Free Electrons

Embedded Linux Experts

  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
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
/*
 * Copyright (c) 2009-2010 Chelsio, Inc. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *	  copyright notice, this list of conditions and the following
 *	  disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *	  copyright notice, this list of conditions and the following
 *	  disclaimer in the documentation and/or other materials
 *	  provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

#include "iw_cxgb4.h"

static int destroy_cq(struct c4iw_rdev *rdev, struct t4_cq *cq,
		      struct c4iw_dev_ucontext *uctx)
{
	struct fw_ri_res_wr *res_wr;
	struct fw_ri_res *res;
	int wr_len;
	struct c4iw_wr_wait wr_wait;
	struct sk_buff *skb;
	int ret;

	wr_len = sizeof *res_wr + sizeof *res;
	skb = alloc_skb(wr_len, GFP_KERNEL);
	if (!skb)
		return -ENOMEM;
	set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0);

	res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len);
	memset(res_wr, 0, wr_len);
	res_wr->op_nres = cpu_to_be32(
			FW_WR_OP(FW_RI_RES_WR) |
			V_FW_RI_RES_WR_NRES(1) |
			FW_WR_COMPL(1));
	res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16));
	res_wr->cookie = (unsigned long) &wr_wait;
	res = res_wr->res;
	res->u.cq.restype = FW_RI_RES_TYPE_CQ;
	res->u.cq.op = FW_RI_RES_OP_RESET;
	res->u.cq.iqid = cpu_to_be32(cq->cqid);

	c4iw_init_wr_wait(&wr_wait);
	ret = c4iw_ofld_send(rdev, skb);
	if (!ret) {
		ret = c4iw_wait_for_reply(rdev, &wr_wait, 0, 0, __func__);
	}

	kfree(cq->sw_queue);
	dma_free_coherent(&(rdev->lldi.pdev->dev),
			  cq->memsize, cq->queue,
			  dma_unmap_addr(cq, mapping));
	c4iw_put_cqid(rdev, cq->cqid, uctx);
	return ret;
}

static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq,
		     struct c4iw_dev_ucontext *uctx)
{
	struct fw_ri_res_wr *res_wr;
	struct fw_ri_res *res;
	int wr_len;
	int user = (uctx != &rdev->uctx);
	struct c4iw_wr_wait wr_wait;
	int ret;
	struct sk_buff *skb;

	cq->cqid = c4iw_get_cqid(rdev, uctx);
	if (!cq->cqid) {
		ret = -ENOMEM;
		goto err1;
	}

	if (!user) {
		cq->sw_queue = kzalloc(cq->memsize, GFP_KERNEL);
		if (!cq->sw_queue) {
			ret = -ENOMEM;
			goto err2;
		}
	}
	cq->queue = dma_alloc_coherent(&rdev->lldi.pdev->dev, cq->memsize,
				       &cq->dma_addr, GFP_KERNEL);
	if (!cq->queue) {
		ret = -ENOMEM;
		goto err3;
	}
	dma_unmap_addr_set(cq, mapping, cq->dma_addr);
	memset(cq->queue, 0, cq->memsize);

	/* build fw_ri_res_wr */
	wr_len = sizeof *res_wr + sizeof *res;

	skb = alloc_skb(wr_len, GFP_KERNEL);
	if (!skb) {
		ret = -ENOMEM;
		goto err4;
	}
	set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0);

