1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
|
/*
* Copyright (C) 2016 Felix Fietkau <nbd@nbd.name>
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#include <linux/dma-mapping.h>
#include "mt76.h"
#include "dma.h"
#define DMA_DUMMY_TXWI ((void *) ~0)
static int
mt76_dma_alloc_queue(struct mt76_dev *dev, struct mt76_queue *q,
int idx, int n_desc, int bufsize,
u32 ring_base)
{
int size;
int i;
spin_lock_init(&q->lock);
q->regs = dev->mmio.regs + ring_base + idx * MT_RING_SIZE;
q->ndesc = n_desc;
q->buf_size = bufsize;
q->hw_idx = idx;
size = q->ndesc * sizeof(struct mt76_desc);
q->desc = dmam_alloc_coherent(dev->dev, size, &q->desc_dma, GFP_KERNEL);
if (!q->desc)
return -ENOMEM;
size = q->ndesc * sizeof(*q->entry);
q->entry = devm_kzalloc(dev->dev, size, GFP_KERNEL);
if (!q->entry)
return -ENOMEM;
/* clear descriptors */
for (i = 0; i < q->ndesc; i++)
q->desc[i].ctrl = cpu_to_le32(MT_DMA_CTL_DMA_DONE);
writel(q->desc_dma, &q->regs->desc_base);
writel(0, &q->regs->cpu_idx);
writel(0, &q->regs->dma_idx);
writel(q->ndesc, &q->regs->ring_size);
return 0;
}
static int
mt76_dma_add_buf(struct mt76_dev *dev, struct mt76_queue *q,
struct mt76_queue_buf *buf, int nbufs, u32 info,
struct sk_buff *skb, void *txwi)
{
struct mt76_desc *desc;
u32 ctrl;
int i, idx = -1;
if (txwi)
q->entry[q->head].txwi = DMA_DUMMY_TXWI;
for (i = 0; i < nbufs; i += 2, buf += 2) {
u32 buf0 = buf[0].addr, buf1 = 0;
ctrl = FIELD_PREP(MT_DMA_CTL_SD_LEN0, buf[0].len);
if (i < nbufs - 1) {
buf1 = buf[1].addr;
ctrl |= FIELD_PREP(MT_DMA_CTL_SD_LEN1, buf[1].len);
}
if (i == nbufs - 1)
ctrl |= MT_DMA_CTL_LAST_SEC0;
else if (i == nbufs - 2)
ctrl |= MT_DMA_CTL_LAST_SEC1;
idx = q->head;
q->head = (q->head + 1) % q->ndesc;
desc = &q->desc[idx];
WRITE_ONCE(desc->buf0, cpu_to_le32(buf0));
WRITE_ONCE(desc->buf1, cpu_to_le32(buf1));
WRITE_ONCE(desc->info, cpu_to_le32(info));
WRITE_ONCE(desc->ctrl, cpu_to_le32(ctrl));
q->queued++;
}
q->entry[idx].txwi = txwi;
q->entry[idx].skb = skb;
return idx;
}
static void
mt76_dma_tx_cleanup_idx(struct mt76_dev *dev, struct mt76_queue *q, int idx,
struct mt76_queue_entry *prev_e)
{
struct mt76_queue_entry *e = &q->entry[idx];
__le32 __ctrl = READ_ONCE(q->desc[idx].ctrl);
u32 ctrl = le32_to_cpu(__ctrl);
if (!e->txwi || !e->skb) {
__le32 addr = READ_ONCE(q->desc[idx].buf0);
u32 len = FIELD_GET(MT_DMA_CTL_SD_LEN0, ctrl);
dma_unmap_single(dev->dev, le32_to_cpu(addr), len,
DMA_TO_DEVICE);
}
if (!(ctrl & MT_DMA_CTL_LAST_SEC0)) {
__le32 addr = READ_ONCE(q->desc[idx].buf1);
u32 len = FIELD_GET(MT_DMA_CTL_SD_LEN1, ctrl);
dma_unmap_single(dev->dev, le32_to_cpu(addr), len,
DMA_TO_DEVICE);
}
if (e->txwi == DMA_DUMMY_TXWI)
e->txwi = NULL;
*prev_e = *e;
memset(e, 0, sizeof(*e));
}
static void
mt76_dma_sync_idx(struct mt76_dev *dev, struct mt76_queue *q)
{
writel(q->desc_dma, &q->regs->desc_base);
writel(q->ndesc, &q->regs->ring_size);
q->head = readl(&q->regs->dma_idx);
q->tail = q->head;
writel(q->head, &q->regs->cpu_idx);
