summaryrefslogtreecommitdiffstats
path: root/drivers/dma/idxd/irq.c
blob: f4b0f59c95ba033e42b391e37fd4c5352ef0622e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
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
// SPDX-License-Identifier: GPL-2.0
/* Copyright(c) 2019 Intel Corporation. All rights rsvd. */
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/io-64-nonatomic-lo-hi.h>
#include <linux/dmaengine.h>
#include <linux/delay.h>
#include <linux/iommu.h>
#include <linux/sched/mm.h>
#include <uapi/linux/idxd.h>
#include "../dmaengine.h"
#include "idxd.h"
#include "registers.h"

enum irq_work_type {
	IRQ_WORK_NORMAL = 0,
	IRQ_WORK_PROCESS_FAULT,
};

struct idxd_resubmit {
	struct work_struct work;
	struct idxd_desc *desc;
};

struct idxd_int_handle_revoke {
	struct work_struct work;
	struct idxd_device *idxd;
};

static void idxd_device_reinit(struct work_struct *work)
{
	struct idxd_device *idxd = container_of(work, struct idxd_device, work);
	struct device *dev = &idxd->pdev->dev;
	int rc, i;

	idxd_device_reset(idxd);
	rc = idxd_device_config(idxd);
	if (rc < 0)
		goto out;

	rc = idxd_device_enable(idxd);
	if (rc < 0)
		goto out;

	for (i = 0; i < idxd->max_wqs; i++) {
		if (test_bit(i, idxd->wq_enable_map)) {
			struct idxd_wq *wq = idxd->wqs[i];

			rc = idxd_wq_enable(wq);
			if (rc < 0) {
				clear_bit(i, idxd->wq_enable_map);
				dev_warn(dev, "Unable to re-enable wq %s\n",
					 dev_name(wq_confdev(wq)));
			}
		}
	}

	return;

 out:
	idxd_device_clear_state(idxd);
}

/*
 * The function sends a drain descriptor for the interrupt handle. The drain ensures
 * all descriptors with this interrupt handle is flushed and the interrupt
 * will allow the cleanup of the outstanding descriptors.
 */
static void idxd_int_handle_revoke_drain(struct idxd_irq_entry *ie)
{
	struct idxd_wq *wq = ie_to_wq(ie);
	struct idxd_device *idxd = wq->idxd;
	struct device *dev = &idxd->pdev->dev;
	struct dsa_hw_desc desc = {};
	void __iomem *portal;
	int rc;

	/* Issue a simple drain operation with interrupt but no completion record */
	desc.flags = IDXD_OP_FLAG_RCI;
	desc.opcode = DSA_OPCODE_DRAIN;
	desc.priv = 1;

	if (ie->pasid != INVALID_IOASID)
		desc.pasid = ie->pasid;
	desc.int_handle = ie->int_handle;
	portal = idxd_wq_portal_addr(wq);

	/*
	 * The wmb() makes sure that the descriptor is all there before we
	 * issue.
	 */
	wmb();
	if (wq_dedicated(wq)) {
		iosubmit_cmds512(portal, &desc, 1);
	} else {
		rc = idxd_enqcmds(wq, portal, &desc);
		/* This should not fail unless hardware failed. */
		if (rc < 0)
			dev_warn(dev, "Failed to submit drain desc on wq %d\n", wq->id);
	}
}

static void idxd_abort_invalid_int_handle_descs(struct idxd_irq_entry *ie)
{
	LIST_HEAD(flist);
	struct idxd_desc *d, *t;
	struct llist_node *head;

	spin_lock(&ie->list_lock);
	head = llist_del_all(&ie->pending_llist);
	if (head) {
		llist_for_each_entry_safe(d, t, head, llnode)
			list_add_tail(&d->list, &ie->work_list);
	}

	list_for_each_entry_safe(d, t, &ie->work_list, list) {
		if (d->completion->status == DSA_COMP_INT_HANDLE_INVAL)
			list_move_tail(&d->list, &flist);
	}
	spin_unlock(&ie->list_lock);