	res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len);
	memset(res_wr, 0, wr_len);
	res_wr->op_nres = cpu_to_be32(
			FW_WR_OP(FW_RI_RES_WR) |
			V_FW_RI_RES_WR_NRES(1) |
			FW_WR_COMPL(1));
	res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16));
	res_wr->cookie = (unsigned long) &wr_wait;
	res = res_wr->res;
	res->u.cq.restype = FW_RI_RES_TYPE_CQ;
	res->u.cq.op = FW_RI_RES_OP_WRITE;
	res->u.cq.iqid = cpu_to_be32(cq->cqid);
	res->u.cq.iqandst_to_iqandstindex = cpu_to_be32(
			V_FW_RI_RES_WR_IQANUS(0) |
			V_FW_RI_RES_WR_IQANUD(1) |
			F_FW_RI_RES_WR_IQANDST |
			V_FW_RI_RES_WR_IQANDSTINDEX(*rdev->lldi.rxq_ids));
	res->u.cq.iqdroprss_to_iqesize = cpu_to_be16(
			F_FW_RI_RES_WR_IQDROPRSS |
			V_FW_RI_RES_WR_IQPCIECH(2) |
			V_FW_RI_RES_WR_IQINTCNTTHRESH(0) |
			F_FW_RI_RES_WR_IQO |
			V_FW_RI_RES_WR_IQESIZE(1));
	res->u.cq.iqsize = cpu_to_be16(cq->size);
	res->u.cq.iqaddr = cpu_to_be64(cq->dma_addr);

	c4iw_init_wr_wait(&wr_wait);

	ret = c4iw_ofld_send(rdev, skb);
	if (ret)
		goto err4;
	PDBG("%s wait_event wr_wait %p\n", __func__, &wr_wait);
	ret = c4iw_wait_for_reply(rdev, &wr_wait, 0, 0, __func__);
	if (ret)
		goto err4;

	cq->gen = 1;
	cq->gts = rdev->lldi.gts_reg;
	cq->rdev = rdev;
	if (user) {
		cq->ugts = (u64)pci_resource_start(rdev->lldi.pdev, 2) +
					(cq->cqid << rdev->cqshift);
		cq->ugts &= PAGE_MASK;
	}
	return 0;
err4:
	dma_free_coherent(&rdev->lldi.pdev->dev, cq->memsize, cq->queue,
			  dma_unmap_addr(cq, mapping));
err3:
	kfree(cq->sw_queue);
err2:
	c4iw_put_cqid(rdev, cq->cqid, uctx);
err1:
	return ret;
}

static void insert_recv_cqe(struct t4_wq *wq, struct t4_cq *cq)
{
	struct t4_cqe cqe;

	PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__,
	     wq, cq, cq->sw_cidx, cq->sw_pidx);
	memset(&cqe, 0, sizeof(cqe));
	cqe.header = cpu_to_be32(V_CQE_STATUS(T4_ERR_SWFLUSH) |
				 V_CQE_OPCODE(FW_RI_SEND) |
				 V_CQE_TYPE(0) |
				 V_CQE_SWCQE(1) |
				 V_CQE_QPID(wq->sq.qid));
	cqe.bits_type_ts = cpu_to_be64(V_CQE_GENBIT((u64)cq->gen));
	cq->sw_queue[cq->sw_pidx] = cqe;
	t4_swcq_produce(cq);
}

int c4iw_flush_rq(struct t4_wq *wq, struct t4_cq *cq, int count)
{
	int flushed = 0;
	int in_use = wq->rq.in_use - count;

	BUG_ON(in_use < 0);
	PDBG("%s wq %p cq %p rq.in_use %u skip count %u\n", __func__,
	     wq, cq, wq->rq.in_use, count);
	while (in_use--) {
		insert_recv_cqe(wq, cq);
		flushed++;
	}
	return flushed;
}

static void insert_sq_cqe(struct t4_wq *wq, struct t4_cq *cq,
			  struct t4_swsqe *swcqe)
{
	struct t4_cqe cqe;

	PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__,
	     wq, cq, cq->sw_cidx, cq->sw_pidx);
	memset(&cqe, 0, sizeof(cqe));
	cqe.header = cpu_to_be32(V_CQE_STATUS(T4_ERR_SWFLUSH) |
				 V_CQE_OPCODE(swcqe->opcode) |
				 V_CQE_TYPE(1) |
				 V_CQE_SWCQE(1) |
				 V_CQE_QPID(wq->sq.qid));
	CQE_WRID_SQ_IDX(&cqe) = swcqe->idx;
	cqe.bits_type_ts = cpu_to_be64(V_CQE_GENBIT((u64)cq->gen));
	cq->sw_queue[cq->sw_pidx] = cqe;
	t4_swcq_produce(cq);
}

int c4iw_flush_sq(struct t4_wq *wq, struct t4_cq *cq, int count)
{
	int flushed = 0;
	struct t4_swsqe *swsqe = &wq->sq.sw_sq[wq->sq.cidx + count];
	int in_use = wq->sq.in_use - count;