}
static void
mt76_dma_tx_cleanup(struct mt76_dev *dev, enum mt76_txq_id qid, bool flush)
{
struct mt76_sw_queue *sq = &dev->q_tx[qid];
struct mt76_queue *q = sq->q;
struct mt76_queue_entry entry;
unsigned int n_swq_queued[4] = {};
unsigned int n_queued = 0;
bool wake = false;
int i, last;
if (!q)
return;
if (flush)
last = -1;
else
last = readl(&q->regs->dma_idx);
while ((q->queued > n_queued) && q->tail != last) {
mt76_dma_tx_cleanup_idx(dev, q, q->tail, &entry);
if (entry.schedule)
n_swq_queued[entry.qid]++;
q->tail = (q->tail + 1) % q->ndesc;
n_queued++;
if (entry.skb)
dev->drv->tx_complete_skb(dev, qid, &entry);
if (entry.txwi) {
mt76_put_txwi(dev, entry.txwi);
wake = !flush;
}
if (!flush && q->tail == last)
last = readl(&q->regs->dma_idx);
}
spin_lock_bh(&q->lock);
q->queued -= n_queued;
for (i = 0; i < ARRAY_SIZE(n_swq_queued); i++) {
if (!n_swq_queued[i])
continue;
dev->q_tx[i].swq_queued -= n_swq_queued[i];
}
if (flush)
mt76_dma_sync_idx(dev, q);
wake = wake && q->stopped &&
qid < IEEE80211_NUM_ACS && q->queued < q->ndesc - 8;
if (wake)
q->stopped = false;
if (!q->queued)
wake_up(&dev->tx_wait);
spin_unlock_bh(&q->lock);
if (wake)
ieee80211_wake_queue(dev->hw, qid);
}
static void *
mt76_dma_get_buf(struct mt76_dev *dev, struct mt76_queue *q, int idx,
int *len, u32 *info, bool *more)
{
struct mt76_queue_entry *e = &q->entry[idx];
struct mt76_desc *desc = &q->desc[idx];
dma_addr_t buf_addr;
void *buf = e->buf;
int buf_len = SKB_WITH_OVERHEAD(q->buf_size);
buf_addr = le32_to_cpu(READ_ONCE(desc->buf0));
if (len) {
u32 ctl = le32_to_cpu(READ_ONCE(desc->ctrl));
*len = FIELD_GET(MT_DMA_CTL_SD_LEN0, ctl);
*more = !(ctl & MT_DMA_CTL_LAST_SEC0);
}
if (info)
*info = le32_to_cpu(desc->info);
dma_unmap_single(dev->dev, buf_addr, buf_len, DMA_FROM_DEVICE);
e->buf = NULL;
return buf;
}
static void *
mt76_dma_dequeue(struct mt76_dev *dev, struct mt76_queue *q, bool flush,
int *len, u32 *info, bool *more)
{
int idx = q->tail;
*more = false;
if (!q->queued)
return NULL;
if (!flush && !(q->desc[idx].ctrl & cpu_to_le32(MT_DMA_CTL_DMA_DONE)))
return NULL;
q->tail = (q->tail + 1) % q->ndesc;
q->queued--;
return mt76_dma_get_buf(dev, q, idx, len, info, more);
}
static void
mt76_dma_kick_queue(struct mt76_dev *dev, struct mt76_queue *q)
{
writel(q->head, &q->regs->cpu_idx);
}
static int
mt76_dma_tx_queue_skb_raw(struct mt76_dev *dev, enum mt76_txq_id qid,
struct sk_buff *skb, u32 tx_info)
{
struct mt76_queue *q = dev->q_tx[qid].q;
struct mt76_queue_buf buf;
dma_addr_t addr;
addr = dma_map_single(dev->dev, skb->data, skb->len,
DMA_TO_DEVICE);
if (dma_mapping_error(dev->dev, addr))
return -ENOMEM;
buf.addr = addr;
buf.len = skb->len;
spin_lock_bh(&q->lock);
mt76_dma_add_buf(dev, q, &buf, 1, tx_info, skb, NULL);
mt76_dma_kick_queue(dev, q);
spin_unlock_bh(&q->lock);
return 0;
}
static int
mt76_dma_tx_queue_skb(struct mt76_dev *dev, enum mt76_txq_id qid,
struct sk_buff *skb, struct mt76_wcid *wcid,
struct ieee80211_sta *sta)
{
