, mapping))) |
b1fb1f28 DLR |
418 | return -ENOMEM; |
419 | ||
4d22de3e DLR |
420 | pci_unmap_addr_set(sd, dma_addr, mapping); |
421 | ||
422 | d->addr_lo = cpu_to_be32(mapping); | |
423 | d->addr_hi = cpu_to_be32((u64) mapping >> 32); | |
424 | wmb(); | |
425 | d->len_gen = cpu_to_be32(V_FLD_GEN1(gen)); | |
426 | d->gen2 = cpu_to_be32(V_FLD_GEN2(gen)); | |
b1fb1f28 | 427 | return 0; |
4d22de3e DLR |
428 | } |
429 | ||
5e68b772 DLR |
430 | static inline int add_one_rx_chunk(dma_addr_t mapping, struct rx_desc *d, |
431 | unsigned int gen) | |
432 | { | |
433 | d->addr_lo = cpu_to_be32(mapping); | |
434 | d->addr_hi = cpu_to_be32((u64) mapping >> 32); | |
435 | wmb(); | |
436 | d->len_gen = cpu_to_be32(V_FLD_GEN1(gen)); | |
437 | d->gen2 = cpu_to_be32(V_FLD_GEN2(gen)); | |
438 | return 0; | |
439 | } | |
440 | ||
441 | static int alloc_pg_chunk(struct adapter *adapter, struct sge_fl *q, | |
442 | struct rx_sw_desc *sd, gfp_t gfp, | |
7385ecf3 | 443 | unsigned int order) |
cf992af5 DLR |
444 | { |
445 | if (!q->pg_chunk.page) { | |
5e68b772 DLR |
446 | dma_addr_t mapping; |
447 | ||
7385ecf3 | 448 | q->pg_chunk.page = alloc_pages(gfp, order); |
cf992af5 DLR |
449 | if (unlikely(!q->pg_chunk.page)) |
450 | return -ENOMEM; | |
451 | q->pg_chunk.va = page_address(q->pg_chunk.page); | |
5e68b772 DLR |
452 | q->pg_chunk.p_cnt = q->pg_chunk.va + (PAGE_SIZE << order) - |
453 | SGE_PG_RSVD; | |
cf992af5 | 454 | q->pg_chunk.offset = 0; |
5e68b772 DLR |
455 | mapping = pci_map_page(adapter->pdev, q->pg_chunk.page, |
456 | 0, q->alloc_size, PCI_DMA_FROMDEVICE); | |
10b6d956 | 457 | q->pg_chunk.mapping = mapping; |
cf992af5 DLR |
458 | } |
459 | sd->pg_chunk = q->pg_chunk; | |
460 | ||
5e68b772 DLR |
461 | prefetch(sd->pg_chunk.p_cnt); |
462 | ||
cf992af5 | 463 | q->pg_chunk.offset += q->buf_size; |
7385ecf3 | 464 | if (q->pg_chunk.offset == (PAGE_SIZE << order)) |
cf992af5 DLR |
465 | q->pg_chunk.page = NULL; |
466 | else { | |
467 | q->pg_chunk.va += q->buf_size; | |
468 | get_page(q->pg_chunk.page); | |
469 | } | |
5e68b772 DLR |
470 | |
471 | if (sd->pg_chunk.offset == 0) | |
472 | *sd->pg_chunk.p_cnt = 1; | |
473 | else | |
474 | *sd->pg_chunk.p_cnt += 1; | |
475 | ||
cf992af5 DLR |
476 | return 0; |
477 | } | |
478 | ||
26b3871d DLR |
479 | static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q) |
480 | { | |
481 | if (q->pend_cred >= q->credits / 4) { | |
482 | q->pend_cred = 0; | |
483 | t3_write_reg(adap, A_SG_KDOORBELL, V_EGRCNTX(q->cntxt_id)); | |
484 | } | |
485 | } | |
486 | ||
4d22de3e DLR |
487 | /** |
488 | * refill_fl - refill an SGE free-buffer list | |
489 | * @adapter: the adapter | |
490 | * @q: the free-list to refill | |
491 | * @n: the number of new buffers to allocate | |
492 | * @gfp: the gfp flags for allocating new buffers | |
493 | * | |
494 | * (Re)populate an SGE free-buffer list with up to @n new packet buffers, | |
495 | * allocated with the supplied gfp flags. The caller must assure that | |
496 | * @n does not exceed the queue's capacity. | |
497 | */ | |
b1fb1f28 | 498 | static int refill_fl(struct adapter *adap, struct sge_fl *q, int n, gfp_t gfp) |
4d22de3e DLR |
499 | { |
500 | struct rx_sw_desc *sd = &q->sdesc[q->pidx]; | |
501 | struct rx_desc *d = &q->desc[q->pidx]; | |
b1fb1f28 | 502 | unsigned int count = 0; |
4d22de3e DLR |
503 | |
504 | while (n--) { | |
5e68b772 | 505 | dma_addr_t mapping; |
b1fb1f28 DLR |
506 | int err; |
507 | ||
cf992af5 | 508 | if (q->use_pages) { |
5e68b772 DLR |
509 | if (unlikely(alloc_pg_chunk(adap, q, sd, gfp, |
510 | q->order))) { | |
cf992af5 | 511 | nomem: q->alloc_failed++; |
e0994eb1 DLR |
512 | break; |
513 | } | |
10b6d956 | 514 | mapping = sd->pg_chunk.mapping + sd->pg_chunk.offset; |
5e68b772 DLR |
515 | pci_unmap_addr_set(sd, dma_addr, mapping); |
516 | ||
517 | add_one_rx_chunk(mapping, d, q->gen); | |
518 | pci_dma_sync_single_for_device(adap->pdev, mapping, | |
519 | q->buf_size - SGE_PG_RSVD, | |
520 | PCI_DMA_FROMDEVICE); | |
e0994eb1 | 521 | } else { |
5e68b772 | 522 | void *buf_start; |
e0994eb1 | 523 | |
5e68b772 | 524 | struct sk_buff *skb = alloc_skb(q->buf_size, gfp); |
cf992af5 DLR |
525 | if (!skb) |
526 | goto nomem; | |
e0994eb1 | 527 | |
cf992af5 DLR |
528 | sd->skb = skb; |
529 | buf_start = skb->data; | |
5e68b772 DLR |
530 | err = add_one_rx_buf(buf_start, q->buf_size, d, sd, |
531 | q->gen, adap->pdev); | |
532 | if (unlikely(err)) { | |
533 | clear_rx_desc(adap->pdev, q, sd); | |
534 | break; | |
535 | } | |
b1fb1f28 DLR |
536 | } |
537 | ||
4d22de3e DLR |
538 | d++; |
539 | sd++; | |
540 | if (++q->pidx == q->size) { | |
541 | q->pidx = 0; | |
542 | q->gen ^= 1; | |
543 | sd = q->sdesc; | |
544 | d = q->desc; | |
545 | } | |
b1fb1f28 | 546 | count++; |
4d22de3e | 547 | } |
26b3871d DLR |
548 | |
549 | q->credits += count; | |
550 | q->pend_cred += count; | |
551 | ring_fl_db(adap, q); | |
b1fb1f28 DLR |
552 | |
553 | return count; | |
4d22de3e DLR |
554 | } |
555 | ||
556 | static inline void __refill_fl(struct adapter *adap, struct sge_fl *fl) | |
557 | { | |
26b3871d | 558 | refill_fl(adap, fl, min(MAX_RX_REFILL, fl->size - fl->credits), |
7385ecf3 | 559 | GFP_ATOMIC | __GFP_COMP); |
4d22de3e DLR |
560 | } |
561 | ||
562 | /** | |
563 | * recycle_rx_buf - recycle a receive buffer | |
564 | * @adapter: the adapter | |
565 | * @q: the SGE free list | |
566 | * @idx: index of buffer to recycle | |
567 | * | |
568 | * Recycles the specified buffer on the given free list by adding it at | |
569 | * the next available slot on the list. | |
570 | */ | |
571 | static void recycle_rx_buf(struct adapter *adap, struct sge_fl *q, | |
572 | unsigned int idx) | |
573 | { | |
574 | struct rx_desc *from = &q->desc[idx]; | |
575 | struct rx_desc *to = &q->desc[q->pidx]; | |
576 | ||
cf992af5 | 577 | q->sdesc[q->pidx] = q->sdesc[idx]; |
4d22de3e DLR |
578 | to->addr_lo = from->addr_lo; /* already big endian */ |
579 | to->addr_hi = from->addr_hi; /* likewise */ | |
580 | wmb(); | |
581 | to->len_gen = cpu_to_be32(V_FLD_GEN1(q->gen)); | |
582 | to->gen2 = cpu_to_be32(V_FLD_GEN2(q->gen)); | |
4d22de3e DLR |
583 | |
584 | if (++q->pidx == q->size) { | |
585 | q->pidx = 0; | |
586 | q->gen ^= 1; | |
587 | } | |
26b3871d DLR |
588 | |
589 | q->credits++; | |
590 | q->pend_cred++; | |
591 | ring_fl_db(adap, q); | |
4d22de3e DLR |
592 | } |
593 | ||
594 | /** | |
595 | * alloc_ring - allocate resources for an SGE descriptor ring | |
596 | * @pdev: the PCI device | |
597 | * @nelem: the number of descriptors | |
598 | * @elem_size: the size of each descriptor | |
599 | * @sw_size: the size of the SW state associated with each ring element | |
600 | * @phys: the physical address of the allocated ring | |
601 | * @metadata: address of the array holding the SW state for the ring | |
602 | * | |
603 | * Allocates resources for an SGE descriptor ring, such as Tx queues, | |
604 | * free buffer lists, or response queues. Each SGE ring requires | |
605 | * space for its HW descriptors plus, optionally, space for the SW state | |
606 | * associated with each HW entry (the metadata). The function returns | |
607 | * three values: the virtual address for the HW ring (the return value | |
608 | * of the function), the physical address of the HW ring, and the address | |
609 | * of the SW ring. | |
610 | */ | |
611 | static void *alloc_ring(struct pci_dev *pdev, size_t nelem, size_t elem_size, | |
e0994eb1 | 612 | size_t sw_size, dma_addr_t * phys, void *metadata) |
4d22de3e DLR |
613 | { |
614 | size_t len = nelem * elem_size; | |
615 | void *s = NULL; | |
616 | void *p = dma_alloc_coherent(&pdev->dev, len, phys, GFP_KERNEL); | |
617 | ||
618 | if (!p) | |
619 | return NULL; | |
52565544 | 620 | if (sw_size && metadata) { |
4d22de3e DLR |
621 | s = kcalloc(nelem, sw_size, GFP_KERNEL); |
622 | ||
623 | if (!s) { | |
624 | dma_free_coherent(&pdev->dev, len, p, *phys); | |
625 | return NULL; | |
626 | } | |
4d22de3e | 627 | *(void **)metadata = s; |
52565544 | 628 | } |
4d22de3e DLR |
629 | memset(p, 0, len); |
630 | return p; | |
631 | } | |
632 | ||
204e2f98 DLR |
633 | /** |
634 | * t3_reset_qset - reset a sge qset | |
635 | * @q: the queue set | |
636 | * | |
637 | * Reset the qset structure. | |
638 | * the NAPI structure is preserved in the event of | |
639 | * the qset's reincarnation, for example during EEH recovery. | |
640 | */ | |
641 | static void t3_reset_qset(struct sge_qset *q) | |
642 | { | |
643 | if (q->adap && | |
644 | !(q->adap->flags & NAPI_INIT)) { | |
645 | memset(q, 0, sizeof(*q)); | |
646 | return; | |
647 | } | |
648 | ||
649 | q->adap = NULL; | |
650 | memset(&q->rspq, 0, sizeof(q->rspq)); | |
651 | memset(q->fl, 0, sizeof(struct sge_fl) * SGE_RXQ_PER_SET); | |
652 | memset(q->txq, 0, sizeof(struct sge_txq) * SGE_TXQ_PER_SET); | |
653 | q->txq_stopped = 0; | |
20d3fc11 | 654 | q->tx_reclaim_timer.function = NULL; /* for t3_stop_sge_timers() */ |
42c8ea17 | 655 | q->rx_reclaim_timer.function = NULL; |
76620aaf HX |
656 | q->nomem = 0; |
657 | napi_free_frags(&q->napi); | |
204e2f98 DLR |
658 | } |
659 | ||
660 | ||
4d22de3e DLR |
661 | /** |
662 | * free_qset - free the resources of an SGE queue set | |
663 | * @adapter: the adapter owning the queue set | |
664 | * @q: the queue set | |
665 | * | |
666 | * Release the HW and SW resources associated with an SGE queue set, such | |
667 | * as HW contexts, packet buffers, and descriptor rings. Traffic to the | |
668 | * queue set must be quiesced prior to calling this. | |
669 | */ | |
9265fabf | 670 | static void t3_free_qset(struct adapter *adapter, struct sge_qset *q) |
4d22de3e DLR |
671 | { |
672 | int i; | |
673 | struct pci_dev *pdev = adapter->pdev; | |
674 | ||
4d22de3e DLR |
675 | for (i = 0; i < SGE_RXQ_PER_SET; ++i) |
676 | if (q->fl[i].desc) { | |
b1186dee | 677 | spin_lock_irq(&adapter->sge.reg_lock); |
4d22de3e | 678 | t3_sge_disable_fl(adapter, q->fl[i].cntxt_id); |
b1186dee | 679 | spin_unlock_irq(&adapter->sge.reg_lock); |
4d22de3e DLR |
680 | free_rx_bufs(pdev, &q->fl[i]); |
681 | kfree(q->fl[i].sdesc); | |
682 | dma_free_coherent(&pdev->dev, | |
683 | q->fl[i].size * | |
684 | sizeof(struct rx_desc), q->fl[i].desc, | |
685 | q->fl[i].phys_addr); | |
686 | } | |
687 | ||
688 | for (i = 0; i < SGE_TXQ_PER_SET; ++i) | |
689 | if (q->txq[i].desc) { | |
b1186dee | 690 | spin_lock_irq(&adapter->sge.reg_lock); |
4d22de3e | 691 | t3_sge_enable_ecntxt(adapter, q->txq[i].cntxt_id, 0); |
b1186dee | 692 | spin_unlock_irq(&adapter->sge.reg_lock); |
4d22de3e DLR |
693 | if (q->txq[i].sdesc) { |
694 | free_tx_desc(adapter, &q->txq[i], | |
695 | q->txq[i].in_use); | |
696 | kfree(q->txq[i].sdesc); | |
697 | } | |
698 | dma_free_coherent(&pdev->dev, | |
699 | q->txq[i].size * | |
700 | sizeof(struct tx_desc), | |
701 | q->txq[i].desc, q->txq[i].phys_addr); | |
702 | __skb_queue_purge(&q->txq[i].sendq); | |
703 | } | |
704 | ||
705 | if (q->rspq.desc) { | |
b1186dee | 706 | spin_lock_irq(&adapter->sge.reg_lock); |
4d22de3e | 707 | t3_sge_disable_rspcntxt(adapter, q->rspq.cntxt_id); |
b1186dee | 708 | spin_unlock_irq(&adapter->sge.reg_lock); |
4d22de3e DLR |
709 | dma_free_coherent(&pdev->dev, |
710 | q->rspq.size * sizeof(struct rsp_desc), | |
711 | q->rspq.desc, q->rspq.phys_addr); | |
712 | } | |
713 | ||
204e2f98 | 714 | t3_reset_qset(q); |
4d22de3e DLR |
715 | } |
716 | ||
717 | /** | |
718 | * init_qset_cntxt - initialize an SGE queue set context info | |
719 | * @qs: the queue set | |
720 | * @id: the queue set id | |
721 | * | |
722 | * Initializes the TIDs and context ids for the queues of a queue set. | |
723 | */ | |
724 | static void init_qset_cntxt(struct sge_qset *qs, unsigned int id) | |
725 | { | |
726 | qs->rspq.cntxt_id = id; | |
727 | qs->fl[0].cntxt_id = 2 * id; | |
728 | qs->fl[1].cntxt_id = 2 * id + 1; | |
729 | qs->txq[TXQ_ETH].cntxt_id = FW_TUNNEL_SGEEC_START + id; | |
730 | qs->txq[TXQ_ETH].token = FW_TUNNEL_TID_START + id; | |
731 | qs->txq[TXQ_OFLD].cntxt_id = FW_OFLD_SGEEC_START + id; | |
732 | qs->txq[TXQ_CTRL].cntxt_id = FW_CTRL_SGEEC_START + id; | |
733 | qs->txq[TXQ_CTRL].token = FW_CTRL_TID_START + id; | |
734 | } | |
735 | ||
736 | /** | |
737 | * sgl_len - calculates the size of an SGL of the given capacity | |
738 | * @n: the number of SGL entries | |
739 | * | |
740 | * Calculates the number of flits needed for a scatter/gather list that | |
741 | * can hold the given number of entries. | |
742 | */ | |
743 | static inline unsigned int sgl_len(unsigned int n) | |
744 | { | |
745 | /* alternatively: 3 * (n / 2) + 2 * (n & 1) */ | |
746 | return (3 * n) / 2 + (n & 1); | |
747 | } | |
748 | ||
749 | /** | |
750 | * flits_to_desc - returns the num of Tx descriptors for the given flits | |
751 | * @n: the number of flits | |
752 | * | |
753 | * Calculates the number of Tx descriptors needed for the supplied number | |
754 | * of flits. | |
755 | */ | |
756 | static inline unsigned int flits_to_desc(unsigned int n) | |
757 | { | |
758 | BUG_ON(n >= ARRAY_SIZE(flit_desc_map)); | |
759 | return flit_desc_map[n]; | |
760 | } | |
761 | ||
cf992af5 DLR |
762 | /** |
763 | * get_packet - return the next ingress packet buffer from a free list | |
764 | * @adap: the adapter that received the packet | |
765 | * @fl: the SGE free list holding the packet | |
766 | * @len: the packet length including any SGE padding | |
767 | * @drop_thres: # of remaining buffers before we start dropping packets | |
768 | * | |
769 | * Get the next packet from a free list and complete setup of the | |
770 | * sk_buff. If the packet is small we make a copy and recycle the | |
771 | * original buffer, otherwise we use the original buffer itself. If a | |
772 | * positive drop threshold is supplied packets are dropped and their | |
773 | * buffers recycled if (a) the number of remaining buffers is under the | |
774 | * threshold and the packet is too big to copy, or (b) the packet should | |
775 | * be copied but there is no memory for the copy. | |
776 | */ | |
777 | static struct sk_buff *get_packet(struct adapter *adap, struct sge_fl *fl, | |
778 | unsigned int len, unsigned int drop_thres) | |
779 | { | |
780 | struct sk_buff *skb = NULL; | |
781 | struct rx_sw_desc *sd = &fl->sdesc[fl->cidx]; | |
782 | ||
783 | prefetch(sd->skb->data); | |
784 | fl->credits--; | |
785 | ||