	BUG_ON(in_use < 0);
	while (in_use--) {
		swsqe->signaled = 0;
		insert_sq_cqe(wq, cq, swsqe);
		swsqe++;
		if (swsqe == (wq->sq.sw_sq + wq->sq.size))
			swsqe = wq->sq.sw_sq;
		flushed++;
	}
	return flushed;
}

/*
 * Move all CQEs from the HWCQ into the SWCQ.
 */
void c4iw_flush_hw_cq(struct t4_cq *cq)
{
	struct t4_cqe *cqe = NULL, *swcqe;
	int ret;

	PDBG("%s cq %p cqid 0x%x\n", __func__, cq, cq->cqid);
	ret = t4_next_hw_cqe(cq, &cqe);
	while (!ret) {
		PDBG("%s flushing hwcq cidx 0x%x swcq pidx 0x%x\n",
		     __func__, cq->cidx, cq->sw_pidx);
		swcqe = &cq->sw_queue[cq->sw_pidx];
		*swcqe = *cqe;
		swcqe->header |= cpu_to_be32(V_CQE_SWCQE(1));
		t4_swcq_produce(cq);
		t4_hwcq_consume(cq);
		ret = t4_next_hw_cqe(cq, &cqe);
	}
}

static int cqe_completes_wr(struct t4_cqe *cqe, struct t4_wq *wq)
{
	if (CQE_OPCODE(cqe) == FW_RI_TERMINATE)
		return 0;

	if ((CQE_OPCODE(cqe) == FW_RI_RDMA_WRITE) && RQ_TYPE(cqe))
		return 0;

	if ((CQE_OPCODE(cqe) == FW_RI_READ_RESP) && SQ_TYPE(cqe))
		return 0;

	if (CQE_SEND_OPCODE(cqe) && RQ_TYPE(cqe) && t4_rq_empty(wq))
		return 0;
	return 1;
}

void c4iw_count_scqes(struct t4_cq *cq, struct t4_wq *wq, int *count)
{
	struct t4_cqe *cqe;
	u32 ptr;

	*count = 0;
	ptr = cq->sw_cidx;
	while (ptr != cq->sw_pidx) {
		cqe = &cq->sw_queue[ptr];
		if ((SQ_TYPE(cqe) || ((CQE_OPCODE(cqe) == FW_RI_READ_RESP) &&
				      wq->sq.oldest_read)) &&
		    (CQE_QPID(cqe) == wq->sq.qid))
			(*count)++;
		if (++ptr == cq->size)
			ptr = 0;
	}
	PDBG("%s cq %p count %d\n", __func__, cq, *count);
}

void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count)
{
	struct t4_cqe *cqe;
	u32 ptr;

	*count = 0;
	PDBG("%s count zero %d\n", __func__, *count);
	ptr = cq->sw_cidx;
	while (ptr != cq->sw_pidx) {
		cqe = &cq->sw_queue[ptr];
		if (RQ_TYPE(cqe) && (CQE_OPCODE(cqe) != FW_RI_READ_RESP) &&
		    (CQE_QPID(cqe) == wq->sq.qid) && cqe_completes_wr(cqe, wq))
			(*count)++;
		if (++ptr == cq->size)
			ptr = 0;
	}
	PDBG("%s cq %p count %d\n", __func__, cq, *count);
}

static void flush_completed_wrs(struct t4_wq *wq, struct t4_cq *cq)
{
	struct t4_swsqe *swsqe;
	u16 ptr = wq->sq.cidx;
	int count = wq->sq.in_use;
	int unsignaled = 0;

	swsqe = &wq->sq.sw_sq[ptr];
	while (count--)
		if (!swsqe->signaled) {
			if (++ptr == wq->sq.size)
				ptr = 0;
			swsqe = &wq->sq.sw_sq[ptr];
			unsignaled++;
		} else if (swsqe->complete) {

