iop-adma.c 39.3 KB
Newer Older
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
/*
 * offload engine driver for the Intel Xscale series of i/o processors
 * Copyright © 2006, Intel Corporation.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 *
 * You should have received a copy of the GNU General Public License along with
 * this program; if not, write to the Free Software Foundation, Inc.,
 * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
 *
 */

/*
 * This driver supports the asynchrounous DMA copy and RAID engines available
 * on the Intel Xscale(R) family of I/O Processors (IOP 32x, 33x, 134x)
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/async_tx.h>
#include <linux/delay.h>
#include <linux/dma-mapping.h>
#include <linux/spinlock.h>
#include <linux/interrupt.h>
#include <linux/platform_device.h>
#include <linux/memory.h>
#include <linux/ioport.h>

#include <asm/arch/adma.h>

#define to_iop_adma_chan(chan) container_of(chan, struct iop_adma_chan, common)
#define to_iop_adma_device(dev) \
	container_of(dev, struct iop_adma_device, common)
#define tx_to_iop_adma_slot(tx) \
	container_of(tx, struct iop_adma_desc_slot, async_tx)

/**
 * iop_adma_free_slots - flags descriptor slots for reuse
 * @slot: Slot to free
 * Caller must hold &iop_chan->lock while calling this function
 */
static void iop_adma_free_slots(struct iop_adma_desc_slot *slot)
{
	int stride = slot->slots_per_op;

	while (stride--) {
		slot->slots_per_op = 0;
		slot = list_entry(slot->slot_node.next,
				struct iop_adma_desc_slot,
				slot_node);
	}
}

static dma_cookie_t
iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
	struct iop_adma_chan *iop_chan, dma_cookie_t cookie)
{
	BUG_ON(desc->async_tx.cookie < 0);
	spin_lock_bh(&desc->async_tx.lock);
	if (desc->async_tx.cookie > 0) {
		cookie = desc->async_tx.cookie;
		desc->async_tx.cookie = 0;

		/* call the callback (must not sleep or submit new
		 * operations to this channel)
		 */
		if (desc->async_tx.callback)
			desc->async_tx.callback(
				desc->async_tx.callback_param);

		/* unmap dma addresses
		 * (unmap_single vs unmap_page?)
		 */
		if (desc->group_head && desc->unmap_len) {
			struct iop_adma_desc_slot *unmap = desc->group_head;
			struct device *dev =
				&iop_chan->device->pdev->dev;
			u32 len = unmap->unmap_len;
			u32 src_cnt = unmap->unmap_src_cnt;
			dma_addr_t addr = iop_desc_get_dest_addr(unmap,
				iop_chan);

			dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
			while (src_cnt--) {
				addr = iop_desc_get_src_addr(unmap,
							iop_chan,
							src_cnt);
				dma_unmap_page(dev, addr, len,
					DMA_TO_DEVICE);
			}
			desc->group_head = NULL;
		}
	}

	/* run dependent operations */
	async_tx_run_dependencies(&desc->async_tx);
	spin_unlock_bh(&desc->async_tx.lock);

	return cookie;
}

static int
iop_adma_clean_slot(struct iop_adma_desc_slot *desc,
	struct iop_adma_chan *iop_chan)
{
	/* the client is allowed to attach dependent operations
	 * until 'ack' is set
	 */
	if (!desc->async_tx.ack)
		return 0;

	/* leave the last descriptor in the chain
	 * so we can append to it
	 */
	if (desc->chain_node.next == &iop_chan->chain)
		return 1;

	dev_dbg(iop_chan->device->common.dev,
		"\tfree slot: %d slots_per_op: %d\n",
		desc->idx, desc->slots_per_op);

	list_del(&desc->chain_node);
	iop_adma_free_slots(desc);

	return 0;
}

static void __iop_adma_slot_cleanup(struct iop_adma_chan *iop_chan)
{
	struct iop_adma_desc_slot *iter, *_iter, *grp_start = NULL;
	dma_cookie_t cookie = 0;
	u32 current_desc = iop_chan_get_current_descriptor(iop_chan);
	int busy = iop_chan_is_busy(iop_chan);
	int seen_current = 0, slot_cnt = 0, slots_per_op = 0;