786 | if (len <= SGE_RX_COPY_THRES) { | |
787 | skb = alloc_skb(len, GFP_ATOMIC); | |
788 | if (likely(skb != NULL)) { | |
789 | __skb_put(skb, len); | |
790 | pci_dma_sync_single_for_cpu(adap->pdev, | |
791 | pci_unmap_addr(sd, dma_addr), len, | |
792 | PCI_DMA_FROMDEVICE); | |
793 | memcpy(skb->data, sd->skb->data, len); | |
794 | pci_dma_sync_single_for_device(adap->pdev, | |
795 | pci_unmap_addr(sd, dma_addr), len, | |
796 | PCI_DMA_FROMDEVICE); | |
797 | } else if (!drop_thres) | |
798 | goto use_orig_buf; | |
799 | recycle: | |
800 | recycle_rx_buf(adap, fl, fl->cidx); | |
801 | return skb; | |
802 | } | |
803 | ||
26b3871d DLR |
804 | if (unlikely(fl->credits < drop_thres) && |
805 | refill_fl(adap, fl, min(MAX_RX_REFILL, fl->size - fl->credits - 1), | |
806 | GFP_ATOMIC | __GFP_COMP) == 0) | |
cf992af5 DLR |
807 | goto recycle; |
808 | ||
809 | use_orig_buf: | |
810 | pci_unmap_single(adap->pdev, pci_unmap_addr(sd, dma_addr), | |
811 | fl->buf_size, PCI_DMA_FROMDEVICE); | |
812 | skb = sd->skb; | |
813 | skb_put(skb, len); | |
814 | __refill_fl(adap, fl); | |
815 | return skb; | |
816 | } | |
817 | ||
818 | /** | |
819 | * get_packet_pg - return the next ingress packet buffer from a free list | |
820 | * @adap: the adapter that received the packet | |
821 | * @fl: the SGE free list holding the packet | |
822 | * @len: the packet length including any SGE padding | |
823 | * @drop_thres: # of remaining buffers before we start dropping packets | |
824 | * | |
825 | * Get the next packet from a free list populated with page chunks. | |
826 | * If the packet is small we make a copy and recycle the original buffer, | |
827 | * otherwise we attach the original buffer as a page fragment to a fresh | |
828 | * sk_buff. If a positive drop threshold is supplied packets are dropped | |
829 | * and their buffers recycled if (a) the number of remaining buffers is | |
830 | * under the threshold and the packet is too big to copy, or (b) there's | |
831 | * no system memory. | |
832 | * | |
833 | * Note: this function is similar to @get_packet but deals with Rx buffers | |
834 | * that are page chunks rather than sk_buffs. | |
835 | */ | |
836 | static struct sk_buff *get_packet_pg(struct adapter *adap, struct sge_fl *fl, | |
7385ecf3 DLR |
837 | struct sge_rspq *q, unsigned int len, |
838 | unsigned int drop_thres) | |
cf992af5 | 839 | { |
7385ecf3 | 840 | struct sk_buff *newskb, *skb; |
cf992af5 DLR |
841 | struct rx_sw_desc *sd = &fl->sdesc[fl->cidx]; |
842 | ||
5e68b772 | 843 | dma_addr_t dma_addr = pci_unmap_addr(sd, dma_addr); |
7385ecf3 | 844 | |
5e68b772 | 845 | newskb = skb = q->pg_skb; |
7385ecf3 DLR |
846 | if (!skb && (len <= SGE_RX_COPY_THRES)) { |
847 | newskb = alloc_skb(len, GFP_ATOMIC); | |
848 | if (likely(newskb != NULL)) { | |
849 | __skb_put(newskb, len); | |
5e68b772 | 850 | pci_dma_sync_single_for_cpu(adap->pdev, dma_addr, len, |
cf992af5 | 851 | PCI_DMA_FROMDEVICE); |
7385ecf3 | 852 | memcpy(newskb->data, sd->pg_chunk.va, len); |
5e68b772 DLR |
853 | pci_dma_sync_single_for_device(adap->pdev, dma_addr, |
854 | len, | |
855 | PCI_DMA_FROMDEVICE); | |
cf992af5 DLR |
856 | } else if (!drop_thres) |
857 | return NULL; | |
858 | recycle: | |
859 | fl->credits--; | |
860 | recycle_rx_buf(adap, fl, fl->cidx); | |
7385ecf3 DLR |
861 | q->rx_recycle_buf++; |
862 | return newskb; | |
cf992af5 DLR |
863 | } |
864 | ||
7385ecf3 | 865 | if (unlikely(q->rx_recycle_buf || (!skb && fl->credits <= drop_thres))) |
cf992af5 DLR |
866 | goto recycle; |
867 | ||
5e68b772 DLR |
868 | prefetch(sd->pg_chunk.p_cnt); |
869 | ||
7385ecf3 | 870 | if (!skb) |
b47385bd | 871 | newskb = alloc_skb(SGE_RX_PULL_LEN, GFP_ATOMIC); |
5e68b772 | 872 | |
7385ecf3 | 873 | if (unlikely(!newskb)) { |
cf992af5 DLR |
874 | if (!drop_thres) |
875 | return NULL; | |
876 | goto recycle; | |
877 | } | |
878 | ||
5e68b772 DLR |
879 | pci_dma_sync_single_for_cpu(adap->pdev, dma_addr, len, |
880 | PCI_DMA_FROMDEVICE); | |
881 | (*sd->pg_chunk.p_cnt)--; | |
882 | if (!*sd->pg_chunk.p_cnt) | |
883 | pci_unmap_page(adap->pdev, | |
10b6d956 | 884 | sd->pg_chunk.mapping, |
5e68b772 DLR |
885 | fl->alloc_size, |
886 | PCI_DMA_FROMDEVICE); | |
7385ecf3 DLR |
887 | if (!skb) { |
888 | __skb_put(newskb, SGE_RX_PULL_LEN); | |
889 | memcpy(newskb->data, sd->pg_chunk.va, SGE_RX_PULL_LEN); | |
890 | skb_fill_page_desc(newskb, 0, sd->pg_chunk.page, | |
891 | sd->pg_chunk.offset + SGE_RX_PULL_LEN, | |
892 | len - SGE_RX_PULL_LEN); | |
893 | newskb->len = len; | |
894 | newskb->data_len = len - SGE_RX_PULL_LEN; | |
8f435804 | 895 | newskb->truesize += newskb->data_len; |
7385ecf3 DLR |
896 | } else { |
897 | skb_fill_page_desc(newskb, skb_shinfo(newskb)->nr_frags, | |
898 | sd->pg_chunk.page, | |
899 | sd->pg_chunk.offset, len); | |
900 | newskb->len += len; | |
901 | newskb->data_len += len; | |
8f435804 | 902 | newskb->truesize += len; |
7385ecf3 | 903 | } |
cf992af5 DLR |
904 | |
905 | fl->credits--; | |
906 | /* | |
907 | * We do not refill FLs here, we let the caller do it to overlap a | |
908 | * prefetch. | |
909 | */ | |
7385ecf3 | 910 | return newskb; |
cf992af5 DLR |
911 | } |
912 | ||
4d22de3e DLR |
913 | /** |
914 | * get_imm_packet - return the next ingress packet buffer from a response | |
915 | * @resp: the response descriptor containing the packet data | |
916 | * | |
917 | * Return a packet containing the immediate data of the given response. | |
918 | */ | |
919 | static inline struct sk_buff *get_imm_packet(const struct rsp_desc *resp) | |
920 | { | |
921 | struct sk_buff *skb = alloc_skb(IMMED_PKT_SIZE, GFP_ATOMIC); | |
922 | ||
923 | if (skb) { | |
924 | __skb_put(skb, IMMED_PKT_SIZE); | |
27d7ff46 | 925 | skb_copy_to_linear_data(skb, resp->imm_data, IMMED_PKT_SIZE); |
4d22de3e DLR |
926 | } |
927 | return skb; | |
928 | } | |
929 | ||
930 | /** | |
931 | * calc_tx_descs - calculate the number of Tx descriptors for a packet | |
932 | * @skb: the packet | |
933 | * | |
934 | * Returns the number of Tx descriptors needed for the given Ethernet | |
935 | * packet. Ethernet packets require addition of WR and CPL headers. | |
936 | */ | |
937 | static inline unsigned int calc_tx_descs(const struct sk_buff *skb) | |
938 | { | |
939 | unsigned int flits; | |
940 | ||
941 | if (skb->len <= WR_LEN - sizeof(struct cpl_tx_pkt)) | |
942 | return 1; | |
943 | ||
944 | flits = sgl_len(skb_shinfo(skb)->nr_frags + 1) + 2; | |
945 | if (skb_shinfo(skb)->gso_size) | |
946 | flits++; | |
947 | return flits_to_desc(flits); | |
948 | } | |
949 | ||
950 | /** | |
951 | * make_sgl - populate a scatter/gather list for a packet | |
952 | * @skb: the packet | |
953 | * @sgp: the SGL to populate | |
954 | * @start: start address of skb main body data to include in the SGL | |
955 | * @len: length of skb main body data to include in the SGL | |
956 | * @pdev: the PCI device | |
957 | * | |
958 | * Generates a scatter/gather list for the buffers that make up a packet | |
959 | * and returns the SGL size in 8-byte words. The caller must size the SGL | |
960 | * appropriately. | |
961 | */ | |
962 | static inline unsigned int make_sgl(const struct sk_buff *skb, | |
963 | struct sg_ent *sgp, unsigned char *start, | |
964 | unsigned int len, struct pci_dev *pdev) | |
965 | { | |
966 | dma_addr_t mapping; | |
967 | unsigned int i, j = 0, nfrags; | |
968 | ||
969 | if (len) { | |
970 | mapping = pci_map_single(pdev, start, len, PCI_DMA_TODEVICE); | |
971 | sgp->len[0] = cpu_to_be32(len); | |
972 | sgp->addr[0] = cpu_to_be64(mapping); | |
973 | j = 1; | |
974 | } | |
975 | ||
976 | nfrags = skb_shinfo(skb)->nr_frags; | |
977 | for (i = 0; i < nfrags; i++) { | |
978 | skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; | |
979 | ||
980 | mapping = pci_map_page(pdev, frag->page, frag->page_offset, | |
981 | frag->size, PCI_DMA_TODEVICE); | |
982 | sgp->len[j] = cpu_to_be32(frag->size); | |
983 | sgp->addr[j] = cpu_to_be64(mapping); | |
984 | j ^= 1; | |
985 | if (j == 0) | |
986 | ++sgp; | |
987 | } | |
988 | if (j) | |
989 | sgp->len[j] = 0; | |
990 | return ((nfrags + (len != 0)) * 3) / 2 + j; | |
991 | } | |
992 | ||
993 | /** | |
994 | * check_ring_tx_db - check and potentially ring a Tx queue's doorbell | |
995 | * @adap: the adapter | |
996 | * @q: the Tx queue | |
997 | * | |
998 | * Ring the doorbel if a Tx queue is asleep. There is a natural race, | |
999 | * where the HW is going to sleep just after we checked, however, | |
1000 | * then the interrupt handler will detect the outstanding TX packet | |
1001 | * and ring the doorbell for us. | |
1002 | * | |
1003 | * When GTS is disabled we unconditionally ring the doorbell. | |
1004 | */ | |
1005 | static inline void check_ring_tx_db(struct adapter *adap, struct sge_txq *q) | |
1006 | { | |
1007 | #if USE_GTS | |
1008 | clear_bit(TXQ_LAST_PKT_DB, &q->flags); | |
1009 | if (test_and_set_bit(TXQ_RUNNING, &q->flags) == 0) { | |
1010 | set_bit(TXQ_LAST_PKT_DB, &q->flags); | |
1011 | t3_write_reg(adap, A_SG_KDOORBELL, | |
1012 | F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id)); | |
1013 | } | |
1014 | #else | |
1015 | wmb(); /* write descriptors before telling HW */ | |
1016 | t3_write_reg(adap, A_SG_KDOORBELL, | |
1017 | F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id)); | |
1018 | #endif | |
1019 | } | |
1020 | ||
1021 | static inline void wr_gen2(struct tx_desc *d, unsigned int gen) | |
1022 | { | |
1023 | #if SGE_NUM_GENBITS == 2 | |
1024 | d->flit[TX_DESC_FLITS - 1] = cpu_to_be64(gen); | |
1025 | #endif | |
1026 | } | |
1027 | ||
1028 | /** | |
1029 | * write_wr_hdr_sgl - write a WR header and, optionally, SGL | |
1030 | * @ndesc: number of Tx descriptors spanned by the SGL | |
1031 | * @skb: the packet corresponding to the WR | |
1032 | * @d: first Tx descriptor to be written | |
1033 | * @pidx: index of above descriptors | |
1034 | * @q: the SGE Tx queue | |
1035 | * @sgl: the SGL | |
1036 | * @flits: number of flits to the start of the SGL in the first descriptor | |
1037 | * @sgl_flits: the SGL size in flits | |
1038 | * @gen: the Tx descriptor generation | |
1039 | * @wr_hi: top 32 bits of WR header based on WR type (big endian) | |
1040 | * @wr_lo: low 32 bits of WR header based on WR type (big endian) | |
1041 | * | |
1042 | * Write a work request header and an associated SGL. If the SGL is | |
1043 | * small enough to fit into one Tx descriptor it has already been written | |
1044 | * and we just need to write the WR header. Otherwise we distribute the | |
1045 | * SGL across the number of descriptors it spans. | |
1046 | */ | |
1047 | static void write_wr_hdr_sgl(unsigned int ndesc, struct sk_buff *skb, | |
1048 | struct tx_desc *d, unsigned int pidx, | |
1049 | const struct sge_txq *q, | |
1050 | const struct sg_ent *sgl, | |
1051 | unsigned int flits, unsigned int sgl_flits, | |
fb8e4444 AV |
1052 | unsigned int gen, __be32 wr_hi, |
1053 | __be32 wr_lo) | |
4d22de3e DLR |
1054 | { |
1055 | struct work_request_hdr *wrp = (struct work_request_hdr *)d; | |
1056 | struct tx_sw_desc *sd = &q->sdesc[pidx]; | |
1057 | ||
1058 | sd->skb = skb; | |
1059 | if (need_skb_unmap()) { | |
23561c94 DLR |
1060 | sd->fragidx = 0; |
1061 | sd->addr_idx = 0; | |
1062 | sd->sflit = flits; | |
4d22de3e DLR |
1063 | } |
1064 | ||
1065 | if (likely(ndesc == 1)) { | |
23561c94 | 1066 | sd->eop = 1; |
4d22de3e DLR |
1067 | wrp->wr_hi = htonl(F_WR_SOP | F_WR_EOP | V_WR_DATATYPE(1) | |
1068 | V_WR_SGLSFLT(flits)) | wr_hi; | |
1069 | wmb(); | |
1070 | wrp->wr_lo = htonl(V_WR_LEN(flits + sgl_flits) | | |
1071 | V_WR_GEN(gen)) | wr_lo; | |
1072 | wr_gen2(d, gen); | |
1073 | } else { | |
1074 | unsigned int ogen = gen; | |
1075 | const u64 *fp = (const u64 *)sgl; | |
1076 | struct work_request_hdr *wp = wrp; | |
1077 | ||
1078 | wrp->wr_hi = htonl(F_WR_SOP | V_WR_DATATYPE(1) | | |
1079 | V_WR_SGLSFLT(flits)) | wr_hi; | |
1080 | ||
1081 | while (sgl_flits) { | |
1082 | unsigned int avail = WR_FLITS - flits; | |
1083 | ||
1084 | if (avail > sgl_flits) | |
1085 | avail = sgl_flits; | |
1086 | memcpy(&d->flit[flits], fp, avail * sizeof(*fp)); | |
1087 | sgl_flits -= avail; | |
1088 | ndesc--; | |
1089 | if (!sgl_flits) | |
1090 | break; | |
1091 | ||
1092 | fp += avail; | |
1093 | d++; | |
23561c94 | 1094 | sd->eop = 0; |
4d22de3e DLR |
1095 | sd++; |
1096 | if (++pidx == q->size) { | |
1097 | pidx = 0; | |
1098 | gen ^= 1; | |
1099 | d = q->desc; | |
1100 | sd = q->sdesc; | |
1101 | } | |
1102 | ||
1103 | sd->skb = skb; | |
1104 | wrp = (struct work_request_hdr *)d; | |
1105 | wrp->wr_hi = htonl(V_WR_DATATYPE(1) | | |
1106 | V_WR_SGLSFLT(1)) | wr_hi; | |
1107 | wrp->wr_lo = htonl(V_WR_LEN(min(WR_FLITS, | |
1108 | sgl_flits + 1)) | | |
1109 | V_WR_GEN(gen)) | wr_lo; | |
1110 | wr_gen2(d, gen); | |
1111 | flits = 1; | |
1112 | } | |
23561c94 | 1113 | sd->eop = 1; |
4d22de3e DLR |
1114 | wrp->wr_hi |= htonl(F_WR_EOP); |
1115 | wmb(); | |
1116 | wp->wr_lo = htonl(V_WR_LEN(WR_FLITS) | V_WR_GEN(ogen)) | wr_lo; | |
1117 | wr_gen2((struct tx_desc *)wp, ogen); | |
1118 | WARN_ON(ndesc != 0); | |
1119 | } | |
1120 | } | |
1121 | ||
1122 | /** | |
1123 | * write_tx_pkt_wr - write a TX_PKT work request | |
1124 | * @adap: the adapter | |
1125 | * @skb: the packet to send | |
1126 | * @pi: the egress interface | |
1127 | * @pidx: index of the first Tx descriptor to write | |
1128 | * @gen: the generation value to use | |
1129 | * @q: the Tx queue | |
1130 | * @ndesc: number of descriptors the packet will occupy | |
1131 | * @compl: the value of the COMPL bit to use | |
1132 | * | |
1133 | * Generate a TX_PKT work request to send the supplied packet. | |
1134 | */ | |
1135 | static void write_tx_pkt_wr(struct adapter *adap, struct sk_buff *skb, | |
1136 | const struct port_info *pi, | |
1137 | unsigned int pidx, unsigned int gen, | |
1138 | struct sge_txq *q, unsigned int ndesc, | |
1139 | unsigned int compl) | |
1140 | { | |
1141 | unsigned int flits, sgl_flits, cntrl, tso_info; | |
1142 | struct sg_ent *sgp, sgl[MAX_SKB_FRAGS / 2 + 1]; | |
1143 | struct tx_desc *d = &q->desc[pidx]; | |
1144 | struct cpl_tx_pkt *cpl = (struct cpl_tx_pkt *)d; | |
1145 | ||
3fa58c88 | 1146 | cpl->len = htonl(skb->len); |
4d22de3e DLR |
1147 | cntrl = V_TXPKT_INTF(pi->port_id); |
1148 | ||
1149 | if (vlan_tx_tag_present(skb) && pi->vlan_grp) | |