			/*
			 * Insert this completed cqe into the swcq.
			 */
			PDBG("%s moving cqe into swcq sq idx %u cq idx %u\n",
			     __func__, ptr, cq->sw_pidx);
			swsqe->cqe.header |= htonl(V_CQE_SWCQE(1));
			cq->sw_queue[cq->sw_pidx] = swsqe->cqe;
			t4_swcq_produce(cq);
			swsqe->signaled = 0;
			wq->sq.in_use -= unsignaled;
			break;
		} else
			break;
}

static void create_read_req_cqe(struct t4_wq *wq, struct t4_cqe *hw_cqe,
				struct t4_cqe *read_cqe)
{
	read_cqe->u.scqe.cidx = wq->sq.oldest_read->idx;
	read_cqe->len = cpu_to_be32(wq->sq.oldest_read->read_len);
	read_cqe->header = htonl(V_CQE_QPID(CQE_QPID(hw_cqe)) |
				 V_CQE_SWCQE(SW_CQE(hw_cqe)) |
				 V_CQE_OPCODE(FW_RI_READ_REQ) |
				 V_CQE_TYPE(1));
	read_cqe->bits_type_ts = hw_cqe->bits_type_ts;
}

/*
 * Return a ptr to the next read wr in the SWSQ or NULL.
 */
static void advance_oldest_read(struct t4_wq *wq)
{

	u32 rptr = wq->sq.oldest_read - wq->sq.sw_sq + 1;

	if (rptr == wq->sq.size)
		rptr = 0;
	while (rptr != wq->sq.pidx) {
		wq->sq.oldest_read = &wq->sq.sw_sq[rptr];

		if (wq->sq.oldest_read->opcode == FW_RI_READ_REQ)
			return;
		if (++rptr == wq->sq.size)
			rptr = 0;
	}
	wq->sq.oldest_read = NULL;
}

/*
 * poll_cq
 *
 * Caller must:
 *     check the validity of the first CQE,
 *     supply the wq assicated with the qpid.
 *
 * credit: cq credit to return to sge.
 * cqe_flushed: 1 iff the CQE is flushed.
 * cqe: copy of the polled CQE.
 *
 * return value:
 *    0		    CQE returned ok.
 *    -EAGAIN       CQE skipped, try again.
 *    -EOVERFLOW    CQ overflow detected.
 */
static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe,
		   u8 *cqe_flushed, u64 *cookie, u32 *credit)
{
	int ret = 0;
	struct t4_cqe *hw_cqe, read_cqe;

	*cqe_flushed = 0;
	*credit = 0;
	ret = t4_next_cqe(cq, &hw_cqe);
	if (ret)
		return ret;

	PDBG("%s CQE OVF %u qpid 0x%0x genbit %u type %u status 0x%0x"
	     " opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n",
	     __func__, CQE_OVFBIT(hw_cqe), CQE_QPID(hw_cqe),
	     CQE_GENBIT(hw_cqe), CQE_TYPE(hw_cqe), CQE_STATUS(hw_cqe),
	     CQE_OPCODE(hw_cqe), CQE_LEN(hw_cqe), CQE_WRID_HI(hw_cqe),
	     CQE_WRID_LOW(hw_cqe));

	/*
	 * skip cqe's not affiliated with a QP.
	 */
	if (wq == NULL) {
		ret = -EAGAIN;
		goto skip_cqe;
	}

	/*
	 * Gotta tweak READ completions:
	 *	1) the cqe doesn't contain the sq_wptr from the wr.
	 *	2) opcode not reflected from the wr.
	 *	3) read_len not reflected from the wr.
	 *	4) cq_type is RQ_TYPE not SQ_TYPE.
	 */
	if (RQ_TYPE(hw_cqe) && (CQE_OPCODE(hw_cqe) == FW_RI_READ_RESP)) {