	list_for_each_entry_safe(d, t, &flist, list) {
		list_del(&d->list);
		idxd_dma_complete_txd(d, IDXD_COMPLETE_ABORT, true);
	}
}

static void idxd_int_handle_revoke(struct work_struct *work)
{
	struct idxd_int_handle_revoke *revoke =
		container_of(work, struct idxd_int_handle_revoke, work);
	struct idxd_device *idxd = revoke->idxd;
	struct pci_dev *pdev = idxd->pdev;
	struct device *dev = &pdev->dev;
	int i, new_handle, rc;

	if (!idxd->request_int_handles) {
		kfree(revoke);
		dev_warn(dev, "Unexpected int handle refresh interrupt.\n");
		return;
	}

	/*
	 * The loop attempts to acquire new interrupt handle for all interrupt
	 * vectors that supports a handle. If a new interrupt handle is acquired and the
	 * wq is kernel type, the driver will kill the percpu_ref to pause all
	 * ongoing descriptor submissions. The interrupt handle is then changed.
	 * After change, the percpu_ref is revived and all the pending submissions
	 * are woken to try again. A drain is sent to for the interrupt handle
	 * at the end to make sure all invalid int handle descriptors are processed.
	 */
	for (i = 1; i < idxd->irq_cnt; i++) {
		struct idxd_irq_entry *ie = idxd_get_ie(idxd, i);
		struct idxd_wq *wq = ie_to_wq(ie);

		if (ie->int_handle == INVALID_INT_HANDLE)
			continue;

		rc = idxd_device_request_int_handle(idxd, i, &new_handle, IDXD_IRQ_MSIX);
		if (rc < 0) {
			dev_warn(dev, "get int handle %d failed: %d\n", i, rc);
			/*
			 * Failed to acquire new interrupt handle. Kill the WQ
			 * and release all the pending submitters. The submitters will
			 * get error return code and handle appropriately.
			 */
			ie->int_handle = INVALID_INT_HANDLE;
			idxd_wq_quiesce(wq);
			idxd_abort_invalid_int_handle_descs(ie);
			continue;
		}

		/* No change in interrupt handle, nothing needs to be done */
		if (ie->int_handle == new_handle)
			continue;

		if (wq->state != IDXD_WQ_ENABLED || wq->type != IDXD_WQT_KERNEL) {
			/*
			 * All the MSIX interrupts are allocated at once during probe.
			 * Therefore we need to update all interrupts even if the WQ
			 * isn't supporting interrupt operations.
			 */
			ie->int_handle = new_handle;
			continue;
		}

		mutex_lock(&wq->wq_lock);
		reinit_completion(&wq->wq_resurrect);

		/* Kill percpu_ref to pause additional descriptor submissions */
		percpu_ref_kill(&wq->wq_active);

		/* Wait for all submitters quiesce before we change interrupt handle */
		wait_for_completion(&wq->wq_dead);

		ie->int_handle = new_handle;

		/* Revive percpu ref and wake up all the waiting submitters */
		percpu_ref_reinit(&wq->wq_active);
		complete_all(&wq->wq_resurrect);
		mutex_unlock(&wq->wq_lock);

		/*
		 * The delay here is to wait for all possible MOVDIR64B that
		 * are issued before percpu_ref_kill() has happened to have
		 * reached the PCIe domain before the drain is issued. The driver
		 * needs to ensure that the drain descriptor issued does not pass
		 * all the other issued descriptors that contain the invalid
		 * interrupt handle in order to ensure that the drain descriptor
		 * interrupt will allow the cleanup of all the descriptors with
		 * invalid interrupt handle.
		 */
		if (wq_dedicated(wq))
			udelay(100);
		idxd_int_handle_revoke_drain(ie);
	}
	kfree(revoke);
}