1150 | cntrl |= F_TXPKT_VLAN_VLD | V_TXPKT_VLAN(vlan_tx_tag_get(skb)); | |
1151 | ||
1152 | tso_info = V_LSO_MSS(skb_shinfo(skb)->gso_size); | |
1153 | if (tso_info) { | |
1154 | int eth_type; | |
1155 | struct cpl_tx_pkt_lso *hdr = (struct cpl_tx_pkt_lso *)cpl; | |
1156 | ||
1157 | d->flit[2] = 0; | |
1158 | cntrl |= V_TXPKT_OPCODE(CPL_TX_PKT_LSO); | |
1159 | hdr->cntrl = htonl(cntrl); | |
bbe735e4 | 1160 | eth_type = skb_network_offset(skb) == ETH_HLEN ? |
4d22de3e DLR |
1161 | CPL_ETH_II : CPL_ETH_II_VLAN; |
1162 | tso_info |= V_LSO_ETH_TYPE(eth_type) | | |
eddc9ec5 | 1163 | V_LSO_IPHDR_WORDS(ip_hdr(skb)->ihl) | |
aa8223c7 | 1164 | V_LSO_TCPHDR_WORDS(tcp_hdr(skb)->doff); |
4d22de3e DLR |
1165 | hdr->lso_info = htonl(tso_info); |
1166 | flits = 3; | |
1167 | } else { | |
1168 | cntrl |= V_TXPKT_OPCODE(CPL_TX_PKT); | |
1169 | cntrl |= F_TXPKT_IPCSUM_DIS; /* SW calculates IP csum */ | |
1170 | cntrl |= V_TXPKT_L4CSUM_DIS(skb->ip_summed != CHECKSUM_PARTIAL); | |
1171 | cpl->cntrl = htonl(cntrl); | |
1172 | ||
1173 | if (skb->len <= WR_LEN - sizeof(*cpl)) { | |
1174 | q->sdesc[pidx].skb = NULL; | |
1175 | if (!skb->data_len) | |
d626f62b ACM |
1176 | skb_copy_from_linear_data(skb, &d->flit[2], |
1177 | skb->len); | |
4d22de3e DLR |
1178 | else |
1179 | skb_copy_bits(skb, 0, &d->flit[2], skb->len); | |
1180 | ||
1181 | flits = (skb->len + 7) / 8 + 2; | |
1182 | cpl->wr.wr_hi = htonl(V_WR_BCNTLFLT(skb->len & 7) | | |
1183 | V_WR_OP(FW_WROPCODE_TUNNEL_TX_PKT) | |
1184 | | F_WR_SOP | F_WR_EOP | compl); | |
1185 | wmb(); | |
1186 | cpl->wr.wr_lo = htonl(V_WR_LEN(flits) | V_WR_GEN(gen) | | |
1187 | V_WR_TID(q->token)); | |
1188 | wr_gen2(d, gen); | |
1189 | kfree_skb(skb); | |
1190 | return; | |
1191 | } | |
1192 | ||
1193 | flits = 2; | |
1194 | } | |
1195 | ||
1196 | sgp = ndesc == 1 ? (struct sg_ent *)&d->flit[flits] : sgl; | |
1197 | sgl_flits = make_sgl(skb, sgp, skb->data, skb_headlen(skb), adap->pdev); | |
4d22de3e DLR |
1198 | |
1199 | write_wr_hdr_sgl(ndesc, skb, d, pidx, q, sgl, flits, sgl_flits, gen, | |
1200 | htonl(V_WR_OP(FW_WROPCODE_TUNNEL_TX_PKT) | compl), | |
1201 | htonl(V_WR_TID(q->token))); | |
1202 | } | |
1203 | ||
82ad3329 DLR |
1204 | static inline void t3_stop_tx_queue(struct netdev_queue *txq, |
1205 | struct sge_qset *qs, struct sge_txq *q) | |
a8cc21f6 | 1206 | { |
82ad3329 | 1207 | netif_tx_stop_queue(txq); |
a8cc21f6 KK |
1208 | set_bit(TXQ_ETH, &qs->txq_stopped); |
1209 | q->stops++; | |
1210 | } | |
1211 | ||
4d22de3e DLR |
1212 | /** |
1213 | * eth_xmit - add a packet to the Ethernet Tx queue | |
1214 | * @skb: the packet | |
1215 | * @dev: the egress net device | |
1216 | * | |
1217 | * Add a packet to an SGE Tx queue. Runs with softirqs disabled. | |
1218 | */ | |
61357325 | 1219 | netdev_tx_t t3_eth_xmit(struct sk_buff *skb, struct net_device *dev) |
4d22de3e | 1220 | { |
82ad3329 | 1221 | int qidx; |
4d22de3e DLR |
1222 | unsigned int ndesc, pidx, credits, gen, compl; |
1223 | const struct port_info *pi = netdev_priv(dev); | |
5fbf816f | 1224 | struct adapter *adap = pi->adapter; |
82ad3329 DLR |
1225 | struct netdev_queue *txq; |
1226 | struct sge_qset *qs; | |
1227 | struct sge_txq *q; | |
4d22de3e DLR |
1228 | |
1229 | /* | |
1230 | * The chip min packet length is 9 octets but play safe and reject | |
1231 | * anything shorter than an Ethernet header. | |
1232 | */ | |
1233 | if (unlikely(skb->len < ETH_HLEN)) { | |
1234 | dev_kfree_skb(skb); | |
1235 | return NETDEV_TX_OK; | |
1236 | } | |
1237 | ||
82ad3329 DLR |
1238 | qidx = skb_get_queue_mapping(skb); |
1239 | qs = &pi->qs[qidx]; | |
1240 | q = &qs->txq[TXQ_ETH]; | |
1241 | txq = netdev_get_tx_queue(dev, qidx); | |
1242 | ||
42c8ea17 | 1243 | reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK); |
4d22de3e DLR |
1244 | |
1245 | credits = q->size - q->in_use; | |
1246 | ndesc = calc_tx_descs(skb); | |
1247 | ||
1248 | if (unlikely(credits < ndesc)) { | |
82ad3329 | 1249 | t3_stop_tx_queue(txq, qs, q); |
a8cc21f6 KK |
1250 | dev_err(&adap->pdev->dev, |
1251 | "%s: Tx ring %u full while queue awake!\n", | |
1252 | dev->name, q->cntxt_id & 7); | |
4d22de3e DLR |
1253 | return NETDEV_TX_BUSY; |
1254 | } | |
1255 | ||
1256 | q->in_use += ndesc; | |
cd7e9034 | 1257 | if (unlikely(credits - ndesc < q->stop_thres)) { |
82ad3329 | 1258 | t3_stop_tx_queue(txq, qs, q); |
cd7e9034 DLR |
1259 | |
1260 | if (should_restart_tx(q) && | |
1261 | test_and_clear_bit(TXQ_ETH, &qs->txq_stopped)) { | |
1262 | q->restarts++; | |
82ad3329 | 1263 | netif_tx_wake_queue(txq); |
cd7e9034 DLR |
1264 | } |
1265 | } | |
4d22de3e DLR |
1266 | |
1267 | gen = q->gen; | |
1268 | q->unacked += ndesc; | |
1269 | compl = (q->unacked & 8) << (S_WR_COMPL - 3); | |
1270 | q->unacked &= 7; | |
1271 | pidx = q->pidx; | |
1272 | q->pidx += ndesc; | |
1273 | if (q->pidx >= q->size) { | |
1274 | q->pidx -= q->size; | |
1275 | q->gen ^= 1; | |
1276 | } | |
1277 | ||
1278 | /* update port statistics */ | |
1279 | if (skb->ip_summed == CHECKSUM_COMPLETE) | |
1280 | qs->port_stats[SGE_PSTAT_TX_CSUM]++; | |
1281 | if (skb_shinfo(skb)->gso_size) | |
1282 | qs->port_stats[SGE_PSTAT_TSO]++; | |
1283 | if (vlan_tx_tag_present(skb) && pi->vlan_grp) | |
1284 | qs->port_stats[SGE_PSTAT_VLANINS]++; | |
1285 | ||
4d22de3e DLR |
1286 | /* |
1287 | * We do not use Tx completion interrupts to free DMAd Tx packets. | |
1288 | * This is good for performamce but means that we rely on new Tx | |
1289 | * packets arriving to run the destructors of completed packets, | |
1290 | * which open up space in their sockets' send queues. Sometimes | |
1291 | * we do not get such new packets causing Tx to stall. A single | |
1292 | * UDP transmitter is a good example of this situation. We have | |
1293 | * a clean up timer that periodically reclaims completed packets | |
1294 | * but it doesn't run often enough (nor do we want it to) to prevent | |
1295 | * lengthy stalls. A solution to this problem is to run the | |
1296 | * destructor early, after the packet is queued but before it's DMAd. | |
1297 | * A cons is that we lie to socket memory accounting, but the amount | |
1298 | * of extra memory is reasonable (limited by the number of Tx | |
1299 | * descriptors), the packets do actually get freed quickly by new | |
1300 | * packets almost always, and for protocols like TCP that wait for | |
1301 | * acks to really free up the data the extra memory is even less. | |
1302 | * On the positive side we run the destructors on the sending CPU | |
1303 | * rather than on a potentially different completing CPU, usually a | |
1304 | * good thing. We also run them without holding our Tx queue lock, | |
1305 | * unlike what reclaim_completed_tx() would otherwise do. | |
1306 | * | |
1307 | * Run the destructor before telling the DMA engine about the packet | |
1308 | * to make sure it doesn't complete and get freed prematurely. | |
1309 | */ | |
1310 | if (likely(!skb_shared(skb))) | |
1311 | skb_orphan(skb); | |
1312 | ||
1313 | write_tx_pkt_wr(adap, skb, pi, pidx, gen, q, ndesc, compl); | |
1314 | check_ring_tx_db(adap, q); | |
1315 | return NETDEV_TX_OK; | |
1316 | } | |
1317 | ||
1318 | /** | |
1319 | * write_imm - write a packet into a Tx descriptor as immediate data | |
1320 | * @d: the Tx descriptor to write | |
1321 | * @skb: the packet | |
1322 | * @len: the length of packet data to write as immediate data | |
1323 | * @gen: the generation bit value to write | |
1324 | * | |
1325 | * Writes a packet as immediate data into a Tx descriptor. The packet | |
1326 | * contains a work request at its beginning. We must write the packet | |
27186dc3 DLR |
1327 | * carefully so the SGE doesn't read it accidentally before it's written |
1328 | * in its entirety. | |
4d22de3e DLR |
1329 | */ |
1330 | static inline void write_imm(struct tx_desc *d, struct sk_buff *skb, | |
1331 | unsigned int len, unsigned int gen) | |
1332 | { | |
1333 | struct work_request_hdr *from = (struct work_request_hdr *)skb->data; | |
1334 | struct work_request_hdr *to = (struct work_request_hdr *)d; | |
1335 | ||
27186dc3 DLR |
1336 | if (likely(!skb->data_len)) |
1337 | memcpy(&to[1], &from[1], len - sizeof(*from)); | |
1338 | else | |
1339 | skb_copy_bits(skb, sizeof(*from), &to[1], len - sizeof(*from)); | |
1340 | ||
4d22de3e DLR |
1341 | to->wr_hi = from->wr_hi | htonl(F_WR_SOP | F_WR_EOP | |
1342 | V_WR_BCNTLFLT(len & 7)); | |
1343 | wmb(); | |
1344 | to->wr_lo = from->wr_lo | htonl(V_WR_GEN(gen) | | |
1345 | V_WR_LEN((len + 7) / 8)); | |
1346 | wr_gen2(d, gen); | |
1347 | kfree_skb(skb); | |
1348 | } | |
1349 | ||
1350 | /** | |
1351 | * check_desc_avail - check descriptor availability on a send queue | |
1352 | * @adap: the adapter | |
1353 | * @q: the send queue | |
1354 | * @skb: the packet needing the descriptors | |
1355 | * @ndesc: the number of Tx descriptors needed | |
1356 | * @qid: the Tx queue number in its queue set (TXQ_OFLD or TXQ_CTRL) | |
1357 | * | |
1358 | * Checks if the requested number of Tx descriptors is available on an | |
1359 | * SGE send queue. If the queue is already suspended or not enough | |
1360 | * descriptors are available the packet is queued for later transmission. | |
1361 | * Must be called with the Tx queue locked. | |
1362 | * | |
1363 | * Returns 0 if enough descriptors are available, 1 if there aren't | |
1364 | * enough descriptors and the packet has been queued, and 2 if the caller | |
1365 | * needs to retry because there weren't enough descriptors at the | |
1366 | * beginning of the call but some freed up in the mean time. | |
1367 | */ | |
1368 | static inline int check_desc_avail(struct adapter *adap, struct sge_txq *q, | |
1369 | struct sk_buff *skb, unsigned int ndesc, | |
1370 | unsigned int qid) | |
1371 | { | |
1372 | if (unlikely(!skb_queue_empty(&q->sendq))) { | |
1373 | addq_exit:__skb_queue_tail(&q->sendq, skb); | |
1374 | return 1; | |
1375 | } | |
1376 | if (unlikely(q->size - q->in_use < ndesc)) { | |
1377 | struct sge_qset *qs = txq_to_qset(q, qid); | |
1378 | ||
1379 | set_bit(qid, &qs->txq_stopped); | |
1380 | smp_mb__after_clear_bit(); | |
1381 | ||
1382 | if (should_restart_tx(q) && | |
1383 | test_and_clear_bit(qid, &qs->txq_stopped)) | |
1384 | return 2; | |
1385 | ||
1386 | q->stops++; | |
1387 | goto addq_exit; | |
1388 | } | |
1389 | return 0; | |
1390 | } | |
1391 | ||
1392 | /** | |
1393 | * reclaim_completed_tx_imm - reclaim completed control-queue Tx descs | |
1394 | * @q: the SGE control Tx queue | |
1395 | * | |
1396 | * This is a variant of reclaim_completed_tx() that is used for Tx queues | |
1397 | * that send only immediate data (presently just the control queues) and | |
1398 | * thus do not have any sk_buffs to release. | |
1399 | */ | |
1400 | static inline void reclaim_completed_tx_imm(struct sge_txq *q) | |
1401 | { | |
1402 | unsigned int reclaim = q->processed - q->cleaned; | |
1403 | ||
1404 | q->in_use -= reclaim; | |
1405 | q->cleaned += reclaim; | |
1406 | } | |
1407 | ||
1408 | static inline int immediate(const struct sk_buff *skb) | |
1409 | { | |
27186dc3 | 1410 | return skb->len <= WR_LEN; |
4d22de3e DLR |
1411 | } |
1412 | ||
1413 | /** | |
1414 | * ctrl_xmit - send a packet through an SGE control Tx queue | |
1415 | * @adap: the adapter | |
1416 | * @q: the control queue | |
1417 | * @skb: the packet | |
1418 | * | |
1419 | * Send a packet through an SGE control Tx queue. Packets sent through | |
1420 | * a control queue must fit entirely as immediate data in a single Tx | |
1421 | * descriptor and have no page fragments. | |
1422 | */ | |
1423 | static int ctrl_xmit(struct adapter *adap, struct sge_txq *q, | |
1424 | struct sk_buff *skb) | |
1425 | { | |
1426 | int ret; | |
1427 | struct work_request_hdr *wrp = (struct work_request_hdr *)skb->data; | |
1428 | ||
1429 | if (unlikely(!immediate(skb))) { | |
1430 | WARN_ON(1); | |
1431 | dev_kfree_skb(skb); | |
1432 | return NET_XMIT_SUCCESS; | |
1433 | } | |
1434 | ||
1435 | wrp->wr_hi |= htonl(F_WR_SOP | F_WR_EOP); | |
1436 | wrp->wr_lo = htonl(V_WR_TID(q->token)); | |
1437 | ||
1438 | spin_lock(&q->lock); | |
1439 | again:reclaim_completed_tx_imm(q); | |
1440 | ||
1441 | ret = check_desc_avail(adap, q, skb, 1, TXQ_CTRL); | |
1442 | if (unlikely(ret)) { | |
1443 | if (ret == 1) { | |
1444 | spin_unlock(&q->lock); | |
1445 | return NET_XMIT_CN; | |
1446 | } | |
1447 | goto again; | |
1448 | } | |
1449 | ||
1450 | write_imm(&q->desc[q->pidx], skb, skb->len, q->gen); | |
1451 | ||
1452 | q->in_use++; | |
1453 | if (++q->pidx >= q->size) { | |
1454 | q->pidx = 0; | |
1455 | q->gen ^= 1; | |
1456 | } | |
1457 | spin_unlock(&q->lock); | |
1458 | wmb(); | |
1459 | t3_write_reg(adap, A_SG_KDOORBELL, | |
1460 | F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id)); | |
1461 | return NET_XMIT_SUCCESS; | |
1462 | } | |
1463 | ||
1464 | /** | |
1465 | * restart_ctrlq - restart a suspended control queue | |
1466 | * @qs: the queue set cotaining the control queue | |
1467 | * | |
1468 | * Resumes transmission on a suspended Tx control queue. | |
1469 | */ | |
1470 | static void restart_ctrlq(unsigned long data) | |
1471 | { | |
1472 | struct sk_buff *skb; | |
1473 | struct sge_qset *qs = (struct sge_qset *)data; | |
1474 | struct sge_txq *q = &qs->txq[TXQ_CTRL]; | |
4d22de3e DLR |
1475 | |
1476 | spin_lock(&q->lock); | |
1477 | again:reclaim_completed_tx_imm(q); | |
1478 | ||
bea3348e SH |
1479 | while (q->in_use < q->size && |
1480 | (skb = __skb_dequeue(&q->sendq)) != NULL) { | |
4d22de3e DLR |
1481 | |
1482 | write_imm(&q->desc[q->pidx], skb, skb->len, q->gen); | |
1483 | ||
1484 | if (++q->pidx >= q->size) { | |
1485 | q->pidx = 0; | |
1486 | q->gen ^= 1; | |
1487 | } | |
1488 | q->in_use++; | |
1489 | } | |
1490 | ||
1491 | if (!skb_queue_empty(&q->sendq)) { | |
1492 | set_bit(TXQ_CTRL, &qs->txq_stopped); | |
1493 | smp_mb__after_clear_bit(); | |
1494 | ||
1495 | if (should_restart_tx(q) && | |
1496 | test_and_clear_bit(TXQ_CTRL, &qs->txq_stopped)) | |
1497 | goto again; | |
1498 | q->stops++; | |
1499 | } | |
1500 | ||
1501 | spin_unlock(&q->lock); | |
afefce66 | 1502 | wmb(); |
bea3348e | 1503 | t3_write_reg(qs->adap, A_SG_KDOORBELL, |
4d22de3e DLR |
1504 | F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id)); |
1505 | } | |
1506 | ||
14ab9892 DLR |
1507 | /* |