		/*
		 * If this is an unsolicited read response, then the read
		 * was generated by the kernel driver as part of peer-2-peer
		 * connection setup.  So ignore the completion.
		 */
		if (!wq->sq.oldest_read) {
			if (CQE_STATUS(hw_cqe))
				t4_set_wq_in_error(wq);
			ret = -EAGAIN;
			goto skip_cqe;
		}

		/*
		 * Don't write to the HWCQ, so create a new read req CQE
		 * in local memory.
		 */
		create_read_req_cqe(wq, hw_cqe, &read_cqe);
		hw_cqe = &read_cqe;
		advance_oldest_read(wq);
	}

	if (CQE_STATUS(hw_cqe) || t4_wq_in_error(wq)) {
		*cqe_flushed = t4_wq_in_error(wq);
		t4_set_wq_in_error(wq);
		goto proc_cqe;
	}

	if (CQE_OPCODE(hw_cqe) == FW_RI_TERMINATE) {
		ret = -EAGAIN;
		goto skip_cqe;
	}

	/*
	 * RECV completion.
	 */
	if (RQ_TYPE(hw_cqe)) {

		/*
		 * HW only validates 4 bits of MSN.  So we must validate that
		 * the MSN in the SEND is the next expected MSN.  If its not,
		 * then we complete this with T4_ERR_MSN and mark the wq in
		 * error.
		 */

		if (t4_rq_empty(wq)) {
			t4_set_wq_in_error(wq);
			ret = -EAGAIN;
			goto skip_cqe;
		}
		if (unlikely((CQE_WRID_MSN(hw_cqe) != (wq->rq.msn)))) {
			t4_set_wq_in_error(wq);
			hw_cqe->header |= htonl(V_CQE_STATUS(T4_ERR_MSN));
			goto proc_cqe;
		}
		goto proc_cqe;
	}

	/*
	 * If we get here its a send completion.
	 *
	 * Handle out of order completion. These get stuffed
	 * in the SW SQ. Then the SW SQ is walked to move any
	 * now in-order completions into the SW CQ.  This handles
	 * 2 cases:
	 *	1) reaping unsignaled WRs when the first subsequent
	 *	   signaled WR is completed.
	 *	2) out of order read completions.
	 */
	if (!SW_CQE(hw_cqe) && (CQE_WRID_SQ_IDX(hw_cqe) != wq->sq.cidx)) {
		struct t4_swsqe *swsqe;

		PDBG("%s out of order completion going in sw_sq at idx %u\n",
		     __func__, CQE_WRID_SQ_IDX(hw_cqe));
		swsqe = &wq->sq.sw_sq[CQE_WRID_SQ_IDX(hw_cqe)];
		swsqe->cqe = *hw_cqe;
		swsqe->complete = 1;
		ret = -EAGAIN;
		goto flush_wq;
	}

proc_cqe:
	*cqe = *hw_cqe;

	/*
	 * Reap the associated WR(s) that are freed up with this
	 * completion.
	 */
	if (SQ_TYPE(hw_cqe)) {
		wq->sq.cidx = CQE_WRID_SQ_IDX(hw_cqe);
		PDBG("%s completing sq idx %u\n", __func__, wq->sq.cidx);
		*cookie = wq->sq.sw_sq[wq->sq.cidx].wr_id;
		t4_sq_consume(wq);
	} else {
		PDBG("%s completing rq idx %u\n", __func__, wq->rq.cidx);
		*cookie = wq->rq.sw_rq[wq->rq.cidx].wr_id;
		BUG_ON(t4_rq_empty(wq));
		t4_rq_consume(wq);
	}

flush_wq:
	/*
	 * Flush any completed cqes that are now in-order.
	 */
	flush_completed_wrs(wq, cq);

skip_cqe:
	if (SW_CQE(hw_cqe)) {
		PDBG("%s cq %p cqid 0x%x skip sw cqe cidx %u\n",
		     __func__, cq, cq->cqid, cq->sw_cidx);
		t4_swcq_consume(cq);
	} else {
		PDBG("%s cq %p cqid 0x%x skip hw cqe cidx %u\n",
		     __func__, cq, cq->cqid, cq->cidx);
		t4_hwcq_consume(cq);
	}
	return ret;
}