struct mt76_queue *q = dev->q_tx[qid].q;
struct mt76_tx_info tx_info = {
.skb = skb,
};
int len, n = 0, ret = -ENOMEM;
struct mt76_queue_entry e;
struct mt76_txwi_cache *t;
struct sk_buff *iter;
dma_addr_t addr;
u8 *txwi;
t = mt76_get_txwi(dev);
if (!t) {
ieee80211_free_txskb(dev->hw, skb);
return -ENOMEM;
}
txwi = mt76_get_txwi_ptr(dev, t);
skb->prev = skb->next = NULL;
if (dev->drv->tx_aligned4_skbs)
mt76_insert_hdr_pad(skb);
len = skb_headlen(skb);
addr = dma_map_single(dev->dev, skb->data, len, DMA_TO_DEVICE);
if (dma_mapping_error(dev->dev, addr))
goto free;
tx_info.buf[n].addr = t->dma_addr;
tx_info.buf[n++].len = dev->drv->txwi_size;
tx_info.buf[n].addr = addr;
tx_info.buf[n++].len = len;
skb_walk_frags(skb, iter) {
if (n == ARRAY_SIZE(tx_info.buf))
goto unmap;
addr = dma_map_single(dev->dev, iter->data, iter->len,
DMA_TO_DEVICE);
if (dma_mapping_error(dev->dev, addr))
goto unmap;
tx_info.buf[n].addr = addr;
tx_info.buf[n++].len = iter->len;
}
tx_info.nbuf = n;
dma_sync_single_for_cpu(dev->dev, t->dma_addr, dev->drv->txwi_size,
DMA_TO_DEVICE);
ret = dev->drv->tx_prepare_skb(dev, txwi, qid, wcid, sta, &tx_info);
dma_sync_single_for_device(dev->dev, t->dma_addr, dev->drv->txwi_size,
DMA_TO_DEVICE);
if (ret < 0)
goto unmap;
if (q->queued + (tx_info.nbuf + 1) / 2 >= q->ndesc - 1) {
ret = -ENOMEM;
goto unmap;
}
return mt76_dma_add_buf(dev, q, tx_info.buf, tx_info.nbuf,
tx_info.info, tx_info.skb, t);
unmap:
for (n--; n > 0; n--)
dma_unmap_single(dev->dev, tx_info.buf[n].addr,
tx_info.buf[n].len, DMA_TO_DEVICE);
free:
e.skb = tx_info.skb;
e.txwi = t;
dev->drv->tx_complete_skb(dev, qid, &e);
mt76_put_txwi(dev, t);
return ret;
}
static int
mt76_dma_rx_fill(struct mt76_dev *dev, struct mt76_queue *q)
{
dma_addr_t addr;
void *buf;
int frames = 0;
int len = SKB_WITH_OVERHEAD(q->buf_size);
int offset = q->buf_offset;
int idx;
spin_lock_bh(&q->lock);
while (q->queued < q->ndesc - 1) {
struct mt76_queue_buf qbuf;
buf = page_frag_alloc(&q->rx_page, q->buf_size, GFP_ATOMIC);
if (!buf)
break;
addr = dma_map_single(dev->dev, buf, len, DMA_FROM_DEVICE);
if (dma_mapping_error(dev->dev, addr)) {
skb_free_frag(buf);
break;
}
qbuf.addr = addr + offset;
qbuf.len = len - offset;
idx = mt76_dma_add_buf(dev, q, &qbuf, 1, 0, buf, NULL);
frames++;
}
if (frames)
mt76_dma_kick_queue(dev, q);
spin_unlock_bh(&q->lock);
return frames;
}
static void
mt76_dma_rx_cleanup(struct mt76_dev *dev, struct mt76_queue *q)
{
struct page *page;
void *buf;
bool more;
spin_lock_bh(&q->lock);
do {
buf = mt76_dma_dequeue(dev, q, true, NULL, NULL, &more);
if (!buf)
break;
skb_free_frag(buf);
} while (1);
spin_unlock_bh(&q->lock);
if (!q->rx_page.va)
return;
page = virt_to_page(q->rx_page.va);
__page_frag_cache_drain(page, q->rx_page.pagecnt_bias);
memset(&q->rx_page, 0, sizeof(q->rx_page));
}
static void
mt76_dma_rx_reset(struct mt76_dev *dev, enum mt76_rxq_id qid)
{
struct mt76_queue *q = &dev->q_rx[qid];
int i;
for (i = 0; i < q->ndesc; i++)