1508 | * Send a management message through control queue 0 | |
1509 | */ | |
1510 | int t3_mgmt_tx(struct adapter *adap, struct sk_buff *skb) | |
1511 | { | |
204e2f98 | 1512 | int ret; |
bc4b6b52 DLR |
1513 | local_bh_disable(); |
1514 | ret = ctrl_xmit(adap, &adap->sge.qs[0].txq[TXQ_CTRL], skb); | |
1515 | local_bh_enable(); | |
1516 | ||
1517 | return ret; | |
14ab9892 DLR |
1518 | } |
1519 | ||
99d7cf30 DLR |
1520 | /** |
1521 | * deferred_unmap_destructor - unmap a packet when it is freed | |
1522 | * @skb: the packet | |
1523 | * | |
1524 | * This is the packet destructor used for Tx packets that need to remain | |
1525 | * mapped until they are freed rather than until their Tx descriptors are | |
1526 | * freed. | |
1527 | */ | |
1528 | static void deferred_unmap_destructor(struct sk_buff *skb) | |
1529 | { | |
1530 | int i; | |
1531 | const dma_addr_t *p; | |
1532 | const struct skb_shared_info *si; | |
1533 | const struct deferred_unmap_info *dui; | |
99d7cf30 DLR |
1534 | |
1535 | dui = (struct deferred_unmap_info *)skb->head; | |
1536 | p = dui->addr; | |
1537 | ||
23561c94 DLR |
1538 | if (skb->tail - skb->transport_header) |
1539 | pci_unmap_single(dui->pdev, *p++, | |
1540 | skb->tail - skb->transport_header, | |
1541 | PCI_DMA_TODEVICE); | |
99d7cf30 DLR |
1542 | |
1543 | si = skb_shinfo(skb); | |
1544 | for (i = 0; i < si->nr_frags; i++) | |
1545 | pci_unmap_page(dui->pdev, *p++, si->frags[i].size, | |
1546 | PCI_DMA_TODEVICE); | |
1547 | } | |
1548 | ||
1549 | static void setup_deferred_unmapping(struct sk_buff *skb, struct pci_dev *pdev, | |
1550 | const struct sg_ent *sgl, int sgl_flits) | |
1551 | { | |
1552 | dma_addr_t *p; | |
1553 | struct deferred_unmap_info *dui; | |
1554 | ||
1555 | dui = (struct deferred_unmap_info *)skb->head; | |
1556 | dui->pdev = pdev; | |
1557 | for (p = dui->addr; sgl_flits >= 3; sgl++, sgl_flits -= 3) { | |
1558 | *p++ = be64_to_cpu(sgl->addr[0]); | |
1559 | *p++ = be64_to_cpu(sgl->addr[1]); | |
1560 | } | |
1561 | if (sgl_flits) | |
1562 | *p = be64_to_cpu(sgl->addr[0]); | |
1563 | } | |
1564 | ||
4d22de3e DLR |
1565 | /** |
1566 | * write_ofld_wr - write an offload work request | |
1567 | * @adap: the adapter | |
1568 | * @skb: the packet to send | |
1569 | * @q: the Tx queue | |
1570 | * @pidx: index of the first Tx descriptor to write | |
1571 | * @gen: the generation value to use | |
1572 | * @ndesc: number of descriptors the packet will occupy | |
1573 | * | |
1574 | * Write an offload work request to send the supplied packet. The packet | |
1575 | * data already carry the work request with most fields populated. | |
1576 | */ | |
1577 | static void write_ofld_wr(struct adapter *adap, struct sk_buff *skb, | |
1578 | struct sge_txq *q, unsigned int pidx, | |
1579 | unsigned int gen, unsigned int ndesc) | |
1580 | { | |
1581 | unsigned int sgl_flits, flits; | |
1582 | struct work_request_hdr *from; | |
1583 | struct sg_ent *sgp, sgl[MAX_SKB_FRAGS / 2 + 1]; | |
1584 | struct tx_desc *d = &q->desc[pidx]; | |
1585 | ||
1586 | if (immediate(skb)) { | |
1587 | q->sdesc[pidx].skb = NULL; | |
1588 | write_imm(d, skb, skb->len, gen); | |
1589 | return; | |
1590 | } | |
1591 | ||
1592 | /* Only TX_DATA builds SGLs */ | |
1593 | ||
1594 | from = (struct work_request_hdr *)skb->data; | |
ea2ae17d ACM |
1595 | memcpy(&d->flit[1], &from[1], |
1596 | skb_transport_offset(skb) - sizeof(*from)); | |
4d22de3e | 1597 | |
ea2ae17d | 1598 | flits = skb_transport_offset(skb) / 8; |
4d22de3e | 1599 | sgp = ndesc == 1 ? (struct sg_ent *)&d->flit[flits] : sgl; |
9c70220b | 1600 | sgl_flits = make_sgl(skb, sgp, skb_transport_header(skb), |
27a884dc | 1601 | skb->tail - skb->transport_header, |
4d22de3e | 1602 | adap->pdev); |
99d7cf30 DLR |
1603 | if (need_skb_unmap()) { |
1604 | setup_deferred_unmapping(skb, adap->pdev, sgp, sgl_flits); | |
1605 | skb->destructor = deferred_unmap_destructor; | |
99d7cf30 | 1606 | } |
4d22de3e DLR |
1607 | |
1608 | write_wr_hdr_sgl(ndesc, skb, d, pidx, q, sgl, flits, sgl_flits, | |
1609 | gen, from->wr_hi, from->wr_lo); | |
1610 | } | |
1611 | ||
1612 | /** | |
1613 | * calc_tx_descs_ofld - calculate # of Tx descriptors for an offload packet | |
1614 | * @skb: the packet | |
1615 | * | |
1616 | * Returns the number of Tx descriptors needed for the given offload | |
1617 | * packet. These packets are already fully constructed. | |
1618 | */ | |
1619 | static inline unsigned int calc_tx_descs_ofld(const struct sk_buff *skb) | |
1620 | { | |
27186dc3 | 1621 | unsigned int flits, cnt; |
4d22de3e | 1622 | |
27186dc3 | 1623 | if (skb->len <= WR_LEN) |
4d22de3e DLR |
1624 | return 1; /* packet fits as immediate data */ |
1625 | ||
ea2ae17d | 1626 | flits = skb_transport_offset(skb) / 8; /* headers */ |
27186dc3 | 1627 | cnt = skb_shinfo(skb)->nr_frags; |
27a884dc | 1628 | if (skb->tail != skb->transport_header) |
4d22de3e DLR |
1629 | cnt++; |
1630 | return flits_to_desc(flits + sgl_len(cnt)); | |
1631 | } | |
1632 | ||
1633 | /** | |
1634 | * ofld_xmit - send a packet through an offload queue | |
1635 | * @adap: the adapter | |
1636 | * @q: the Tx offload queue | |
1637 | * @skb: the packet | |
1638 | * | |
1639 | * Send an offload packet through an SGE offload queue. | |
1640 | */ | |
1641 | static int ofld_xmit(struct adapter *adap, struct sge_txq *q, | |
1642 | struct sk_buff *skb) | |
1643 | { | |
1644 | int ret; | |
1645 | unsigned int ndesc = calc_tx_descs_ofld(skb), pidx, gen; | |
1646 | ||
1647 | spin_lock(&q->lock); | |
42c8ea17 | 1648 | again: reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK); |
4d22de3e DLR |
1649 | |
1650 | ret = check_desc_avail(adap, q, skb, ndesc, TXQ_OFLD); | |
1651 | if (unlikely(ret)) { | |
1652 | if (ret == 1) { | |
1653 | skb->priority = ndesc; /* save for restart */ | |
1654 | spin_unlock(&q->lock); | |
1655 | return NET_XMIT_CN; | |
1656 | } | |
1657 | goto again; | |
1658 | } | |
1659 | ||
1660 | gen = q->gen; | |
1661 | q->in_use += ndesc; | |
1662 | pidx = q->pidx; | |
1663 | q->pidx += ndesc; | |
1664 | if (q->pidx >= q->size) { | |
1665 | q->pidx -= q->size; | |
1666 | q->gen ^= 1; | |
1667 | } | |
1668 | spin_unlock(&q->lock); | |
1669 | ||
1670 | write_ofld_wr(adap, skb, q, pidx, gen, ndesc); | |
1671 | check_ring_tx_db(adap, q); | |
1672 | return NET_XMIT_SUCCESS; | |
1673 | } | |
1674 | ||
1675 | /** | |
1676 | * restart_offloadq - restart a suspended offload queue | |
1677 | * @qs: the queue set cotaining the offload queue | |
1678 | * | |
1679 | * Resumes transmission on a suspended Tx offload queue. | |
1680 | */ | |
1681 | static void restart_offloadq(unsigned long data) | |
1682 | { | |
1683 | struct sk_buff *skb; | |
1684 | struct sge_qset *qs = (struct sge_qset *)data; | |
1685 | struct sge_txq *q = &qs->txq[TXQ_OFLD]; | |
5fbf816f DLR |
1686 | const struct port_info *pi = netdev_priv(qs->netdev); |
1687 | struct adapter *adap = pi->adapter; | |
4d22de3e DLR |
1688 | |
1689 | spin_lock(&q->lock); | |
42c8ea17 | 1690 | again: reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK); |
4d22de3e DLR |
1691 | |
1692 | while ((skb = skb_peek(&q->sendq)) != NULL) { | |
1693 | unsigned int gen, pidx; | |
1694 | unsigned int ndesc = skb->priority; | |
1695 | ||
1696 | if (unlikely(q->size - q->in_use < ndesc)) { | |
1697 | set_bit(TXQ_OFLD, &qs->txq_stopped); | |
1698 | smp_mb__after_clear_bit(); | |
1699 | ||
1700 | if (should_restart_tx(q) && | |
1701 | test_and_clear_bit(TXQ_OFLD, &qs->txq_stopped)) | |
1702 | goto again; | |
1703 | q->stops++; | |
1704 | break; | |
1705 | } | |
1706 | ||
1707 | gen = q->gen; | |
1708 | q->in_use += ndesc; | |
1709 | pidx = q->pidx; | |
1710 | q->pidx += ndesc; | |
1711 | if (q->pidx >= q->size) { | |
1712 | q->pidx -= q->size; | |
1713 | q->gen ^= 1; | |
1714 | } | |
1715 | __skb_unlink(skb, &q->sendq); | |
1716 | spin_unlock(&q->lock); | |
1717 | ||
1718 | write_ofld_wr(adap, skb, q, pidx, gen, ndesc); | |
1719 | spin_lock(&q->lock); | |
1720 | } | |
1721 | spin_unlock(&q->lock); | |
1722 | ||
1723 | #if USE_GTS | |
1724 | set_bit(TXQ_RUNNING, &q->flags); | |
1725 | set_bit(TXQ_LAST_PKT_DB, &q->flags); | |
1726 | #endif | |
afefce66 | 1727 | wmb(); |
4d22de3e DLR |
1728 | t3_write_reg(adap, A_SG_KDOORBELL, |
1729 | F_SELEGRCNTX | V_EGRCNTX(q->cntxt_id)); | |
1730 | } | |
1731 | ||
1732 | /** | |
1733 | * queue_set - return the queue set a packet should use | |
1734 | * @skb: the packet | |
1735 | * | |
1736 | * Maps a packet to the SGE queue set it should use. The desired queue | |
1737 | * set is carried in bits 1-3 in the packet's priority. | |
1738 | */ | |
1739 | static inline int queue_set(const struct sk_buff *skb) | |
1740 | { | |
1741 | return skb->priority >> 1; | |
1742 | } | |
1743 | ||
1744 | /** | |
1745 | * is_ctrl_pkt - return whether an offload packet is a control packet | |
1746 | * @skb: the packet | |
1747 | * | |
1748 | * Determines whether an offload packet should use an OFLD or a CTRL | |
1749 | * Tx queue. This is indicated by bit 0 in the packet's priority. | |
1750 | */ | |
1751 | static inline int is_ctrl_pkt(const struct sk_buff *skb) | |
1752 | { | |
1753 | return skb->priority & 1; | |
1754 | } | |
1755 | ||
1756 | /** | |
1757 | * t3_offload_tx - send an offload packet | |
1758 | * @tdev: the offload device to send to | |
1759 | * @skb: the packet | |
1760 | * | |
1761 | * Sends an offload packet. We use the packet priority to select the | |
1762 | * appropriate Tx queue as follows: bit 0 indicates whether the packet | |
1763 | * should be sent as regular or control, bits 1-3 select the queue set. | |
1764 | */ | |
1765 | int t3_offload_tx(struct t3cdev *tdev, struct sk_buff *skb) | |
1766 | { | |
1767 | struct adapter *adap = tdev2adap(tdev); | |
1768 | struct sge_qset *qs = &adap->sge.qs[queue_set(skb)]; | |
1769 | ||
1770 | if (unlikely(is_ctrl_pkt(skb))) | |
1771 | return ctrl_xmit(adap, &qs->txq[TXQ_CTRL], skb); | |
1772 | ||
1773 | return ofld_xmit(adap, &qs->txq[TXQ_OFLD], skb); | |
1774 | } | |
1775 | ||
1776 | /** | |
1777 | * offload_enqueue - add an offload packet to an SGE offload receive queue | |
1778 | * @q: the SGE response queue | |
1779 | * @skb: the packet | |
1780 | * | |
1781 | * Add a new offload packet to an SGE response queue's offload packet | |
1782 | * queue. If the packet is the first on the queue it schedules the RX | |
1783 | * softirq to process the queue. | |
1784 | */ | |
1785 | static inline void offload_enqueue(struct sge_rspq *q, struct sk_buff *skb) | |
1786 | { | |
147e70e6 DM |
1787 | int was_empty = skb_queue_empty(&q->rx_queue); |
1788 | ||
1789 | __skb_queue_tail(&q->rx_queue, skb); | |
1790 | ||
1791 | if (was_empty) { | |
4d22de3e DLR |
1792 | struct sge_qset *qs = rspq_to_qset(q); |
1793 | ||
bea3348e | 1794 | napi_schedule(&qs->napi); |
4d22de3e | 1795 | } |
4d22de3e DLR |
1796 | } |
1797 | ||
1798 | /** | |
1799 | * deliver_partial_bundle - deliver a (partial) bundle of Rx offload pkts | |
1800 | * @tdev: the offload device that will be receiving the packets | |
1801 | * @q: the SGE response queue that assembled the bundle | |
1802 | * @skbs: the partial bundle | |
1803 | * @n: the number of packets in the bundle | |
1804 | * | |
1805 | * Delivers a (partial) bundle of Rx offload packets to an offload device. | |
1806 | */ | |
1807 | static inline void deliver_partial_bundle(struct t3cdev *tdev, | |
1808 | struct sge_rspq *q, | |
1809 | struct sk_buff *skbs[], int n) | |
1810 | { | |
1811 | if (n) { | |
1812 | q->offload_bundles++; | |
1813 | tdev->recv(tdev, skbs, n); | |
1814 | } | |
1815 | } | |
1816 | ||
1817 | /** | |
1818 | * ofld_poll - NAPI handler for offload packets in interrupt mode | |
1819 | * @dev: the network device doing the polling | |
1820 | * @budget: polling budget | |
1821 | * | |
1822 | * The NAPI handler for offload packets when a response queue is serviced | |
1823 | * by the hard interrupt handler, i.e., when it's operating in non-polling | |
1824 | * mode. Creates small packet batches and sends them through the offload | |
1825 | * receive handler. Batches need to be of modest size as we do prefetches | |
1826 | * on the packets in each. | |
1827 | */ | |
bea3348e | 1828 | static int ofld_poll(struct napi_struct *napi, int budget) |
4d22de3e | 1829 | { |
bea3348e | 1830 | struct sge_qset *qs = container_of(napi, struct sge_qset, napi); |
4d22de3e | 1831 | struct sge_rspq *q = &qs->rspq; |
bea3348e SH |
1832 | struct adapter *adapter = qs->adap; |
1833 | int work_done = 0; | |
4d22de3e | 1834 | |
bea3348e | 1835 | while (work_done < budget) { |
147e70e6 DM |
1836 | struct sk_buff *skb, *tmp, *skbs[RX_BUNDLE_SIZE]; |
1837 | struct sk_buff_head queue; | |
4d22de3e DLR |
1838 | int ngathered; |
1839 | ||
1840 | spin_lock_irq(&q->lock); | |
147e70e6 DM |
1841 | __skb_queue_head_init(&queue); |
1842 | skb_queue_splice_init(&q->rx_queue, &queue); | |
1843 | if (skb_queue_empty(&queue)) { | |
bea3348e | 1844 | napi_complete(napi); |
4d22de3e | 1845 | spin_unlock_irq(&q->lock); |
bea3348e | 1846 | return work_done; |
4d22de3e | 1847 | } |
4d22de3e DLR |
1848 | spin_unlock_irq(&q->lock); |
1849 | ||
147e70e6 DM |
1850 | ngathered = 0; |
1851 | skb_queue_walk_safe(&queue, skb, tmp) { | |
1852 | if (work_done >= budget) | |
1853 | break; | |
1854 | work_done++; | |
1855 | ||
1856 | __skb_unlink(skb, &queue); | |
1857 | prefetch(skb->data); | |
1858 | skbs[ngathered] = skb; | |
4d22de3e DLR |
1859 | if (++ngathered == RX_BUNDLE_SIZE) { |
1860 | q->offload_bundles++; | |
1861 | adapter->tdev.recv(&adapter->tdev, skbs, | |
1862 | ngathered); | |
1863 | ngathered = 0; | |
1864 | } | |
1865 | } | |
147e70e6 DM |
1866 | if (!skb_queue_empty(&queue)) { |
1867 | /* splice remaining packets back onto Rx queue */ | |
4d22de3e | 1868 | spin_lock_irq(&q->lock); |
147e70e6 | 1869 | skb_queue_splice(&queue, &q->rx_queue); |
4d22de3e DLR |
1870 | spin_unlock_irq(&q->lock); |
1871 | } | |
1872 | deliver_partial_bundle(&adapter->tdev, q, skbs, ngathered); | |
1873 | } | |
bea3348e SH |
1874 | |
1875 | return work_done; | |
4d22de3e DLR |
1876 | } |
1877 | ||
1878 | /** | |
1879 | * rx_offload - process a received offload packet | |
1880 | * @tdev: the offload device receiving the packet | |
1881 | * @rq: the response queue that received the packet | |
1882 | * @skb: the packet | |