/*
 * Get one cq entry from c4iw and map it to openib.
 *
 * Returns:
 *	0			cqe returned
 *	-ENODATA		EMPTY;
 *	-EAGAIN			caller must try again
 *	any other -errno	fatal error
 */
static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc)
{
	struct c4iw_qp *qhp = NULL;
	struct t4_cqe cqe = {0, 0}, *rd_cqe;
	struct t4_wq *wq;
	u32 credit = 0;
	u8 cqe_flushed;
	u64 cookie = 0;
	int ret;

	ret = t4_next_cqe(&chp->cq, &rd_cqe);

	if (ret)
		return ret;

	qhp = get_qhp(chp->rhp, CQE_QPID(rd_cqe));
	if (!qhp)
		wq = NULL;
	else {
		spin_lock(&qhp->lock);
		wq = &(qhp->wq);
	}
	ret = poll_cq(wq, &(chp->cq), &cqe, &cqe_flushed, &cookie, &credit);
	if (ret)
		goto out;

	wc->wr_id = cookie;
	wc->qp = &qhp->ibqp;
	wc->vendor_err = CQE_STATUS(&cqe);
	wc->wc_flags = 0;

	PDBG("%s qpid 0x%x type %d opcode %d status 0x%x len %u wrid hi 0x%x "
	     "lo 0x%x cookie 0x%llx\n", __func__, CQE_QPID(&cqe),
	     CQE_TYPE(&cqe), CQE_OPCODE(&cqe), CQE_STATUS(&cqe), CQE_LEN(&cqe),
	     CQE_WRID_HI(&cqe), CQE_WRID_LOW(&cqe), (unsigned long long)cookie);

	if (CQE_TYPE(&cqe) == 0) {
		if (!CQE_STATUS(&cqe))
			wc->byte_len = CQE_LEN(&cqe);
		else
			wc->byte_len = 0;
		wc->opcode = IB_WC_RECV;
		if (CQE_OPCODE(&cqe) == FW_RI_SEND_WITH_INV ||
		    CQE_OPCODE(&cqe) == FW_RI_SEND_WITH_SE_INV) {
			wc->ex.invalidate_rkey = CQE_WRID_STAG(&cqe);
			wc->wc_flags |= IB_WC_WITH_INVALIDATE;
		}
	} else {
		switch (CQE_OPCODE(&cqe)) {
		case FW_RI_RDMA_WRITE:
			wc->opcode = IB_WC_RDMA_WRITE;
			break;
		case FW_RI_READ_REQ:
			wc->opcode = IB_WC_RDMA_READ;
			wc->byte_len = CQE_LEN(&cqe);
			break;
		case FW_RI_SEND_WITH_INV:
		case FW_RI_SEND_WITH_SE_INV:
			wc->opcode = IB_WC_SEND;
			wc->wc_flags |= IB_WC_WITH_INVALIDATE;
			break;
		case FW_RI_SEND:
		case FW_RI_SEND_WITH_SE:
			wc->opcode = IB_WC_SEND;
			break;
		case FW_RI_BIND_MW:
			wc->opcode = IB_WC_BIND_MW;
			break;

		case FW_RI_LOCAL_INV:
			wc->opcode = IB_WC_LOCAL_INV;
			break;
		case FW_RI_FAST_REGISTER:
			wc->opcode = IB_WC_FAST_REG_MR;
			break;
		default:
			printk(KERN_ERR MOD "Unexpected opcode %d "
			       "in the CQE received for QPID=0x%0x\n",
			       CQE_OPCODE(&cqe), CQE_QPID(&cqe));
			ret = -EINVAL;
			goto out;
		}
	}