static void idxd_evl_fault_work(struct work_struct *work)
{
	struct idxd_evl_fault *fault = container_of(work, struct idxd_evl_fault, work);
	struct idxd_wq *wq = fault->wq;
	struct idxd_device *idxd = wq->idxd;
	struct device *dev = &idxd->pdev->dev;
	struct idxd_evl *evl = idxd->evl;
	struct __evl_entry *entry_head = fault->entry;
	void *cr = (void *)entry_head + idxd->data->evl_cr_off;
	int cr_size = idxd->data->compl_size;
	u8 *status = (u8 *)cr + idxd->data->cr_status_off;
	u8 *result = (u8 *)cr + idxd->data->cr_result_off;
	int copied, copy_size;
	bool *bf;

	switch (fault->status) {
	case DSA_COMP_CRA_XLAT:
		if (entry_head->batch && entry_head->first_err_in_batch)
			evl->batch_fail[entry_head->batch_id] = false;

		copy_size = cr_size;
		idxd_user_counter_increment(wq, entry_head->pasid, COUNTER_FAULTS);
		break;
	case DSA_COMP_BATCH_EVL_ERR:
		bf = &evl->batch_fail[entry_head->batch_id];

		copy_size = entry_head->rcr || *bf ? cr_size : 0;
		if (*bf) {
			if (*status == DSA_COMP_SUCCESS)
				*status = DSA_COMP_BATCH_FAIL;
			*result = 1;
			*bf = false;
		}
		idxd_user_counter_increment(wq, entry_head->pasid, COUNTER_FAULTS);
		break;
	case DSA_COMP_DRAIN_EVL:
		copy_size = cr_size;
		break;
	default:
		copy_size = 0;
		dev_dbg_ratelimited(dev, "Unrecognized error code: %#x\n", fault->status);
		break;
	}

	if (copy_size == 0)
		return;

	/*
	 * Copy completion record to fault_addr in user address space
	 * that is found by wq and PASID.
	 */
	copied = idxd_copy_cr(wq, entry_head->pasid, entry_head->fault_addr,
			      cr, copy_size);
	/*
	 * The task that triggered the page fault is unknown currently
	 * because multiple threads may share the user address
	 * space or the task exits already before this fault.
	 * So if the copy fails, SIGSEGV can not be sent to the task.
	 * Just print an error for the failure. The user application
	 * waiting for the completion record will time out on this
	 * failure.
	 */
	switch (fault->status) {
	case DSA_COMP_CRA_XLAT:
		if (copied != copy_size) {
			idxd_user_counter_increment(wq, entry_head->pasid, COUNTER_FAULT_FAILS);
			dev_dbg_ratelimited(dev, "Failed to write to completion record: (%d:%d)\n",
					    copy_size, copied);
			if (entry_head->batch)
				evl->batch_fail[entry_head->batch_id] = true;
		}
		break;
	case DSA_COMP_BATCH_EVL_ERR:
		if (copied != copy_size) {
			idxd_user_counter_increment(wq, entry_head->pasid, COUNTER_FAULT_FAILS);
			dev_dbg_ratelimited(dev, "Failed to write to batch completion record: (%d:%d)\n",
					    copy_size, copied);
		}
		break;
	case DSA_COMP_DRAIN_EVL:
		if (copied != copy_size)
			dev_dbg_ratelimited(dev, "Failed to write to drain completion record: (%d:%d)\n",
					    copy_size, copied);
		break;
	}

	kmem_cache_free(idxd->evl_cache, fault);
}

static void process_evl_entry(struct idxd_device *idxd,
			      struct __evl_entry *entry_head, unsigned int index)
{
	struct device *dev = &idxd->pdev->dev;
	struct idxd_evl *evl = idxd->evl;
	u8 status;

	if (test_bit(index, evl->bmap)) {
		clear_bit(index, evl->bmap);
	} else {
		status = DSA_COMP_STATUS(entry_head->error);

		if (status == DSA_COMP_CRA_XLAT || status == DSA_COMP_DRAIN_EVL ||
		    status == DSA_COMP_BATCH_EVL_ERR) {
			struct idxd_evl_fault *fault;
			int ent_size = evl_ent_size(idxd);