1883 | * @rx_gather: a gather list of packets if we are building a bundle | |
1884 | * @gather_idx: index of the next available slot in the bundle | |
1885 | * | |
1886 | * Process an ingress offload pakcet and add it to the offload ingress | |
1887 | * queue. Returns the index of the next available slot in the bundle. | |
1888 | */ | |
1889 | static inline int rx_offload(struct t3cdev *tdev, struct sge_rspq *rq, | |
1890 | struct sk_buff *skb, struct sk_buff *rx_gather[], | |
1891 | unsigned int gather_idx) | |
1892 | { | |
459a98ed | 1893 | skb_reset_mac_header(skb); |
c1d2bbe1 | 1894 | skb_reset_network_header(skb); |
badff6d0 | 1895 | skb_reset_transport_header(skb); |
4d22de3e DLR |
1896 | |
1897 | if (rq->polling) { | |
1898 | rx_gather[gather_idx++] = skb; | |
1899 | if (gather_idx == RX_BUNDLE_SIZE) { | |
1900 | tdev->recv(tdev, rx_gather, RX_BUNDLE_SIZE); | |
1901 | gather_idx = 0; | |
1902 | rq->offload_bundles++; | |
1903 | } | |
1904 | } else | |
1905 | offload_enqueue(rq, skb); | |
1906 | ||
1907 | return gather_idx; | |
1908 | } | |
1909 | ||
4d22de3e DLR |
1910 | /** |
1911 | * restart_tx - check whether to restart suspended Tx queues | |
1912 | * @qs: the queue set to resume | |
1913 | * | |
1914 | * Restarts suspended Tx queues of an SGE queue set if they have enough | |
1915 | * free resources to resume operation. | |
1916 | */ | |
1917 | static void restart_tx(struct sge_qset *qs) | |
1918 | { | |
1919 | if (test_bit(TXQ_ETH, &qs->txq_stopped) && | |
1920 | should_restart_tx(&qs->txq[TXQ_ETH]) && | |
1921 | test_and_clear_bit(TXQ_ETH, &qs->txq_stopped)) { | |
1922 | qs->txq[TXQ_ETH].restarts++; | |
1923 | if (netif_running(qs->netdev)) | |
82ad3329 | 1924 | netif_tx_wake_queue(qs->tx_q); |
4d22de3e DLR |
1925 | } |
1926 | ||
1927 | if (test_bit(TXQ_OFLD, &qs->txq_stopped) && | |
1928 | should_restart_tx(&qs->txq[TXQ_OFLD]) && | |
1929 | test_and_clear_bit(TXQ_OFLD, &qs->txq_stopped)) { | |
1930 | qs->txq[TXQ_OFLD].restarts++; | |
1931 | tasklet_schedule(&qs->txq[TXQ_OFLD].qresume_tsk); | |
1932 | } | |
1933 | if (test_bit(TXQ_CTRL, &qs->txq_stopped) && | |
1934 | should_restart_tx(&qs->txq[TXQ_CTRL]) && | |
1935 | test_and_clear_bit(TXQ_CTRL, &qs->txq_stopped)) { | |
1936 | qs->txq[TXQ_CTRL].restarts++; | |
1937 | tasklet_schedule(&qs->txq[TXQ_CTRL].qresume_tsk); | |
1938 | } | |
1939 | } | |
1940 | ||
a109a5b9 KX |
1941 | /** |
1942 | * cxgb3_arp_process - process an ARP request probing a private IP address | |
1943 | * @adapter: the adapter | |
1944 | * @skb: the skbuff containing the ARP request | |
1945 | * | |
1946 | * Check if the ARP request is probing the private IP address | |
1947 | * dedicated to iSCSI, generate an ARP reply if so. | |
1948 | */ | |
1949 | static void cxgb3_arp_process(struct adapter *adapter, struct sk_buff *skb) | |
1950 | { | |
1951 | struct net_device *dev = skb->dev; | |
1952 | struct port_info *pi; | |
1953 | struct arphdr *arp; | |
1954 | unsigned char *arp_ptr; | |
1955 | unsigned char *sha; | |
1956 | __be32 sip, tip; | |
1957 | ||
1958 | if (!dev) | |
1959 | return; | |
1960 | ||
1961 | skb_reset_network_header(skb); | |
1962 | arp = arp_hdr(skb); | |
1963 | ||
1964 | if (arp->ar_op != htons(ARPOP_REQUEST)) | |
1965 | return; | |
1966 | ||
1967 | arp_ptr = (unsigned char *)(arp + 1); | |
1968 | sha = arp_ptr; | |
1969 | arp_ptr += dev->addr_len; | |
1970 | memcpy(&sip, arp_ptr, sizeof(sip)); | |
1971 | arp_ptr += sizeof(sip); | |
1972 | arp_ptr += dev->addr_len; | |
1973 | memcpy(&tip, arp_ptr, sizeof(tip)); | |
1974 | ||
1975 | pi = netdev_priv(dev); | |
1976 | if (tip != pi->iscsi_ipv4addr) | |
1977 | return; | |
1978 | ||
1979 | arp_send(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip, sha, | |
1980 | dev->dev_addr, sha); | |
1981 | ||
1982 | } | |
1983 | ||
1984 | static inline int is_arp(struct sk_buff *skb) | |
1985 | { | |
1986 | return skb->protocol == htons(ETH_P_ARP); | |
1987 | } | |
1988 | ||
4d22de3e DLR |
1989 | /** |
1990 | * rx_eth - process an ingress ethernet packet | |
1991 | * @adap: the adapter | |
1992 | * @rq: the response queue that received the packet | |
1993 | * @skb: the packet | |
1994 | * @pad: amount of padding at the start of the buffer | |
1995 | * | |
1996 | * Process an ingress ethernet pakcet and deliver it to the stack. | |
1997 | * The padding is 2 if the packet was delivered in an Rx buffer and 0 | |
1998 | * if it was immediate data in a response. | |
1999 | */ | |
2000 | static void rx_eth(struct adapter *adap, struct sge_rspq *rq, | |
b47385bd | 2001 | struct sk_buff *skb, int pad, int lro) |
4d22de3e DLR |
2002 | { |
2003 | struct cpl_rx_pkt *p = (struct cpl_rx_pkt *)(skb->data + pad); | |
b47385bd | 2004 | struct sge_qset *qs = rspq_to_qset(rq); |
4d22de3e DLR |
2005 | struct port_info *pi; |
2006 | ||
4d22de3e | 2007 | skb_pull(skb, sizeof(*p) + pad); |
4c13eb66 | 2008 | skb->protocol = eth_type_trans(skb, adap->port[p->iff]); |
4d22de3e | 2009 | pi = netdev_priv(skb->dev); |
5e68b772 DLR |
2010 | if ((pi->rx_offload & T3_RX_CSUM) && p->csum_valid && |
2011 | p->csum == htons(0xffff) && !p->fragment) { | |
a109a5b9 | 2012 | qs->port_stats[SGE_PSTAT_RX_CSUM_GOOD]++; |
4d22de3e DLR |
2013 | skb->ip_summed = CHECKSUM_UNNECESSARY; |
2014 | } else | |
2015 | skb->ip_summed = CHECKSUM_NONE; | |
0c8dfc83 | 2016 | skb_record_rx_queue(skb, qs - &adap->sge.qs[0]); |
4d22de3e DLR |
2017 | |
2018 | if (unlikely(p->vlan_valid)) { | |
2019 | struct vlan_group *grp = pi->vlan_grp; | |
2020 | ||
b47385bd | 2021 | qs->port_stats[SGE_PSTAT_VLANEX]++; |
4d22de3e | 2022 | if (likely(grp)) |
b47385bd | 2023 | if (lro) |
7be2df45 HX |
2024 | vlan_gro_receive(&qs->napi, grp, |
2025 | ntohs(p->vlan), skb); | |
a109a5b9 KX |
2026 | else { |
2027 | if (unlikely(pi->iscsi_ipv4addr && | |
2028 | is_arp(skb))) { | |
2029 | unsigned short vtag = ntohs(p->vlan) & | |
2030 | VLAN_VID_MASK; | |
2031 | skb->dev = vlan_group_get_device(grp, | |
2032 | vtag); | |
2033 | cxgb3_arp_process(adap, skb); | |
2034 | } | |
b47385bd DLR |
2035 | __vlan_hwaccel_rx(skb, grp, ntohs(p->vlan), |
2036 | rq->polling); | |
a109a5b9 | 2037 | } |
4d22de3e DLR |
2038 | else |
2039 | dev_kfree_skb_any(skb); | |
b47385bd DLR |
2040 | } else if (rq->polling) { |
2041 | if (lro) | |
7be2df45 | 2042 | napi_gro_receive(&qs->napi, skb); |
a109a5b9 KX |
2043 | else { |
2044 | if (unlikely(pi->iscsi_ipv4addr && is_arp(skb))) | |
2045 | cxgb3_arp_process(adap, skb); | |
b47385bd | 2046 | netif_receive_skb(skb); |
a109a5b9 | 2047 | } |
b47385bd | 2048 | } else |
4d22de3e DLR |
2049 | netif_rx(skb); |
2050 | } | |
2051 | ||
b47385bd DLR |
2052 | static inline int is_eth_tcp(u32 rss) |
2053 | { | |
2054 | return G_HASHTYPE(ntohl(rss)) == RSS_HASH_4_TUPLE; | |
2055 | } | |
2056 | ||
b47385bd DLR |
2057 | /** |
2058 | * lro_add_page - add a page chunk to an LRO session | |
2059 | * @adap: the adapter | |
2060 | * @qs: the associated queue set | |
2061 | * @fl: the free list containing the page chunk to add | |
2062 | * @len: packet length | |
2063 | * @complete: Indicates the last fragment of a frame | |
2064 | * | |
2065 | * Add a received packet contained in a page chunk to an existing LRO | |
2066 | * session. | |
2067 | */ | |
2068 | static void lro_add_page(struct adapter *adap, struct sge_qset *qs, | |
2069 | struct sge_fl *fl, int len, int complete) | |
2070 | { | |
2071 | struct rx_sw_desc *sd = &fl->sdesc[fl->cidx]; | |
76620aaf | 2072 | struct sk_buff *skb = NULL; |
b47385bd | 2073 | struct cpl_rx_pkt *cpl; |
76620aaf HX |
2074 | struct skb_frag_struct *rx_frag; |
2075 | int nr_frags; | |
b47385bd DLR |
2076 | int offset = 0; |
2077 | ||
76620aaf HX |
2078 | if (!qs->nomem) { |
2079 | skb = napi_get_frags(&qs->napi); | |
2080 | qs->nomem = !skb; | |
b47385bd DLR |
2081 | } |
2082 | ||
2083 | fl->credits--; | |
2084 | ||
5e68b772 DLR |
2085 | pci_dma_sync_single_for_cpu(adap->pdev, |
2086 | pci_unmap_addr(sd, dma_addr), | |
2087 | fl->buf_size - SGE_PG_RSVD, | |
2088 | PCI_DMA_FROMDEVICE); | |
2089 | ||
2090 | (*sd->pg_chunk.p_cnt)--; | |
2091 | if (!*sd->pg_chunk.p_cnt) | |
2092 | pci_unmap_page(adap->pdev, | |
10b6d956 | 2093 | sd->pg_chunk.mapping, |
5e68b772 DLR |
2094 | fl->alloc_size, |
2095 | PCI_DMA_FROMDEVICE); | |
b47385bd | 2096 | |
76620aaf HX |
2097 | if (!skb) { |
2098 | put_page(sd->pg_chunk.page); | |
2099 | if (complete) | |
2100 | qs->nomem = 0; | |
2101 | return; | |
2102 | } | |
2103 | ||
2104 | rx_frag = skb_shinfo(skb)->frags; | |
2105 | nr_frags = skb_shinfo(skb)->nr_frags; | |
2106 | ||
2107 | if (!nr_frags) { | |
2108 | offset = 2 + sizeof(struct cpl_rx_pkt); | |
2109 | qs->lro_va = sd->pg_chunk.va + 2; | |
2110 | } | |
2111 | len -= offset; | |
2112 | ||
5e68b772 | 2113 | prefetch(qs->lro_va); |
b2b964f0 | 2114 | |
b47385bd DLR |
2115 | rx_frag += nr_frags; |
2116 | rx_frag->page = sd->pg_chunk.page; | |
2117 | rx_frag->page_offset = sd->pg_chunk.offset + offset; | |
2118 | rx_frag->size = len; | |
b47385bd | 2119 | |
76620aaf HX |
2120 | skb->len += len; |
2121 | skb->data_len += len; | |
2122 | skb->truesize += len; | |
2123 | skb_shinfo(skb)->nr_frags++; | |
5e68b772 | 2124 | |
b47385bd DLR |
2125 | if (!complete) |
2126 | return; | |
2127 | ||
76620aaf | 2128 | skb->ip_summed = CHECKSUM_UNNECESSARY; |
b47385bd DLR |
2129 | cpl = qs->lro_va; |
2130 | ||
2131 | if (unlikely(cpl->vlan_valid)) { | |
2132 | struct net_device *dev = qs->netdev; | |
2133 | struct port_info *pi = netdev_priv(dev); | |
2134 | struct vlan_group *grp = pi->vlan_grp; | |
2135 | ||
2136 | if (likely(grp != NULL)) { | |
76620aaf HX |
2137 | vlan_gro_frags(&qs->napi, grp, ntohs(cpl->vlan)); |
2138 | return; | |
b47385bd DLR |
2139 | } |
2140 | } | |
76620aaf | 2141 | napi_gro_frags(&qs->napi); |
b47385bd DLR |
2142 | } |
2143 | ||
4d22de3e DLR |
2144 | /** |
2145 | * handle_rsp_cntrl_info - handles control information in a response | |
2146 | * @qs: the queue set corresponding to the response | |
2147 | * @flags: the response control flags | |
4d22de3e DLR |
2148 | * |
2149 | * Handles the control information of an SGE response, such as GTS | |
2150 | * indications and completion credits for the queue set's Tx queues. | |
6195c71d | 2151 | * HW coalesces credits, we don't do any extra SW coalescing. |
4d22de3e | 2152 | */ |
6195c71d | 2153 | static inline void handle_rsp_cntrl_info(struct sge_qset *qs, u32 flags) |
4d22de3e DLR |
2154 | { |
2155 | unsigned int credits; | |
2156 | ||
2157 | #if USE_GTS | |
2158 | if (flags & F_RSPD_TXQ0_GTS) | |
2159 | clear_bit(TXQ_RUNNING, &qs->txq[TXQ_ETH].flags); | |
2160 | #endif | |
2161 | ||
4d22de3e DLR |
2162 | credits = G_RSPD_TXQ0_CR(flags); |
2163 | if (credits) | |
2164 | qs->txq[TXQ_ETH].processed += credits; | |
2165 | ||
6195c71d DLR |
2166 | credits = G_RSPD_TXQ2_CR(flags); |
2167 | if (credits) | |
2168 | qs->txq[TXQ_CTRL].processed += credits; | |
2169 | ||
4d22de3e DLR |
2170 | # if USE_GTS |
2171 | if (flags & F_RSPD_TXQ1_GTS) | |
2172 | clear_bit(TXQ_RUNNING, &qs->txq[TXQ_OFLD].flags); | |
2173 | # endif | |
6195c71d DLR |
2174 | credits = G_RSPD_TXQ1_CR(flags); |
2175 | if (credits) | |
2176 | qs->txq[TXQ_OFLD].processed += credits; | |
4d22de3e DLR |
2177 | } |
2178 | ||
2179 | /** | |
2180 | * check_ring_db - check if we need to ring any doorbells | |
2181 | * @adapter: the adapter | |
2182 | * @qs: the queue set whose Tx queues are to be examined | |
2183 | * @sleeping: indicates which Tx queue sent GTS | |
2184 | * | |
2185 | * Checks if some of a queue set's Tx queues need to ring their doorbells | |
2186 | * to resume transmission after idling while they still have unprocessed | |
2187 | * descriptors. | |
2188 | */ | |
2189 | static void check_ring_db(struct adapter *adap, struct sge_qset *qs, | |
2190 | unsigned int sleeping) | |
2191 | { | |
2192 | if (sleeping & F_RSPD_TXQ0_GTS) { | |
2193 | struct sge_txq *txq = &qs->txq[TXQ_ETH]; | |
2194 | ||
2195 | if (txq->cleaned + txq->in_use != txq->processed && | |
2196 | !test_and_set_bit(TXQ_LAST_PKT_DB, &txq->flags)) { | |
2197 | set_bit(TXQ_RUNNING, &txq->flags); | |
2198 | t3_write_reg(adap, A_SG_KDOORBELL, F_SELEGRCNTX | | |
2199 | V_EGRCNTX(txq->cntxt_id)); | |
2200 | } | |
2201 | } | |
2202 | ||
2203 | if (sleeping & F_RSPD_TXQ1_GTS) { | |
2204 | struct sge_txq *txq = &qs->txq[TXQ_OFLD]; | |
2205 | ||
2206 | if (txq->cleaned + txq->in_use != txq->processed && | |
2207 | !test_and_set_bit(TXQ_LAST_PKT_DB, &txq->flags)) { | |
2208 | set_bit(TXQ_RUNNING, &txq->flags); | |
2209 | t3_write_reg(adap, A_SG_KDOORBELL, F_SELEGRCNTX | | |
2210 | V_EGRCNTX(txq->cntxt_id)); | |
2211 | } | |
2212 | } | |
2213 | } | |
2214 | ||
2215 | /** | |
2216 | * is_new_response - check if a response is newly written | |
2217 | * @r: the response descriptor | |
2218 | * @q: the response queue | |
2219 | * | |
2220 | * Returns true if a response descriptor contains a yet unprocessed | |
2221 | * response. | |
2222 | */ | |
2223 | static inline int is_new_response(const struct rsp_desc *r, | |
2224 | const struct sge_rspq *q) | |
2225 | { | |
2226 | return (r->intr_gen & F_RSPD_GEN2) == q->gen; | |
2227 | } | |
2228 | ||
7385ecf3 DLR |
2229 | static inline void clear_rspq_bufstate(struct sge_rspq * const q) |
2230 | { | |
2231 | q->pg_skb = NULL; | |
2232 | q->rx_recycle_buf = 0; | |
2233 | } | |
2234 | ||
4d22de3e DLR |
2235 | #define RSPD_GTS_MASK (F_RSPD_TXQ0_GTS | F_RSPD_TXQ1_GTS) |
2236 | #define RSPD_CTRL_MASK (RSPD_GTS_MASK | \ | |
2237 | V_RSPD_TXQ0_CR(M_RSPD_TXQ0_CR) | \ | |
2238 | V_RSPD_TXQ1_CR(M_RSPD_TXQ1_CR) | \ | |
2239 | V_RSPD_TXQ2_CR(M_RSPD_TXQ2_CR)) | |
2240 | ||
2241 | /* How long to delay the next interrupt in case of memory shortage, in 0.1us. */ | |
2242 | #define NOMEM_INTR_DELAY 2500 | |
2243 | ||
2244 | /** | |
2245 | * process_responses - process responses from an SGE response queue | |
2246 | * @adap: the adapter | |
2247 | * @qs: the queue set to which the response queue belongs | |
2248 | * @budget: how many responses can be processed in this round | |
2249 | * | |
2250 | * Process responses from an SGE response queue up to the supplied budget. | |
2251 | * Responses include received packets as well as credits and other events | |
2252 | * for the queues that belong to the response queue's queue set. | |
2253 | * A negative budget is effectively unlimited. | |
2254 | * | |