	if (cqe_flushed)
		wc->status = IB_WC_WR_FLUSH_ERR;
	else {

		switch (CQE_STATUS(&cqe)) {
		case T4_ERR_SUCCESS:
			wc->status = IB_WC_SUCCESS;
			break;
		case T4_ERR_STAG:
			wc->status = IB_WC_LOC_ACCESS_ERR;
			break;
		case T4_ERR_PDID:
			wc->status = IB_WC_LOC_PROT_ERR;
			break;
		case T4_ERR_QPID:
		case T4_ERR_ACCESS:
			wc->status = IB_WC_LOC_ACCESS_ERR;
			break;
		case T4_ERR_WRAP:
			wc->status = IB_WC_GENERAL_ERR;
			break;
		case T4_ERR_BOUND:
			wc->status = IB_WC_LOC_LEN_ERR;
			break;
		case T4_ERR_INVALIDATE_SHARED_MR:
		case T4_ERR_INVALIDATE_MR_WITH_MW_BOUND:
			wc->status = IB_WC_MW_BIND_ERR;
			break;
		case T4_ERR_CRC:
		case T4_ERR_MARKER:
		case T4_ERR_PDU_LEN_ERR:
		case T4_ERR_OUT_OF_RQE:
		case T4_ERR_DDP_VERSION:
		case T4_ERR_RDMA_VERSION:
		case T4_ERR_DDP_QUEUE_NUM:
		case T4_ERR_MSN:
		case T4_ERR_TBIT:
		case T4_ERR_MO:
		case T4_ERR_MSN_RANGE:
		case T4_ERR_IRD_OVERFLOW:
		case T4_ERR_OPCODE:
		case T4_ERR_INTERNAL_ERR:
			wc->status = IB_WC_FATAL_ERR;
			break;
		case T4_ERR_SWFLUSH:
			wc->status = IB_WC_WR_FLUSH_ERR;
			break;
		default:
			printk(KERN_ERR MOD
			       "Unexpected cqe_status 0x%x for QPID=0x%0x\n",
			       CQE_STATUS(&cqe), CQE_QPID(&cqe));
			ret = -EINVAL;
		}
	}
out:
	if (wq)
		spin_unlock(&qhp->lock);
	return ret;
}

int c4iw_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc)
{
	struct c4iw_cq *chp;
	unsigned long flags;
	int npolled;
	int err = 0;

	chp = to_c4iw_cq(ibcq);

	spin_lock_irqsave(&chp->lock, flags);
	for (npolled = 0; npolled < num_entries; ++npolled) {
		do {
			err = c4iw_poll_cq_one(chp, wc + npolled);
		} while (err == -EAGAIN);
		if (err)
			break;
	}
	spin_unlock_irqrestore(&chp->lock, flags);
	return !err || err == -ENODATA ? npolled : err;
}

int c4iw_destroy_cq(struct ib_cq *ib_cq)
{
	struct c4iw_cq *chp;
	struct c4iw_ucontext *ucontext;

	PDBG("%s ib_cq %p\n", __func__, ib_cq);
	chp = to_c4iw_cq(ib_cq);

	remove_handle(chp->rhp, &chp->rhp->cqidr, chp->cq.cqid);
	atomic_dec(&chp->refcnt);
	wait_event(chp->wait, !atomic_read(&chp->refcnt));

	ucontext = ib_cq->uobject ? to_c4iw_ucontext(ib_cq->uobject->context)
				  : NULL;
	destroy_cq(&chp->rhp->rdev, &chp->cq,
		   ucontext ? &ucontext->uctx : &chp->cq.rdev->uctx);
	kfree(chp);
	return 0;
}

struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries,
			     int vector, struct ib_ucontext *ib_context,
			     struct ib_udata *udata)
{
	struct c4iw_dev *rhp;
	struct c4iw_cq *chp;
	struct c4iw_create_cq_resp uresp;
	struct c4iw_ucontext *ucontext = NULL;
	int ret;
	size_t memsize, hwentries;
	struct c4iw_mm_entry *mm, *mm2;

	PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries);

	rhp = to_c4iw_dev(ibdev);

	chp = kzalloc(sizeof(*chp), GFP_KERNEL);
	if (!chp)
		return ERR_PTR(-ENOMEM);

	if (ib_context)
		ucontext = to_c4iw_ucontext(ib_context);

	/* account for the status page. */
	entries++;

	/* IQ needs one extra entry to differentiate full vs empty. */
	entries++;