			if (entry_head->rci)
				dev_dbg(dev, "Completion Int Req set, ignoring!\n");

			if (!entry_head->rcr && status == DSA_COMP_DRAIN_EVL)
				return;

			fault = kmem_cache_alloc(idxd->evl_cache, GFP_ATOMIC);
			if (fault) {
				struct idxd_wq *wq = idxd->wqs[entry_head->wq_idx];

				fault->wq = wq;
				fault->status = status;
				memcpy(&fault->entry, entry_head, ent_size);
				INIT_WORK(&fault->work, idxd_evl_fault_work);
				queue_work(wq->wq, &fault->work);
			} else {
				dev_warn(dev, "Failed to service fault work.\n");
			}
		} else {
			dev_warn_ratelimited(dev, "Device error %#x operation: %#x fault addr: %#llx\n",
					     status, entry_head->operation,
					     entry_head->fault_addr);
		}
	}
}

static void process_evl_entries(struct idxd_device *idxd)
{
	union evl_status_reg evl_status;
	unsigned int h, t;
	struct idxd_evl *evl = idxd->evl;
	struct __evl_entry *entry_head;
	unsigned int ent_size = evl_ent_size(idxd);
	u32 size;

	evl_status.bits = 0;
	evl_status.int_pending = 1;

	spin_lock(&evl->lock);
	/* Clear interrupt pending bit */
	iowrite32(evl_status.bits_upper32,
		  idxd->reg_base + IDXD_EVLSTATUS_OFFSET + sizeof(u32));
	h = evl->head;
	evl_status.bits = ioread64(idxd->reg_base + IDXD_EVLSTATUS_OFFSET);
	t = evl_status.tail;
	size = idxd->evl->size;

	while (h != t) {
		entry_head = (struct __evl_entry *)(evl->log + (h * ent_size));
		process_evl_entry(idxd, entry_head, h);
		h = (h + 1) % size;
	}

	evl->head = h;
	evl_status.head = h;
	iowrite32(evl_status.bits_lower32, idxd->reg_base + IDXD_EVLSTATUS_OFFSET);
	spin_unlock(&evl->lock);
}

irqreturn_t idxd_misc_thread(int vec, void *data)
{
	struct idxd_irq_entry *irq_entry = data;
	struct idxd_device *idxd = ie_to_idxd(irq_entry);
	struct device *dev = &idxd->pdev->dev;
	union gensts_reg gensts;
	u32 val = 0;
	int i;
	bool err = false;
	u32 cause;

	cause = ioread32(idxd->reg_base + IDXD_INTCAUSE_OFFSET);
	if (!cause)
		return IRQ_NONE;

	iowrite32(cause, idxd->reg_base + IDXD_INTCAUSE_OFFSET);

	if (cause & IDXD_INTC_HALT_STATE)
		goto halt;

	if (cause & IDXD_INTC_ERR) {
		spin_lock(&idxd->dev_lock);
		for (i = 0; i < 4; i++)
			idxd->sw_err.bits[i] = ioread64(idxd->reg_base +
					IDXD_SWERR_OFFSET + i * sizeof(u64));

		iowrite64(idxd->sw_err.bits[0] & IDXD_SWERR_ACK,
			  idxd->reg_base + IDXD_SWERR_OFFSET);

		if (idxd->sw_err.valid && idxd->sw_err.wq_idx_valid) {
			int id = idxd->sw_err.wq_idx;
			struct idxd_wq *wq = idxd->wqs[id];

			if (wq->type == IDXD_WQT_USER)
				wake_up_interruptible(&wq->err_queue);
		} else {
			int i;

			for (i = 0; i < idxd->max_wqs; i++) {
				struct idxd_wq *wq = idxd->wqs[i];

				if (wq->type == IDXD_WQT_USER)
					wake_up_interruptible(&wq->err_queue);
			}
		}