2255 | * Additionally choose the interrupt holdoff time for the next interrupt | |
2256 | * on this queue. If the system is under memory shortage use a fairly | |
2257 | * long delay to help recovery. | |
2258 | */ | |
2259 | static int process_responses(struct adapter *adap, struct sge_qset *qs, | |
2260 | int budget) | |
2261 | { | |
2262 | struct sge_rspq *q = &qs->rspq; | |
2263 | struct rsp_desc *r = &q->desc[q->cidx]; | |
2264 | int budget_left = budget; | |
6195c71d | 2265 | unsigned int sleeping = 0; |
4d22de3e DLR |
2266 | struct sk_buff *offload_skbs[RX_BUNDLE_SIZE]; |
2267 | int ngathered = 0; | |
2268 | ||
2269 | q->next_holdoff = q->holdoff_tmr; | |
2270 | ||
2271 | while (likely(budget_left && is_new_response(r, q))) { | |
b47385bd | 2272 | int packet_complete, eth, ethpad = 2, lro = qs->lro_enabled; |
4d22de3e DLR |
2273 | struct sk_buff *skb = NULL; |
2274 | u32 len, flags = ntohl(r->flags); | |
7385ecf3 DLR |
2275 | __be32 rss_hi = *(const __be32 *)r, |
2276 | rss_lo = r->rss_hdr.rss_hash_val; | |
4d22de3e DLR |
2277 | |
2278 | eth = r->rss_hdr.opcode == CPL_RX_PKT; | |
2279 | ||
2280 | if (unlikely(flags & F_RSPD_ASYNC_NOTIF)) { | |
2281 | skb = alloc_skb(AN_PKT_SIZE, GFP_ATOMIC); | |
2282 | if (!skb) | |
2283 | goto no_mem; | |
2284 | ||
2285 | memcpy(__skb_put(skb, AN_PKT_SIZE), r, AN_PKT_SIZE); | |
2286 | skb->data[0] = CPL_ASYNC_NOTIF; | |
2287 | rss_hi = htonl(CPL_ASYNC_NOTIF << 24); | |
2288 | q->async_notif++; | |
2289 | } else if (flags & F_RSPD_IMM_DATA_VALID) { | |
2290 | skb = get_imm_packet(r); | |
2291 | if (unlikely(!skb)) { | |
cf992af5 | 2292 | no_mem: |
4d22de3e DLR |
2293 | q->next_holdoff = NOMEM_INTR_DELAY; |
2294 | q->nomem++; | |
2295 | /* consume one credit since we tried */ | |
2296 | budget_left--; | |
2297 | break; | |
2298 | } | |
2299 | q->imm_data++; | |
e0994eb1 | 2300 | ethpad = 0; |
4d22de3e | 2301 | } else if ((len = ntohl(r->len_cq)) != 0) { |
cf992af5 | 2302 | struct sge_fl *fl; |
e0994eb1 | 2303 | |
65ab8385 | 2304 | lro &= eth && is_eth_tcp(rss_hi); |
b47385bd | 2305 | |
cf992af5 DLR |
2306 | fl = (len & F_RSPD_FLQ) ? &qs->fl[1] : &qs->fl[0]; |
2307 | if (fl->use_pages) { | |
2308 | void *addr = fl->sdesc[fl->cidx].pg_chunk.va; | |
e0994eb1 | 2309 | |
cf992af5 DLR |
2310 | prefetch(addr); |
2311 | #if L1_CACHE_BYTES < 128 | |
2312 | prefetch(addr + L1_CACHE_BYTES); | |
2313 | #endif | |
e0994eb1 | 2314 | __refill_fl(adap, fl); |
b47385bd DLR |
2315 | if (lro > 0) { |
2316 | lro_add_page(adap, qs, fl, | |
2317 | G_RSPD_LEN(len), | |
2318 | flags & F_RSPD_EOP); | |
2319 | goto next_fl; | |
2320 | } | |
e0994eb1 | 2321 | |
7385ecf3 DLR |
2322 | skb = get_packet_pg(adap, fl, q, |
2323 | G_RSPD_LEN(len), | |
2324 | eth ? | |
2325 | SGE_RX_DROP_THRES : 0); | |
2326 | q->pg_skb = skb; | |
cf992af5 | 2327 | } else |
e0994eb1 DLR |
2328 | skb = get_packet(adap, fl, G_RSPD_LEN(len), |
2329 | eth ? SGE_RX_DROP_THRES : 0); | |
cf992af5 DLR |
2330 | if (unlikely(!skb)) { |
2331 | if (!eth) | |
2332 | goto no_mem; | |
2333 | q->rx_drops++; | |
2334 | } else if (unlikely(r->rss_hdr.opcode == CPL_TRACE_PKT)) | |
2335 | __skb_pull(skb, 2); | |
b47385bd | 2336 | next_fl: |
4d22de3e DLR |
2337 | if (++fl->cidx == fl->size) |
2338 | fl->cidx = 0; | |
2339 | } else | |
2340 | q->pure_rsps++; | |
2341 | ||
2342 | if (flags & RSPD_CTRL_MASK) { | |
2343 | sleeping |= flags & RSPD_GTS_MASK; | |
6195c71d | 2344 | handle_rsp_cntrl_info(qs, flags); |
4d22de3e DLR |
2345 | } |
2346 | ||
2347 | r++; | |
2348 | if (unlikely(++q->cidx == q->size)) { | |
2349 | q->cidx = 0; | |
2350 | q->gen ^= 1; | |
2351 | r = q->desc; | |
2352 | } | |
2353 | prefetch(r); | |
2354 | ||
2355 | if (++q->credits >= (q->size / 4)) { | |
2356 | refill_rspq(adap, q, q->credits); | |
2357 | q->credits = 0; | |
2358 | } | |
2359 | ||
7385ecf3 DLR |
2360 | packet_complete = flags & |
2361 | (F_RSPD_EOP | F_RSPD_IMM_DATA_VALID | | |
2362 | F_RSPD_ASYNC_NOTIF); | |
2363 | ||
2364 | if (skb != NULL && packet_complete) { | |
4d22de3e | 2365 | if (eth) |
b47385bd | 2366 | rx_eth(adap, q, skb, ethpad, lro); |
4d22de3e | 2367 | else { |
afefce66 | 2368 | q->offload_pkts++; |
cf992af5 DLR |
2369 | /* Preserve the RSS info in csum & priority */ |
2370 | skb->csum = rss_hi; | |
2371 | skb->priority = rss_lo; | |
2372 | ngathered = rx_offload(&adap->tdev, q, skb, | |
2373 | offload_skbs, | |
e0994eb1 | 2374 | ngathered); |
4d22de3e | 2375 | } |
7385ecf3 DLR |
2376 | |
2377 | if (flags & F_RSPD_EOP) | |
b47385bd | 2378 | clear_rspq_bufstate(q); |
4d22de3e | 2379 | } |
4d22de3e DLR |
2380 | --budget_left; |
2381 | } | |
2382 | ||
4d22de3e | 2383 | deliver_partial_bundle(&adap->tdev, q, offload_skbs, ngathered); |
b47385bd | 2384 | |
4d22de3e DLR |
2385 | if (sleeping) |
2386 | check_ring_db(adap, qs, sleeping); | |
2387 | ||
2388 | smp_mb(); /* commit Tx queue .processed updates */ | |
2389 | if (unlikely(qs->txq_stopped != 0)) | |
2390 | restart_tx(qs); | |
2391 | ||
2392 | budget -= budget_left; | |
2393 | return budget; | |
2394 | } | |
2395 | ||
2396 | static inline int is_pure_response(const struct rsp_desc *r) | |
2397 | { | |
c5419e6f | 2398 | __be32 n = r->flags & htonl(F_RSPD_ASYNC_NOTIF | F_RSPD_IMM_DATA_VALID); |
4d22de3e DLR |
2399 | |
2400 | return (n | r->len_cq) == 0; | |
2401 | } | |
2402 | ||
2403 | /** | |
2404 | * napi_rx_handler - the NAPI handler for Rx processing | |
bea3348e | 2405 | * @napi: the napi instance |
4d22de3e DLR |
2406 | * @budget: how many packets we can process in this round |
2407 | * | |
2408 | * Handler for new data events when using NAPI. | |
2409 | */ | |
bea3348e | 2410 | static int napi_rx_handler(struct napi_struct *napi, int budget) |
4d22de3e | 2411 | { |
bea3348e SH |
2412 | struct sge_qset *qs = container_of(napi, struct sge_qset, napi); |
2413 | struct adapter *adap = qs->adap; | |
2414 | int work_done = process_responses(adap, qs, budget); | |
4d22de3e | 2415 | |
bea3348e SH |
2416 | if (likely(work_done < budget)) { |
2417 | napi_complete(napi); | |
4d22de3e | 2418 | |
bea3348e SH |
2419 | /* |
2420 | * Because we don't atomically flush the following | |
2421 | * write it is possible that in very rare cases it can | |
2422 | * reach the device in a way that races with a new | |
2423 | * response being written plus an error interrupt | |
2424 | * causing the NAPI interrupt handler below to return | |
2425 | * unhandled status to the OS. To protect against | |
2426 | * this would require flushing the write and doing | |
2427 | * both the write and the flush with interrupts off. | |
2428 | * Way too expensive and unjustifiable given the | |
2429 | * rarity of the race. | |
2430 | * | |
2431 | * The race cannot happen at all with MSI-X. | |
2432 | */ | |
2433 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(qs->rspq.cntxt_id) | | |
2434 | V_NEWTIMER(qs->rspq.next_holdoff) | | |
2435 | V_NEWINDEX(qs->rspq.cidx)); | |
2436 | } | |
2437 | return work_done; | |
4d22de3e DLR |
2438 | } |
2439 | ||
2440 | /* | |
2441 | * Returns true if the device is already scheduled for polling. | |
2442 | */ | |
bea3348e | 2443 | static inline int napi_is_scheduled(struct napi_struct *napi) |
4d22de3e | 2444 | { |
bea3348e | 2445 | return test_bit(NAPI_STATE_SCHED, &napi->state); |
4d22de3e DLR |
2446 | } |
2447 | ||
2448 | /** | |
2449 | * process_pure_responses - process pure responses from a response queue | |
2450 | * @adap: the adapter | |
2451 | * @qs: the queue set owning the response queue | |
2452 | * @r: the first pure response to process | |
2453 | * | |
2454 | * A simpler version of process_responses() that handles only pure (i.e., | |
2455 | * non data-carrying) responses. Such respones are too light-weight to | |
2456 | * justify calling a softirq under NAPI, so we handle them specially in | |
2457 | * the interrupt handler. The function is called with a pointer to a | |
2458 | * response, which the caller must ensure is a valid pure response. | |
2459 | * | |
2460 | * Returns 1 if it encounters a valid data-carrying response, 0 otherwise. | |
2461 | */ | |
2462 | static int process_pure_responses(struct adapter *adap, struct sge_qset *qs, | |
2463 | struct rsp_desc *r) | |
2464 | { | |
2465 | struct sge_rspq *q = &qs->rspq; | |
6195c71d | 2466 | unsigned int sleeping = 0; |
4d22de3e DLR |
2467 | |
2468 | do { | |
2469 | u32 flags = ntohl(r->flags); | |
2470 | ||
2471 | r++; | |
2472 | if (unlikely(++q->cidx == q->size)) { | |
2473 | q->cidx = 0; | |
2474 | q->gen ^= 1; | |
2475 | r = q->desc; | |
2476 | } | |
2477 | prefetch(r); | |
2478 | ||
2479 | if (flags & RSPD_CTRL_MASK) { | |
2480 | sleeping |= flags & RSPD_GTS_MASK; | |
6195c71d | 2481 | handle_rsp_cntrl_info(qs, flags); |
4d22de3e DLR |
2482 | } |
2483 | ||
2484 | q->pure_rsps++; | |
2485 | if (++q->credits >= (q->size / 4)) { | |
2486 | refill_rspq(adap, q, q->credits); | |
2487 | q->credits = 0; | |
2488 | } | |
2489 | } while (is_new_response(r, q) && is_pure_response(r)); | |
2490 | ||
4d22de3e DLR |
2491 | if (sleeping) |
2492 | check_ring_db(adap, qs, sleeping); | |
2493 | ||
2494 | smp_mb(); /* commit Tx queue .processed updates */ | |
2495 | if (unlikely(qs->txq_stopped != 0)) | |
2496 | restart_tx(qs); | |
2497 | ||
2498 | return is_new_response(r, q); | |
2499 | } | |
2500 | ||
2501 | /** | |
2502 | * handle_responses - decide what to do with new responses in NAPI mode | |
2503 | * @adap: the adapter | |
2504 | * @q: the response queue | |
2505 | * | |
2506 | * This is used by the NAPI interrupt handlers to decide what to do with | |
2507 | * new SGE responses. If there are no new responses it returns -1. If | |
2508 | * there are new responses and they are pure (i.e., non-data carrying) | |
2509 | * it handles them straight in hard interrupt context as they are very | |
2510 | * cheap and don't deliver any packets. Finally, if there are any data | |
2511 | * signaling responses it schedules the NAPI handler. Returns 1 if it | |
2512 | * schedules NAPI, 0 if all new responses were pure. | |
2513 | * | |
2514 | * The caller must ascertain NAPI is not already running. | |
2515 | */ | |
2516 | static inline int handle_responses(struct adapter *adap, struct sge_rspq *q) | |
2517 | { | |
2518 | struct sge_qset *qs = rspq_to_qset(q); | |
2519 | struct rsp_desc *r = &q->desc[q->cidx]; | |
2520 | ||
2521 | if (!is_new_response(r, q)) | |
2522 | return -1; | |
2523 | if (is_pure_response(r) && process_pure_responses(adap, qs, r) == 0) { | |
2524 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(q->cntxt_id) | | |
2525 | V_NEWTIMER(q->holdoff_tmr) | V_NEWINDEX(q->cidx)); | |
2526 | return 0; | |
2527 | } | |
bea3348e | 2528 | napi_schedule(&qs->napi); |
4d22de3e DLR |
2529 | return 1; |
2530 | } | |
2531 | ||
2532 | /* | |
2533 | * The MSI-X interrupt handler for an SGE response queue for the non-NAPI case | |
2534 | * (i.e., response queue serviced in hard interrupt). | |
2535 | */ | |
2536 | irqreturn_t t3_sge_intr_msix(int irq, void *cookie) | |
2537 | { | |
2538 | struct sge_qset *qs = cookie; | |
bea3348e | 2539 | struct adapter *adap = qs->adap; |
4d22de3e DLR |
2540 | struct sge_rspq *q = &qs->rspq; |
2541 | ||
2542 | spin_lock(&q->lock); | |
2543 | if (process_responses(adap, qs, -1) == 0) | |
2544 | q->unhandled_irqs++; | |
2545 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(q->cntxt_id) | | |
2546 | V_NEWTIMER(q->next_holdoff) | V_NEWINDEX(q->cidx)); | |
2547 | spin_unlock(&q->lock); | |
2548 | return IRQ_HANDLED; | |
2549 | } | |
2550 | ||
2551 | /* | |
2552 | * The MSI-X interrupt handler for an SGE response queue for the NAPI case | |
2553 | * (i.e., response queue serviced by NAPI polling). | |
2554 | */ | |
9265fabf | 2555 | static irqreturn_t t3_sge_intr_msix_napi(int irq, void *cookie) |
4d22de3e DLR |
2556 | { |
2557 | struct sge_qset *qs = cookie; | |
4d22de3e DLR |
2558 | struct sge_rspq *q = &qs->rspq; |
2559 | ||
2560 | spin_lock(&q->lock); | |
4d22de3e | 2561 | |
bea3348e | 2562 | if (handle_responses(qs->adap, q) < 0) |
4d22de3e DLR |
2563 | q->unhandled_irqs++; |
2564 | spin_unlock(&q->lock); | |
2565 | return IRQ_HANDLED; | |
2566 | } | |
2567 | ||
2568 | /* | |
2569 | * The non-NAPI MSI interrupt handler. This needs to handle data events from | |
2570 | * SGE response queues as well as error and other async events as they all use | |
2571 | * the same MSI vector. We use one SGE response queue per port in this mode | |
2572 | * and protect all response queues with queue 0's lock. | |
2573 | */ | |
2574 | static irqreturn_t t3_intr_msi(int irq, void *cookie) | |
2575 | { | |
2576 | int new_packets = 0; | |
2577 | struct adapter *adap = cookie; | |
2578 | struct sge_rspq *q = &adap->sge.qs[0].rspq; | |
2579 | ||
2580 | spin_lock(&q->lock); | |
2581 | ||
2582 | if (process_responses(adap, &adap->sge.qs[0], -1)) { | |
2583 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(q->cntxt_id) | | |
2584 | V_NEWTIMER(q->next_holdoff) | V_NEWINDEX(q->cidx)); | |
2585 | new_packets = 1; | |
2586 | } | |
2587 | ||
2588 | if (adap->params.nports == 2 && | |
2589 | process_responses(adap, &adap->sge.qs[1], -1)) { | |
2590 | struct sge_rspq *q1 = &adap->sge.qs[1].rspq; | |
2591 | ||
2592 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(q1->cntxt_id) | | |
2593 | V_NEWTIMER(q1->next_holdoff) | | |
2594 | V_NEWINDEX(q1->cidx)); | |
2595 | new_packets = 1; | |
2596 | } | |
2597 | ||
2598 | if (!new_packets && t3_slow_intr_handler(adap) == 0) | |
2599 | q->unhandled_irqs++; | |
2600 | ||
2601 | spin_unlock(&q->lock); | |
2602 | return IRQ_HANDLED; | |
2603 | } | |
2604 | ||
bea3348e | 2605 | static int rspq_check_napi(struct sge_qset *qs) |
4d22de3e | 2606 | { |
bea3348e SH |
2607 | struct sge_rspq *q = &qs->rspq; |
2608 | ||
2609 | if (!napi_is_scheduled(&qs->napi) && | |
2610 | is_new_response(&q->desc[q->cidx], q)) { | |
2611 | napi_schedule(&qs->napi); | |
4d22de3e DLR |
2612 | return 1; |
2613 | } | |
2614 | return 0; | |
2615 | } | |
2616 | ||
2617 | /* | |
2618 | * The MSI interrupt handler for the NAPI case (i.e., response queues serviced | |
2619 | * by NAPI polling). Handles data events from SGE response queues as well as | |
2620 | * error and other async events as they all use the same MSI vector. We use | |
2621 | * one SGE response queue per port in this mode and protect all response | |