	/*
	 * entries must be multiple of 16 for HW.
	 */
	entries = roundup(entries, 16);

	/*
	 * Make actual HW queue 2x to avoid cdix_inc overflows.
	 */
	hwentries = entries * 2;

	/*
	 * Make HW queue at least 64 entries so GTS updates aren't too
	 * frequent.
	 */
	if (hwentries < 64)
		hwentries = 64;

	memsize = hwentries * sizeof *chp->cq.queue;

	/*
	 * memsize must be a multiple of the page size if its a user cq.
	 */
	if (ucontext) {
		memsize = roundup(memsize, PAGE_SIZE);
		hwentries = memsize / sizeof *chp->cq.queue;
		while (hwentries > T4_MAX_IQ_SIZE) {
			memsize -= PAGE_SIZE;
			hwentries = memsize / sizeof *chp->cq.queue;
		}
	}
	chp->cq.size = hwentries;
	chp->cq.memsize = memsize;

	ret = create_cq(&rhp->rdev, &chp->cq,
			ucontext ? &ucontext->uctx : &rhp->rdev.uctx);
	if (ret)
		goto err1;

	chp->rhp = rhp;
	chp->cq.size--;				/* status page */
	chp->ibcq.cqe = entries - 2;
	spin_lock_init(&chp->lock);
	spin_lock_init(&chp->comp_handler_lock);
	atomic_set(&chp->refcnt, 1);
	init_waitqueue_head(&chp->wait);
	ret = insert_handle(rhp, &rhp->cqidr, chp, chp->cq.cqid);
	if (ret)
		goto err2;

	if (ucontext) {
		mm = kmalloc(sizeof *mm, GFP_KERNEL);
		if (!mm)
			goto err3;
		mm2 = kmalloc(sizeof *mm2, GFP_KERNEL);
		if (!mm2)
			goto err4;

		uresp.qid_mask = rhp->rdev.cqmask;
		uresp.cqid = chp->cq.cqid;
		uresp.size = chp->cq.size;
		uresp.memsize = chp->cq.memsize;
		spin_lock(&ucontext->mmap_lock);
		uresp.key = ucontext->key;
		ucontext->key += PAGE_SIZE;
		uresp.gts_key = ucontext->key;
		ucontext->key += PAGE_SIZE;
		spin_unlock(&ucontext->mmap_lock);
		ret = ib_copy_to_udata(udata, &uresp, sizeof uresp);
		if (ret)
			goto err5;

		mm->key = uresp.key;
		mm->addr = virt_to_phys(chp->cq.queue);
		mm->len = chp->cq.memsize;
		insert_mmap(ucontext, mm);

		mm2->key = uresp.gts_key;
		mm2->addr = chp->cq.ugts;
		mm2->len = PAGE_SIZE;
		insert_mmap(ucontext, mm2);
	}
	PDBG("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n",
	     __func__, chp->cq.cqid, chp, chp->cq.size,
	     chp->cq.memsize,
	     (unsigned long long) chp->cq.dma_addr);
	return &chp->ibcq;
err5:
	kfree(mm2);
err4:
	kfree(mm);
err3:
	remove_handle(rhp, &rhp->cqidr, chp->cq.cqid);
err2:
	destroy_cq(&chp->rhp->rdev, &chp->cq,
		   ucontext ? &ucontext->uctx : &rhp->rdev.uctx);
err1:
	kfree(chp);
	return ERR_PTR(ret);
}

int c4iw_resize_cq(struct ib_cq *cq, int cqe, struct ib_udata *udata)
{
	return -ENOSYS;
}

int c4iw_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags)
{
	struct c4iw_cq *chp;
	int ret;
	unsigned long flag;

	chp = to_c4iw_cq(ibcq);
	spin_lock_irqsave(&chp->lock, flag);
	ret = t4_arm_cq(&chp->cq,
			(flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED);
	spin_unlock_irqrestore(&chp->lock, flag);
	if (ret && !(flags & IB_CQ_REPORT_MISSED_EVENTS))
		ret = 0;
	return ret;
}