		spin_unlock(&idxd->dev_lock);
		val |= IDXD_INTC_ERR;

		for (i = 0; i < 4; i++)
			dev_warn(dev, "err[%d]: %#16.16llx\n",
				 i, idxd->sw_err.bits[i]);
		err = true;
	}

	if (cause & IDXD_INTC_INT_HANDLE_REVOKED) {
		struct idxd_int_handle_revoke *revoke;

		val |= IDXD_INTC_INT_HANDLE_REVOKED;

		revoke = kzalloc(sizeof(*revoke), GFP_ATOMIC);
		if (revoke) {
			revoke->idxd = idxd;
			INIT_WORK(&revoke->work, idxd_int_handle_revoke);
			queue_work(idxd->wq, &revoke->work);

		} else {
			dev_err(dev, "Failed to allocate work for int handle revoke\n");
			idxd_wqs_quiesce(idxd);
		}
	}

	if (cause & IDXD_INTC_CMD) {
		val |= IDXD_INTC_CMD;
		complete(idxd->cmd_done);
	}

	if (cause & IDXD_INTC_OCCUPY) {
		/* Driver does not utilize occupancy interrupt */
		val |= IDXD_INTC_OCCUPY;
	}

	if (cause & IDXD_INTC_PERFMON_OVFL) {
		val |= IDXD_INTC_PERFMON_OVFL;
		perfmon_counter_overflow(idxd);
	}

	if (cause & IDXD_INTC_EVL) {
		val |= IDXD_INTC_EVL;
		process_evl_entries(idxd);
	}

	val ^= cause;
	if (val)
		dev_warn_once(dev, "Unexpected interrupt cause bits set: %#x\n",
			      val);

	if (!err)
		goto out;

halt:
	gensts.bits = ioread32(idxd->reg_base + IDXD_GENSTATS_OFFSET);
	if (gensts.state == IDXD_DEVICE_STATE_HALT) {
		idxd->state = IDXD_DEV_HALTED;
		if (gensts.reset_type == IDXD_DEVICE_RESET_SOFTWARE) {
			/*
			 * If we need a software reset, we will throw the work
			 * on a system workqueue in order to allow interrupts
			 * for the device command completions.
			 */
			INIT_WORK(&idxd->work, idxd_device_reinit);
			queue_work(idxd->wq, &idxd->work);
		} else {
			idxd->state = IDXD_DEV_HALTED;
			idxd_wqs_quiesce(idxd);
			idxd_wqs_unmap_portal(idxd);
			idxd_device_clear_state(idxd);
			dev_err(&idxd->pdev->dev,
				"idxd halted, need %s.\n",
				gensts.reset_type == IDXD_DEVICE_RESET_FLR ?
				"FLR" : "system reset");
		}
	}

out:
	return IRQ_HANDLED;
}

static void idxd_int_handle_resubmit_work(struct work_struct *work)
{
	struct idxd_resubmit *irw = container_of(work, struct idxd_resubmit, work);
	struct idxd_desc *desc = irw->desc;
	struct idxd_wq *wq = desc->wq;
	int rc;

	desc->completion->status = 0;
	rc = idxd_submit_desc(wq, desc);
	if (rc < 0) {
		dev_dbg(&wq->idxd->pdev->dev, "Failed to resubmit desc %d to wq %d.\n",
			desc->id, wq->id);
		/*
		 * If the error is not -EAGAIN, it means the submission failed due to wq
		 * has been killed instead of ENQCMDS failure. Here the driver needs to
		 * notify the submitter of the failure by reporting abort status.
		 *
		 * -EAGAIN comes from ENQCMDS failure. idxd_submit_desc() will handle the
		 * abort.
		 */
		if (rc != -EAGAIN) {
			desc->completion->status = IDXD_COMP_DESC_ABORT;
			idxd_dma_complete_txd(desc, IDXD_COMPLETE_ABORT, false);
		}
		idxd_free_desc(wq, desc);
	}
	kfree(irw);
}