2622 | * queues with queue 0's lock. | |
2623 | */ | |
9265fabf | 2624 | static irqreturn_t t3_intr_msi_napi(int irq, void *cookie) |
4d22de3e DLR |
2625 | { |
2626 | int new_packets; | |
2627 | struct adapter *adap = cookie; | |
2628 | struct sge_rspq *q = &adap->sge.qs[0].rspq; | |
2629 | ||
2630 | spin_lock(&q->lock); | |
2631 | ||
bea3348e | 2632 | new_packets = rspq_check_napi(&adap->sge.qs[0]); |
4d22de3e | 2633 | if (adap->params.nports == 2) |
bea3348e | 2634 | new_packets += rspq_check_napi(&adap->sge.qs[1]); |
4d22de3e DLR |
2635 | if (!new_packets && t3_slow_intr_handler(adap) == 0) |
2636 | q->unhandled_irqs++; | |
2637 | ||
2638 | spin_unlock(&q->lock); | |
2639 | return IRQ_HANDLED; | |
2640 | } | |
2641 | ||
2642 | /* | |
2643 | * A helper function that processes responses and issues GTS. | |
2644 | */ | |
2645 | static inline int process_responses_gts(struct adapter *adap, | |
2646 | struct sge_rspq *rq) | |
2647 | { | |
2648 | int work; | |
2649 | ||
2650 | work = process_responses(adap, rspq_to_qset(rq), -1); | |
2651 | t3_write_reg(adap, A_SG_GTS, V_RSPQ(rq->cntxt_id) | | |
2652 | V_NEWTIMER(rq->next_holdoff) | V_NEWINDEX(rq->cidx)); | |
2653 | return work; | |
2654 | } | |
2655 | ||
2656 | /* | |
2657 | * The legacy INTx interrupt handler. This needs to handle data events from | |
2658 | * SGE response queues as well as error and other async events as they all use | |
2659 | * the same interrupt pin. We use one SGE response queue per port in this mode | |
2660 | * and protect all response queues with queue 0's lock. | |
2661 | */ | |
2662 | static irqreturn_t t3_intr(int irq, void *cookie) | |
2663 | { | |
2664 | int work_done, w0, w1; | |
2665 | struct adapter *adap = cookie; | |
2666 | struct sge_rspq *q0 = &adap->sge.qs[0].rspq; | |
2667 | struct sge_rspq *q1 = &adap->sge.qs[1].rspq; | |
2668 | ||
2669 | spin_lock(&q0->lock); | |
2670 | ||
2671 | w0 = is_new_response(&q0->desc[q0->cidx], q0); | |
2672 | w1 = adap->params.nports == 2 && | |
2673 | is_new_response(&q1->desc[q1->cidx], q1); | |
2674 | ||
2675 | if (likely(w0 | w1)) { | |
2676 | t3_write_reg(adap, A_PL_CLI, 0); | |
2677 | t3_read_reg(adap, A_PL_CLI); /* flush */ | |
2678 | ||
2679 | if (likely(w0)) | |
2680 | process_responses_gts(adap, q0); | |
2681 | ||
2682 | if (w1) | |
2683 | process_responses_gts(adap, q1); | |
2684 | ||
2685 | work_done = w0 | w1; | |
2686 | } else | |
2687 | work_done = t3_slow_intr_handler(adap); | |
2688 | ||
2689 | spin_unlock(&q0->lock); | |
2690 | return IRQ_RETVAL(work_done != 0); | |
2691 | } | |
2692 | ||
2693 | /* | |
2694 | * Interrupt handler for legacy INTx interrupts for T3B-based cards. | |
2695 | * Handles data events from SGE response queues as well as error and other | |
2696 | * async events as they all use the same interrupt pin. We use one SGE | |
2697 | * response queue per port in this mode and protect all response queues with | |
2698 | * queue 0's lock. | |
2699 | */ | |
2700 | static irqreturn_t t3b_intr(int irq, void *cookie) | |
2701 | { | |
2702 | u32 map; | |
2703 | struct adapter *adap = cookie; | |
2704 | struct sge_rspq *q0 = &adap->sge.qs[0].rspq; | |
2705 | ||
2706 | t3_write_reg(adap, A_PL_CLI, 0); | |
2707 | map = t3_read_reg(adap, A_SG_DATA_INTR); | |
2708 | ||
2709 | if (unlikely(!map)) /* shared interrupt, most likely */ | |
2710 | return IRQ_NONE; | |
2711 | ||
2712 | spin_lock(&q0->lock); | |
2713 | ||
2714 | if (unlikely(map & F_ERRINTR)) | |
2715 | t3_slow_intr_handler(adap); | |
2716 | ||
2717 | if (likely(map & 1)) | |
2718 | process_responses_gts(adap, q0); | |
2719 | ||
2720 | if (map & 2) | |
2721 | process_responses_gts(adap, &adap->sge.qs[1].rspq); | |
2722 | ||
2723 | spin_unlock(&q0->lock); | |
2724 | return IRQ_HANDLED; | |
2725 | } | |
2726 | ||
2727 | /* | |
2728 | * NAPI interrupt handler for legacy INTx interrupts for T3B-based cards. | |
2729 | * Handles data events from SGE response queues as well as error and other | |
2730 | * async events as they all use the same interrupt pin. We use one SGE | |
2731 | * response queue per port in this mode and protect all response queues with | |
2732 | * queue 0's lock. | |
2733 | */ | |
2734 | static irqreturn_t t3b_intr_napi(int irq, void *cookie) | |
2735 | { | |
2736 | u32 map; | |
4d22de3e | 2737 | struct adapter *adap = cookie; |
bea3348e SH |
2738 | struct sge_qset *qs0 = &adap->sge.qs[0]; |
2739 | struct sge_rspq *q0 = &qs0->rspq; | |
4d22de3e DLR |
2740 | |
2741 | t3_write_reg(adap, A_PL_CLI, 0); | |
2742 | map = t3_read_reg(adap, A_SG_DATA_INTR); | |
2743 | ||
2744 | if (unlikely(!map)) /* shared interrupt, most likely */ | |
2745 | return IRQ_NONE; | |
2746 | ||
2747 | spin_lock(&q0->lock); | |
2748 | ||
2749 | if (unlikely(map & F_ERRINTR)) | |
2750 | t3_slow_intr_handler(adap); | |
2751 | ||
bea3348e SH |
2752 | if (likely(map & 1)) |
2753 | napi_schedule(&qs0->napi); | |
4d22de3e | 2754 | |
bea3348e SH |
2755 | if (map & 2) |
2756 | napi_schedule(&adap->sge.qs[1].napi); | |
4d22de3e DLR |
2757 | |
2758 | spin_unlock(&q0->lock); | |
2759 | return IRQ_HANDLED; | |
2760 | } | |
2761 | ||
2762 | /** | |
2763 | * t3_intr_handler - select the top-level interrupt handler | |
2764 | * @adap: the adapter | |
2765 | * @polling: whether using NAPI to service response queues | |
2766 | * | |
2767 | * Selects the top-level interrupt handler based on the type of interrupts | |
2768 | * (MSI-X, MSI, or legacy) and whether NAPI will be used to service the | |
2769 | * response queues. | |
2770 | */ | |
7c239975 | 2771 | irq_handler_t t3_intr_handler(struct adapter *adap, int polling) |
4d22de3e DLR |
2772 | { |
2773 | if (adap->flags & USING_MSIX) | |
2774 | return polling ? t3_sge_intr_msix_napi : t3_sge_intr_msix; | |
2775 | if (adap->flags & USING_MSI) | |
2776 | return polling ? t3_intr_msi_napi : t3_intr_msi; | |
2777 | if (adap->params.rev > 0) | |
2778 | return polling ? t3b_intr_napi : t3b_intr; | |
2779 | return t3_intr; | |
2780 | } | |
2781 | ||
b881955b DLR |
2782 | #define SGE_PARERR (F_CPPARITYERROR | F_OCPARITYERROR | F_RCPARITYERROR | \ |
2783 | F_IRPARITYERROR | V_ITPARITYERROR(M_ITPARITYERROR) | \ | |
2784 | V_FLPARITYERROR(M_FLPARITYERROR) | F_LODRBPARITYERROR | \ | |
2785 | F_HIDRBPARITYERROR | F_LORCQPARITYERROR | \ | |
2786 | F_HIRCQPARITYERROR) | |
2787 | #define SGE_FRAMINGERR (F_UC_REQ_FRAMINGERROR | F_R_REQ_FRAMINGERROR) | |
2788 | #define SGE_FATALERR (SGE_PARERR | SGE_FRAMINGERR | F_RSPQCREDITOVERFOW | \ | |
2789 | F_RSPQDISABLED) | |
2790 | ||
4d22de3e DLR |
2791 | /** |
2792 | * t3_sge_err_intr_handler - SGE async event interrupt handler | |
2793 | * @adapter: the adapter | |
2794 | * | |
2795 | * Interrupt handler for SGE asynchronous (non-data) events. | |
2796 | */ | |
2797 | void t3_sge_err_intr_handler(struct adapter *adapter) | |
2798 | { | |
fc882196 DLR |
2799 | unsigned int v, status = t3_read_reg(adapter, A_SG_INT_CAUSE) & |
2800 | ~F_FLEMPTY; | |
4d22de3e | 2801 | |
b881955b DLR |
2802 | if (status & SGE_PARERR) |
2803 | CH_ALERT(adapter, "SGE parity error (0x%x)\n", | |
2804 | status & SGE_PARERR); | |
2805 | if (status & SGE_FRAMINGERR) | |
2806 | CH_ALERT(adapter, "SGE framing error (0x%x)\n", | |
2807 | status & SGE_FRAMINGERR); | |
2808 | ||
4d22de3e DLR |
2809 | if (status & F_RSPQCREDITOVERFOW) |
2810 | CH_ALERT(adapter, "SGE response queue credit overflow\n"); | |
2811 | ||
2812 | if (status & F_RSPQDISABLED) { | |
2813 | v = t3_read_reg(adapter, A_SG_RSPQ_FL_STATUS); | |
2814 | ||
2815 | CH_ALERT(adapter, | |
2816 | "packet delivered to disabled response queue " | |
2817 | "(0x%x)\n", (v >> S_RSPQ0DISABLED) & 0xff); | |
2818 | } | |
2819 | ||
6e3f03b7 DLR |
2820 | if (status & (F_HIPIODRBDROPERR | F_LOPIODRBDROPERR)) |
2821 | CH_ALERT(adapter, "SGE dropped %s priority doorbell\n", | |
2822 | status & F_HIPIODRBDROPERR ? "high" : "lo"); | |
2823 | ||
4d22de3e | 2824 | t3_write_reg(adapter, A_SG_INT_CAUSE, status); |
b881955b | 2825 | if (status & SGE_FATALERR) |
4d22de3e DLR |
2826 | t3_fatal_err(adapter); |
2827 | } | |
2828 | ||
2829 | /** | |
42c8ea17 | 2830 | * sge_timer_tx - perform periodic maintenance of an SGE qset |
4d22de3e DLR |
2831 | * @data: the SGE queue set to maintain |
2832 | * | |
2833 | * Runs periodically from a timer to perform maintenance of an SGE queue | |
2834 | * set. It performs two tasks: | |
2835 | * | |
42c8ea17 | 2836 | * Cleans up any completed Tx descriptors that may still be pending. |
4d22de3e DLR |
2837 | * Normal descriptor cleanup happens when new packets are added to a Tx |
2838 | * queue so this timer is relatively infrequent and does any cleanup only | |
2839 | * if the Tx queue has not seen any new packets in a while. We make a | |
2840 | * best effort attempt to reclaim descriptors, in that we don't wait | |
2841 | * around if we cannot get a queue's lock (which most likely is because | |
2842 | * someone else is queueing new packets and so will also handle the clean | |
2843 | * up). Since control queues use immediate data exclusively we don't | |
2844 | * bother cleaning them up here. | |
2845 | * | |
4d22de3e | 2846 | */ |
42c8ea17 | 2847 | static void sge_timer_tx(unsigned long data) |
4d22de3e | 2848 | { |
4d22de3e | 2849 | struct sge_qset *qs = (struct sge_qset *)data; |
42c8ea17 DLR |
2850 | struct port_info *pi = netdev_priv(qs->netdev); |
2851 | struct adapter *adap = pi->adapter; | |
2852 | unsigned int tbd[SGE_TXQ_PER_SET] = {0, 0}; | |
2853 | unsigned long next_period; | |
4d22de3e | 2854 | |
c3a8c5b6 DLR |
2855 | if (__netif_tx_trylock(qs->tx_q)) { |
2856 | tbd[TXQ_ETH] = reclaim_completed_tx(adap, &qs->txq[TXQ_ETH], | |
2857 | TX_RECLAIM_TIMER_CHUNK); | |
2858 | __netif_tx_unlock(qs->tx_q); | |
4d22de3e | 2859 | } |
c3a8c5b6 | 2860 | |
4d22de3e | 2861 | if (spin_trylock(&qs->txq[TXQ_OFLD].lock)) { |
42c8ea17 DLR |
2862 | tbd[TXQ_OFLD] = reclaim_completed_tx(adap, &qs->txq[TXQ_OFLD], |
2863 | TX_RECLAIM_TIMER_CHUNK); | |
4d22de3e DLR |
2864 | spin_unlock(&qs->txq[TXQ_OFLD].lock); |
2865 | } | |
42c8ea17 DLR |
2866 | |
2867 | next_period = TX_RECLAIM_PERIOD >> | |
c3a8c5b6 DLR |
2868 | (max(tbd[TXQ_ETH], tbd[TXQ_OFLD]) / |
2869 | TX_RECLAIM_TIMER_CHUNK); | |
42c8ea17 DLR |
2870 | mod_timer(&qs->tx_reclaim_timer, jiffies + next_period); |
2871 | } | |
2872 | ||
2873 | /* | |
2874 | * sge_timer_rx - perform periodic maintenance of an SGE qset | |
2875 | * @data: the SGE queue set to maintain | |
2876 | * | |
2877 | * a) Replenishes Rx queues that have run out due to memory shortage. | |
2878 | * Normally new Rx buffers are added when existing ones are consumed but | |
2879 | * when out of memory a queue can become empty. We try to add only a few | |
2880 | * buffers here, the queue will be replenished fully as these new buffers | |
2881 | * are used up if memory shortage has subsided. | |
2882 | * | |
2883 | * b) Return coalesced response queue credits in case a response queue is | |
2884 | * starved. | |
2885 | * | |
2886 | */ | |
2887 | static void sge_timer_rx(unsigned long data) | |
2888 | { | |
2889 | spinlock_t *lock; | |
2890 | struct sge_qset *qs = (struct sge_qset *)data; | |
2891 | struct port_info *pi = netdev_priv(qs->netdev); | |
2892 | struct adapter *adap = pi->adapter; | |
2893 | u32 status; | |
2894 | ||
2895 | lock = adap->params.rev > 0 ? | |
2896 | &qs->rspq.lock : &adap->sge.qs[0].rspq.lock; | |
2897 | ||
2898 | if (!spin_trylock_irq(lock)) | |
2899 | goto out; | |
2900 | ||
2901 | if (napi_is_scheduled(&qs->napi)) | |
2902 | goto unlock; | |
2903 | ||
2904 | if (adap->params.rev < 4) { | |
2905 | status = t3_read_reg(adap, A_SG_RSPQ_FL_STATUS); | |
2906 | ||
2907 | if (status & (1 << qs->rspq.cntxt_id)) { | |
2908 | qs->rspq.starved++; | |
2909 | if (qs->rspq.credits) { | |
2910 | qs->rspq.credits--; | |
2911 | refill_rspq(adap, &qs->rspq, 1); | |
2912 | qs->rspq.restarted++; | |
2913 | t3_write_reg(adap, A_SG_RSPQ_FL_STATUS, | |
2914 | 1 << qs->rspq.cntxt_id); | |
bae73f44 | 2915 | } |
4d22de3e | 2916 | } |
4d22de3e | 2917 | } |
42c8ea17 DLR |
2918 | |
2919 | if (qs->fl[0].credits < qs->fl[0].size) | |
2920 | __refill_fl(adap, &qs->fl[0]); | |
2921 | if (qs->fl[1].credits < qs->fl[1].size) | |
2922 | __refill_fl(adap, &qs->fl[1]); | |
2923 | ||
2924 | unlock: | |
2925 | spin_unlock_irq(lock); | |
2926 | out: | |
2927 | mod_timer(&qs->rx_reclaim_timer, jiffies + RX_RECLAIM_PERIOD); | |
4d22de3e DLR |
2928 | } |
2929 | ||
2930 | /** | |
2931 | * t3_update_qset_coalesce - update coalescing settings for a queue set | |
2932 | * @qs: the SGE queue set | |
2933 | * @p: new queue set parameters | |
2934 | * | |
2935 | * Update the coalescing settings for an SGE queue set. Nothing is done | |
2936 | * if the queue set is not initialized yet. | |
2937 | */ | |
2938 | void t3_update_qset_coalesce(struct sge_qset *qs, const struct qset_params *p) | |
2939 | { | |
4d22de3e DLR |
2940 | qs->rspq.holdoff_tmr = max(p->coalesce_usecs * 10, 1U);/* can't be 0 */ |
2941 | qs->rspq.polling = p->polling; | |
bea3348e | 2942 | qs->napi.poll = p->polling ? napi_rx_handler : ofld_poll; |
4d22de3e DLR |
2943 | } |
2944 | ||
2945 | /** | |
2946 | * t3_sge_alloc_qset - initialize an SGE queue set | |
2947 | * @adapter: the adapter | |
2948 | * @id: the queue set id | |
2949 | * @nports: how many Ethernet ports will be using this queue set | |
2950 | * @irq_vec_idx: the IRQ vector index for response queue interrupts | |
2951 | * @p: configuration parameters for this queue set | |
2952 | * @ntxq: number of Tx queues for the queue set | |
2953 | * @netdev: net device associated with this queue set | |
82ad3329 | 2954 | * @netdevq: net device TX queue associated with this queue set |
4d22de3e DLR |
2955 | * |
2956 | * Allocate resources and initialize an SGE queue set. A queue set | |
2957 | * comprises a response queue, two Rx free-buffer queues, and up to 3 | |
2958 | * Tx queues. The Tx queues are assigned roles in the order Ethernet | |
2959 | * queue, offload queue, and control queue. | |
2960 | */ | |
2961 | int t3_sge_alloc_qset(struct adapter *adapter, unsigned int id, int nports, | |
2962 | int irq_vec_idx, const struct qset_params *p, | |
82ad3329 DLR |
2963 | int ntxq, struct net_device *dev, |
2964 | struct netdev_queue *netdevq) | |
4d22de3e | 2965 | { |
b1fb1f28 | 2966 | int i, avail, ret = -ENOMEM; |
4d22de3e DLR |