q->desc[i].ctrl &= ~cpu_to_le32(MT_DMA_CTL_DMA_DONE);
mt76_dma_rx_cleanup(dev, q);
mt76_dma_sync_idx(dev, q);
mt76_dma_rx_fill(dev, q);
}
static void
mt76_add_fragment(struct mt76_dev *dev, struct mt76_queue *q, void *data,
int len, bool more)
{
struct page *page = virt_to_head_page(data);
int offset = data - page_address(page);
struct sk_buff *skb = q->rx_head;
offset += q->buf_offset;
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, offset, len,
q->buf_size);
if (more)
return;
q->rx_head = NULL;
dev->drv->rx_skb(dev, q - dev->q_rx, skb);
}
static int
mt76_dma_rx_process(struct mt76_dev *dev, struct mt76_queue *q, int budget)
{
int len, data_len, done = 0;
struct sk_buff *skb;
unsigned char *data;
bool more;
while (done < budget) {
u32 info;
data = mt76_dma_dequeue(dev, q, false, &len, &info, &more);
if (!data)
break;
if (q->rx_head)
data_len = q->buf_size;
else
data_len = SKB_WITH_OVERHEAD(q->buf_size);
if (data_len < len + q->buf_offset) {
dev_kfree_skb(q->rx_head);
q->rx_head = NULL;
skb_free_frag(data);
continue;
}
if (q->rx_head) {
mt76_add_fragment(dev, q, data, len, more);
continue;
}
skb = build_skb(data, q->buf_size);
if (!skb) {
skb_free_frag(data);
continue;
}
skb_reserve(skb, q->buf_offset);
if (q == &dev->q_rx[MT_RXQ_MCU]) {
u32 *rxfce = (u32 *) skb->cb;
*rxfce = info;
}
__skb_put(skb, len);
done++;
if (more) {
q->rx_head = skb;
continue;
}
dev->drv->rx_skb(dev, q - dev->q_rx, skb);
}
mt76_dma_rx_fill(dev, q);
return done;
}
static int
mt76_dma_rx_poll(struct napi_struct *napi, int budget)
{
struct mt76_dev *dev;
int qid, done = 0, cur;
dev = container_of(napi->dev, struct mt76_dev, napi_dev);
qid = napi - dev->napi;
rcu_read_lock();
do {
cur = mt76_dma_rx_process(dev, &dev->q_rx[qid], budget - done);
mt76_rx_poll_complete(dev, qid, napi);
done += cur;
} while (cur && done < budget);
rcu_read_unlock();
if (done < budget) {
napi_complete(napi);
dev->drv->rx_poll_complete(dev, qid);
}
return done;
}
static int
mt76_dma_init(struct mt76_dev *dev)
{
int i;
init_dummy_netdev(&dev->napi_dev);
for (i = 0; i < ARRAY_SIZE(dev->q_rx); i++) {
netif_napi_add(&dev->napi_dev, &dev->napi[i], mt76_dma_rx_poll,
64);
mt76_dma_rx_fill(dev, &dev->q_rx[i]);
skb_queue_head_init(&dev->rx_skb[i]);
napi_enable(&dev->napi[i]);
}
return 0;
}
static const struct mt76_queue_ops mt76_dma_ops = {
.init = mt76_dma_init,
.alloc = mt76_dma_alloc_queue,
.tx_queue_skb_raw = mt76_dma_tx_queue_skb_raw,
.tx_queue_skb = mt76_dma_tx_queue_skb,
.tx_cleanup = mt76_dma_tx_cleanup,
.rx_reset = mt76_dma_rx_reset,
.kick = mt76_dma_kick_queue,
};
void mt76_dma_attach(struct mt76_dev *dev)
{
dev->queue_ops = &mt76_dma_ops;
}
EXPORT_SYMBOL_GPL(mt76_dma_attach);
void mt76_dma_cleanup(struct mt76_dev *dev)
{
int i;
for (i = 0; i < ARRAY_SIZE(dev->q_tx); i++)
mt76_dma_tx_cleanup(dev, i, true);
for (i = 0; i < ARRAY_SIZE(dev->q_rx); i++) {
netif_napi_del(&dev->napi[i]);
mt76_dma_rx_cleanup(dev, &dev->q_rx[i]);
}
}
EXPORT_SYMBOL_GPL(mt76_dma_cleanup);
|