bool idxd_queue_int_handle_resubmit(struct idxd_desc *desc)
{
	struct idxd_wq *wq = desc->wq;
	struct idxd_device *idxd = wq->idxd;
	struct idxd_resubmit *irw;

	irw = kzalloc(sizeof(*irw), GFP_KERNEL);
	if (!irw)
		return false;

	irw->desc = desc;
	INIT_WORK(&irw->work, idxd_int_handle_resubmit_work);
	queue_work(idxd->wq, &irw->work);
	return true;
}

static void irq_process_pending_llist(struct idxd_irq_entry *irq_entry)
{
	struct idxd_desc *desc, *t;
	struct llist_node *head;

	head = llist_del_all(&irq_entry->pending_llist);
	if (!head)
		return;

	llist_for_each_entry_safe(desc, t, head, llnode) {
		u8 status = desc->completion->status & DSA_COMP_STATUS_MASK;

		if (status) {
			/*
			 * Check against the original status as ABORT is software defined
			 * and 0xff, which DSA_COMP_STATUS_MASK can mask out.
			 */
			if (unlikely(desc->completion->status == IDXD_COMP_DESC_ABORT)) {
				idxd_dma_complete_txd(desc, IDXD_COMPLETE_ABORT, true);
				continue;
			}

			idxd_dma_complete_txd(desc, IDXD_COMPLETE_NORMAL, true);
		} else {
			spin_lock(&irq_entry->list_lock);
			list_add_tail(&desc->list,
				      &irq_entry->work_list);
			spin_unlock(&irq_entry->list_lock);
		}
	}
}

static void irq_process_work_list(struct idxd_irq_entry *irq_entry)
{
	LIST_HEAD(flist);
	struct idxd_desc *desc, *n;

	/*
	 * This lock protects list corruption from access of list outside of the irq handler
	 * thread.
	 */
	spin_lock(&irq_entry->list_lock);
	if (list_empty(&irq_entry->work_list)) {
		spin_unlock(&irq_entry->list_lock);
		return;
	}

	list_for_each_entry_safe(desc, n, &irq_entry->work_list, list) {
		if (desc->completion->status) {
			list_move_tail(&desc->list, &flist);
		}
	}

	spin_unlock(&irq_entry->list_lock);

	list_for_each_entry(desc, &flist, list) {
		/*
		 * Check against the original status as ABORT is software defined
		 * and 0xff, which DSA_COMP_STATUS_MASK can mask out.
		 */
		if (unlikely(desc->completion->status == IDXD_COMP_DESC_ABORT)) {
			idxd_dma_complete_txd(desc, IDXD_COMPLETE_ABORT, true);
			continue;
		}

		idxd_dma_complete_txd(desc, IDXD_COMPLETE_NORMAL, true);
	}
}

irqreturn_t idxd_wq_thread(int irq, void *data)
{
	struct idxd_irq_entry *irq_entry = data;

	/*
	 * There are two lists we are processing. The pending_llist is where
	 * submmiter adds all the submitted descriptor after sending it to
	 * the workqueue. It's a lockless singly linked list. The work_list
	 * is the common linux double linked list. We are in a scenario of
	 * multiple producers and a single consumer. The producers are all
	 * the kernel submitters of descriptors, and the consumer is the
	 * kernel irq handler thread for the msix vector when using threaded
	 * irq. To work with the restrictions of llist to remain lockless,
	 * we are doing the following steps:
	 * 1. Iterate through the work_list and process any completed
	 *    descriptor. Delete the completed entries during iteration.
	 * 2. llist_del_all() from the pending list.
	 * 3. Iterate through the llist that was deleted from the pending list
	 *    and process the completed entries.
	 * 4. If the entry is still waiting on hardware, list_add_tail() to
	 *    the work_list.
	 */
	irq_process_work_list(irq_entry);
	irq_process_pending_llist(irq_entry);

	return IRQ_HANDLED;
}