2967 | struct sge_qset *q = &adapter->sge.qs[id]; |
2968 | ||
2969 | init_qset_cntxt(q, id); | |
42c8ea17 DLR |
2970 | setup_timer(&q->tx_reclaim_timer, sge_timer_tx, (unsigned long)q); |
2971 | setup_timer(&q->rx_reclaim_timer, sge_timer_rx, (unsigned long)q); | |
4d22de3e DLR |
2972 | |
2973 | q->fl[0].desc = alloc_ring(adapter->pdev, p->fl_size, | |
2974 | sizeof(struct rx_desc), | |
2975 | sizeof(struct rx_sw_desc), | |
2976 | &q->fl[0].phys_addr, &q->fl[0].sdesc); | |
2977 | if (!q->fl[0].desc) | |
2978 | goto err; | |
2979 | ||
2980 | q->fl[1].desc = alloc_ring(adapter->pdev, p->jumbo_size, | |
2981 | sizeof(struct rx_desc), | |
2982 | sizeof(struct rx_sw_desc), | |
2983 | &q->fl[1].phys_addr, &q->fl[1].sdesc); | |
2984 | if (!q->fl[1].desc) | |
2985 | goto err; | |
2986 | ||
2987 | q->rspq.desc = alloc_ring(adapter->pdev, p->rspq_size, | |
2988 | sizeof(struct rsp_desc), 0, | |
2989 | &q->rspq.phys_addr, NULL); | |
2990 | if (!q->rspq.desc) | |
2991 | goto err; | |
2992 | ||
2993 | for (i = 0; i < ntxq; ++i) { | |
2994 | /* | |
2995 | * The control queue always uses immediate data so does not | |
2996 | * need to keep track of any sk_buffs. | |
2997 | */ | |
2998 | size_t sz = i == TXQ_CTRL ? 0 : sizeof(struct tx_sw_desc); | |
2999 | ||
3000 | q->txq[i].desc = alloc_ring(adapter->pdev, p->txq_size[i], | |
3001 | sizeof(struct tx_desc), sz, | |
3002 | &q->txq[i].phys_addr, | |
3003 | &q->txq[i].sdesc); | |
3004 | if (!q->txq[i].desc) | |
3005 | goto err; | |
3006 | ||
3007 | q->txq[i].gen = 1; | |
3008 | q->txq[i].size = p->txq_size[i]; | |
3009 | spin_lock_init(&q->txq[i].lock); | |
3010 | skb_queue_head_init(&q->txq[i].sendq); | |
3011 | } | |
3012 | ||
3013 | tasklet_init(&q->txq[TXQ_OFLD].qresume_tsk, restart_offloadq, | |
3014 | (unsigned long)q); | |
3015 | tasklet_init(&q->txq[TXQ_CTRL].qresume_tsk, restart_ctrlq, | |
3016 | (unsigned long)q); | |
3017 | ||
3018 | q->fl[0].gen = q->fl[1].gen = 1; | |
3019 | q->fl[0].size = p->fl_size; | |
3020 | q->fl[1].size = p->jumbo_size; | |
3021 | ||
3022 | q->rspq.gen = 1; | |
3023 | q->rspq.size = p->rspq_size; | |
3024 | spin_lock_init(&q->rspq.lock); | |
147e70e6 | 3025 | skb_queue_head_init(&q->rspq.rx_queue); |
4d22de3e DLR |
3026 | |
3027 | q->txq[TXQ_ETH].stop_thres = nports * | |
3028 | flits_to_desc(sgl_len(MAX_SKB_FRAGS + 1) + 3); | |
3029 | ||
cf992af5 DLR |
3030 | #if FL0_PG_CHUNK_SIZE > 0 |
3031 | q->fl[0].buf_size = FL0_PG_CHUNK_SIZE; | |
e0994eb1 | 3032 | #else |
cf992af5 | 3033 | q->fl[0].buf_size = SGE_RX_SM_BUF_SIZE + sizeof(struct cpl_rx_data); |
e0994eb1 | 3034 | #endif |
7385ecf3 DLR |
3035 | #if FL1_PG_CHUNK_SIZE > 0 |
3036 | q->fl[1].buf_size = FL1_PG_CHUNK_SIZE; | |
3037 | #else | |
cf992af5 DLR |
3038 | q->fl[1].buf_size = is_offload(adapter) ? |
3039 | (16 * 1024) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : | |
3040 | MAX_FRAME_SIZE + 2 + sizeof(struct cpl_rx_pkt); | |
7385ecf3 DLR |
3041 | #endif |
3042 | ||
3043 | q->fl[0].use_pages = FL0_PG_CHUNK_SIZE > 0; | |
3044 | q->fl[1].use_pages = FL1_PG_CHUNK_SIZE > 0; | |
3045 | q->fl[0].order = FL0_PG_ORDER; | |
3046 | q->fl[1].order = FL1_PG_ORDER; | |
5e68b772 DLR |
3047 | q->fl[0].alloc_size = FL0_PG_ALLOC_SIZE; |
3048 | q->fl[1].alloc_size = FL1_PG_ALLOC_SIZE; | |
4d22de3e | 3049 | |
b1186dee | 3050 | spin_lock_irq(&adapter->sge.reg_lock); |
4d22de3e DLR |
3051 | |
3052 | /* FL threshold comparison uses < */ | |
3053 | ret = t3_sge_init_rspcntxt(adapter, q->rspq.cntxt_id, irq_vec_idx, | |
3054 | q->rspq.phys_addr, q->rspq.size, | |
5e68b772 | 3055 | q->fl[0].buf_size - SGE_PG_RSVD, 1, 0); |
4d22de3e DLR |
3056 | if (ret) |
3057 | goto err_unlock; | |
3058 | ||
3059 | for (i = 0; i < SGE_RXQ_PER_SET; ++i) { | |
3060 | ret = t3_sge_init_flcntxt(adapter, q->fl[i].cntxt_id, 0, | |
3061 | q->fl[i].phys_addr, q->fl[i].size, | |
5e68b772 DLR |
3062 | q->fl[i].buf_size - SGE_PG_RSVD, |
3063 | p->cong_thres, 1, 0); | |
4d22de3e DLR |
3064 | if (ret) |
3065 | goto err_unlock; | |
3066 | } | |
3067 | ||
3068 | ret = t3_sge_init_ecntxt(adapter, q->txq[TXQ_ETH].cntxt_id, USE_GTS, | |
3069 | SGE_CNTXT_ETH, id, q->txq[TXQ_ETH].phys_addr, | |
3070 | q->txq[TXQ_ETH].size, q->txq[TXQ_ETH].token, | |
3071 | 1, 0); | |
3072 | if (ret) | |
3073 | goto err_unlock; | |
3074 | ||
3075 | if (ntxq > 1) { | |
3076 | ret = t3_sge_init_ecntxt(adapter, q->txq[TXQ_OFLD].cntxt_id, | |
3077 | USE_GTS, SGE_CNTXT_OFLD, id, | |
3078 | q->txq[TXQ_OFLD].phys_addr, | |
3079 | q->txq[TXQ_OFLD].size, 0, 1, 0); | |
3080 | if (ret) | |
3081 | goto err_unlock; | |
3082 | } | |
3083 | ||
3084 | if (ntxq > 2) { | |
3085 | ret = t3_sge_init_ecntxt(adapter, q->txq[TXQ_CTRL].cntxt_id, 0, | |
3086 | SGE_CNTXT_CTRL, id, | |
3087 | q->txq[TXQ_CTRL].phys_addr, | |
3088 | q->txq[TXQ_CTRL].size, | |
3089 | q->txq[TXQ_CTRL].token, 1, 0); | |
3090 | if (ret) | |
3091 | goto err_unlock; | |
3092 | } | |
3093 | ||
b1186dee | 3094 | spin_unlock_irq(&adapter->sge.reg_lock); |
4d22de3e | 3095 | |
bea3348e SH |
3096 | q->adap = adapter; |
3097 | q->netdev = dev; | |
82ad3329 | 3098 | q->tx_q = netdevq; |
bea3348e | 3099 | t3_update_qset_coalesce(q, p); |
b47385bd | 3100 | |
7385ecf3 DLR |
3101 | avail = refill_fl(adapter, &q->fl[0], q->fl[0].size, |
3102 | GFP_KERNEL | __GFP_COMP); | |
b1fb1f28 DLR |
3103 | if (!avail) { |
3104 | CH_ALERT(adapter, "free list queue 0 initialization failed\n"); | |
3105 | goto err; | |
3106 | } | |
3107 | if (avail < q->fl[0].size) | |
3108 | CH_WARN(adapter, "free list queue 0 enabled with %d credits\n", | |
3109 | avail); | |
3110 | ||
7385ecf3 DLR |
3111 | avail = refill_fl(adapter, &q->fl[1], q->fl[1].size, |
3112 | GFP_KERNEL | __GFP_COMP); | |
b1fb1f28 DLR |
3113 | if (avail < q->fl[1].size) |
3114 | CH_WARN(adapter, "free list queue 1 enabled with %d credits\n", | |
3115 | avail); | |
4d22de3e DLR |
3116 | refill_rspq(adapter, &q->rspq, q->rspq.size - 1); |
3117 | ||
3118 | t3_write_reg(adapter, A_SG_GTS, V_RSPQ(q->rspq.cntxt_id) | | |
3119 | V_NEWTIMER(q->rspq.holdoff_tmr)); | |
3120 | ||
4d22de3e DLR |
3121 | return 0; |
3122 | ||
b1fb1f28 | 3123 | err_unlock: |
b1186dee | 3124 | spin_unlock_irq(&adapter->sge.reg_lock); |
b1fb1f28 | 3125 | err: |
4d22de3e DLR |
3126 | t3_free_qset(adapter, q); |
3127 | return ret; | |
3128 | } | |
3129 | ||
31563789 DLR |
3130 | /** |
3131 | * t3_start_sge_timers - start SGE timer call backs | |
3132 | * @adap: the adapter | |
3133 | * | |
3134 | * Starts each SGE queue set's timer call back | |
3135 | */ | |
3136 | void t3_start_sge_timers(struct adapter *adap) | |
3137 | { | |
3138 | int i; | |
3139 | ||
3140 | for (i = 0; i < SGE_QSETS; ++i) { | |
3141 | struct sge_qset *q = &adap->sge.qs[i]; | |
3142 | ||
3143 | if (q->tx_reclaim_timer.function) | |
3144 | mod_timer(&q->tx_reclaim_timer, jiffies + TX_RECLAIM_PERIOD); | |
3145 | ||
3146 | if (q->rx_reclaim_timer.function) | |
3147 | mod_timer(&q->rx_reclaim_timer, jiffies + RX_RECLAIM_PERIOD); | |
3148 | } | |
3149 | } | |
3150 | ||
0ca41c04 DLR |
3151 | /** |
3152 | * t3_stop_sge_timers - stop SGE timer call backs | |
3153 | * @adap: the adapter | |
3154 | * | |
3155 | * Stops each SGE queue set's timer call back | |
3156 | */ | |
3157 | void t3_stop_sge_timers(struct adapter *adap) | |
3158 | { | |
3159 | int i; | |
3160 | ||
3161 | for (i = 0; i < SGE_QSETS; ++i) { | |
3162 | struct sge_qset *q = &adap->sge.qs[i]; | |
3163 | ||
3164 | if (q->tx_reclaim_timer.function) | |
3165 | del_timer_sync(&q->tx_reclaim_timer); | |
42c8ea17 DLR |
3166 | if (q->rx_reclaim_timer.function) |
3167 | del_timer_sync(&q->rx_reclaim_timer); | |
0ca41c04 DLR |
3168 | } |
3169 | } | |
3170 | ||
4d22de3e DLR |
3171 | /** |
3172 | * t3_free_sge_resources - free SGE resources | |
3173 | * @adap: the adapter | |
3174 | * | |
3175 | * Frees resources used by the SGE queue sets. | |
3176 | */ | |
3177 | void t3_free_sge_resources(struct adapter *adap) | |
3178 | { | |
3179 | int i; | |
3180 | ||
3181 | for (i = 0; i < SGE_QSETS; ++i) | |
3182 | t3_free_qset(adap, &adap->sge.qs[i]); | |
3183 | } | |
3184 | ||
3185 | /** | |
3186 | * t3_sge_start - enable SGE | |
3187 | * @adap: the adapter | |
3188 | * | |
3189 | * Enables the SGE for DMAs. This is the last step in starting packet | |
3190 | * transfers. | |
3191 | */ | |
3192 | void t3_sge_start(struct adapter *adap) | |
3193 | { | |
3194 | t3_set_reg_field(adap, A_SG_CONTROL, F_GLOBALENABLE, F_GLOBALENABLE); | |
3195 | } | |
3196 | ||
3197 | /** | |
3198 | * t3_sge_stop - disable SGE operation | |
3199 | * @adap: the adapter | |
3200 | * | |
3201 | * Disables the DMA engine. This can be called in emeregencies (e.g., | |
3202 | * from error interrupts) or from normal process context. In the latter | |
3203 | * case it also disables any pending queue restart tasklets. Note that | |
3204 | * if it is called in interrupt context it cannot disable the restart | |
3205 | * tasklets as it cannot wait, however the tasklets will have no effect | |
3206 | * since the doorbells are disabled and the driver will call this again | |
3207 | * later from process context, at which time the tasklets will be stopped | |
3208 | * if they are still running. | |
3209 | */ | |
3210 | void t3_sge_stop(struct adapter *adap) | |
3211 | { | |
3212 | t3_set_reg_field(adap, A_SG_CONTROL, F_GLOBALENABLE, 0); | |
3213 | if (!in_interrupt()) { | |
3214 | int i; | |
3215 | ||
3216 | for (i = 0; i < SGE_QSETS; ++i) { | |
3217 | struct sge_qset *qs = &adap->sge.qs[i]; | |
3218 | ||
3219 | tasklet_kill(&qs->txq[TXQ_OFLD].qresume_tsk); | |
3220 | tasklet_kill(&qs->txq[TXQ_CTRL].qresume_tsk); | |
3221 | } | |
3222 | } | |
3223 | } | |
3224 | ||
3225 | /** | |
3226 | * t3_sge_init - initialize SGE | |
3227 | * @adap: the adapter | |
3228 | * @p: the SGE parameters | |
3229 | * | |
3230 | * Performs SGE initialization needed every time after a chip reset. | |
3231 | * We do not initialize any of the queue sets here, instead the driver | |
3232 | * top-level must request those individually. We also do not enable DMA | |
3233 | * here, that should be done after the queues have been set up. | |
3234 | */ | |
3235 | void t3_sge_init(struct adapter *adap, struct sge_params *p) | |
3236 | { | |
3237 | unsigned int ctrl, ups = ffs(pci_resource_len(adap->pdev, 2) >> 12); | |
3238 | ||
3239 | ctrl = F_DROPPKT | V_PKTSHIFT(2) | F_FLMODE | F_AVOIDCQOVFL | | |
b881955b | 3240 | F_CQCRDTCTRL | F_CONGMODE | F_TNLFLMODE | F_FATLPERREN | |
4d22de3e DLR |
3241 | V_HOSTPAGESIZE(PAGE_SHIFT - 11) | F_BIGENDIANINGRESS | |
3242 | V_USERSPACESIZE(ups ? ups - 1 : 0) | F_ISCSICOALESCING; | |
3243 | #if SGE_NUM_GENBITS == 1 | |
3244 | ctrl |= F_EGRGENCTRL; | |
3245 | #endif | |
3246 | if (adap->params.rev > 0) { | |
3247 | if (!(adap->flags & (USING_MSIX | USING_MSI))) | |
3248 | ctrl |= F_ONEINTMULTQ | F_OPTONEINTMULTQ; | |
4d22de3e DLR |
3249 | } |
3250 | t3_write_reg(adap, A_SG_CONTROL, ctrl); | |
3251 | t3_write_reg(adap, A_SG_EGR_RCQ_DRB_THRSH, V_HIRCQDRBTHRSH(512) | | |
3252 | V_LORCQDRBTHRSH(512)); | |
3253 | t3_write_reg(adap, A_SG_TIMER_TICK, core_ticks_per_usec(adap) / 10); | |
3254 | t3_write_reg(adap, A_SG_CMDQ_CREDIT_TH, V_THRESHOLD(32) | | |
6195c71d | 3255 | V_TIMEOUT(200 * core_ticks_per_usec(adap))); |
b881955b DLR |
3256 | t3_write_reg(adap, A_SG_HI_DRB_HI_THRSH, |
3257 | adap->params.rev < T3_REV_C ? 1000 : 500); | |
4d22de3e DLR |
3258 | t3_write_reg(adap, A_SG_HI_DRB_LO_THRSH, 256); |
3259 | t3_write_reg(adap, A_SG_LO_DRB_HI_THRSH, 1000); | |
3260 | t3_write_reg(adap, A_SG_LO_DRB_LO_THRSH, 256); | |
3261 | t3_write_reg(adap, A_SG_OCO_BASE, V_BASE1(0xfff)); | |
3262 | t3_write_reg(adap, A_SG_DRB_PRI_THRESH, 63 * 1024); | |
3263 | } | |
3264 | ||
3265 | /** | |
3266 | * t3_sge_prep - one-time SGE initialization | |
3267 | * @adap: the associated adapter | |
3268 | * @p: SGE parameters | |
3269 | * | |
3270 | * Performs one-time initialization of SGE SW state. Includes determining | |
3271 | * defaults for the assorted SGE parameters, which admins can change until | |
3272 | * they are used to initialize the SGE. | |
3273 | */ | |
7b9b0943 | 3274 | void t3_sge_prep(struct adapter *adap, struct sge_params *p) |
4d22de3e DLR |
3275 | { |
3276 | int i; | |
3277 | ||
3278 | p->max_pkt_size = (16 * 1024) - sizeof(struct cpl_rx_data) - | |
3279 | SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); | |
3280 | ||
3281 | for (i = 0; i < SGE_QSETS; ++i) { | |
3282 | struct qset_params *q = p->qset + i; | |
3283 | ||
3284 | q->polling = adap->params.rev > 0; | |
3285 | q->coalesce_usecs = 5; | |
3286 | q->rspq_size = 1024; | |
e0994eb1 | 3287 | q->fl_size = 1024; |
7385ecf3 | 3288 | q->jumbo_size = 512; |
4d22de3e DLR |
3289 | q->txq_size[TXQ_ETH] = 1024; |
3290 | q->txq_size[TXQ_OFLD] = 1024; | |
3291 | q->txq_size[TXQ_CTRL] = 256; | |
3292 | q->cong_thres = 0; | |
3293 | } | |
3294 | ||
3295 | spin_lock_init(&adap->sge.reg_lock); | |
3296 | } | |
3297 | ||
3298 | /** | |
3299 | * t3_get_desc - dump an SGE descriptor for debugging purposes | |
3300 | * @qs: the queue set | |
3301 | * @qnum: identifies the specific queue (0..2: Tx, 3:response, 4..5: Rx) | |
3302 | * @idx: the descriptor index in the queue | |
3303 | * @data: where to dump the descriptor contents | |
3304 | * | |
3305 | * Dumps the contents of a HW descriptor of an SGE queue. Returns the | |
3306 | * size of the descriptor. | |
3307 | */ | |
3308 | int t3_get_desc(const struct sge_qset *qs, unsigned int qnum, unsigned int idx, | |
3309 | unsigned char *data) | |
3310 | { | |
3311 | if (qnum >= 6) | |
3312 | return -EINVAL; | |
3313 | ||
3314 | if (qnum < 3) { | |
3315 | if (!qs->txq[qnum].desc || idx >= qs->txq[qnum].size) | |
3316 | return -EINVAL; | |
3317 | memcpy(data, &qs->txq[qnum].desc[idx], sizeof(struct tx_desc)); | |
3318 | return sizeof(struct tx_desc); | |
3319 | } | |
3320 | ||
3321 | if (qnum == 3) { | |
3322 | if (!qs->rspq.desc || idx >= qs->rspq.size) | |
3323 | return -EINVAL; | |
3324 | memcpy(data, &qs->rspq.desc[idx], sizeof(struct rsp_desc)); | |
3325 | return sizeof(struct rsp_desc); | |
3326 | } | |
3327 | ||
3328 | qnum -= 4; | |
3329 | if (!qs->fl[qnum].desc || idx >= qs->fl[qnum].size) | |
3330 | return -EINVAL; | |
3331 | memcpy(data, &qs->fl[qnum].desc[idx], sizeof(struct rx_desc)); | |
3332 | return sizeof(struct rx_desc); | |
3333 | } |