143
	dev_dbg(iop_chan->device->common.dev, "%s\n", __func__);
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
	/* free completed slots from the chain starting with
	 * the oldest descriptor
	 */
	list_for_each_entry_safe(iter, _iter, &iop_chan->chain,
					chain_node) {
		pr_debug("\tcookie: %d slot: %d busy: %d "
			"this_desc: %#x next_desc: %#x ack: %d\n",
			iter->async_tx.cookie, iter->idx, busy,
			iter->async_tx.phys, iop_desc_get_next_desc(iter),
			iter->async_tx.ack);
		prefetch(_iter);
		prefetch(&_iter->async_tx);

		/* do not advance past the current descriptor loaded into the
		 * hardware channel, subsequent descriptors are either in
		 * process or have not been submitted
		 */
		if (seen_current)
			break;

		/* stop the search if we reach the current descriptor and the
		 * channel is busy, or if it appears that the current descriptor
		 * needs to be re-read (i.e. has been appended to)
		 */
		if (iter->async_tx.phys == current_desc) {
			BUG_ON(seen_current++);
			if (busy || iop_desc_get_next_desc(iter))
				break;
		}

		/* detect the start of a group transaction */
		if (!slot_cnt && !slots_per_op) {
			slot_cnt = iter->slot_cnt;
			slots_per_op = iter->slots_per_op;
			if (slot_cnt <= slots_per_op) {
				slot_cnt = 0;
				slots_per_op = 0;
			}
		}

		if (slot_cnt) {
			pr_debug("\tgroup++\n");
			if (!grp_start)
				grp_start = iter;
			slot_cnt -= slots_per_op;
		}

		/* all the members of a group are complete */
		if (slots_per_op != 0 && slot_cnt == 0) {
			struct iop_adma_desc_slot *grp_iter, *_grp_iter;
			int end_of_chain = 0;
			pr_debug("\tgroup end\n");

			/* collect the total results */
			if (grp_start->xor_check_result) {
				u32 zero_sum_result = 0;
				slot_cnt = grp_start->slot_cnt;
				grp_iter = grp_start;

				list_for_each_entry_from(grp_iter,
					&iop_chan->chain, chain_node) {
					zero_sum_result |=
					    iop_desc_get_zero_result(grp_iter);
					    pr_debug("\titer%d result: %d\n",
					    grp_iter->idx, zero_sum_result);
					slot_cnt -= slots_per_op;
					if (slot_cnt == 0)
						break;
				}
				pr_debug("\tgrp_start->xor_check_result: %p\n",
					grp_start->xor_check_result);
				*grp_start->xor_check_result = zero_sum_result;
			}

			/* clean up the group */
			slot_cnt = grp_start->slot_cnt;
			grp_iter = grp_start;
			list_for_each_entry_safe_from(grp_iter, _grp_iter,
				&iop_chan->chain, chain_node) {
				cookie = iop_adma_run_tx_complete_actions(
					grp_iter, iop_chan, cookie);

				slot_cnt -= slots_per_op;
				end_of_chain = iop_adma_clean_slot(grp_iter,
					iop_chan);

				if (slot_cnt == 0 || end_of_chain)
					break;
			}

			/* the group should be complete at this point */
			BUG_ON(slot_cnt);

			slots_per_op = 0;
			grp_start = NULL;
			if (end_of_chain)
				break;
			else
				continue;
		} else if (slots_per_op) /* wait for group completion */
			continue;

		/* write back zero sum results (single descriptor case) */
		if (iter->xor_check_result && iter->async_tx.cookie)
			*iter->xor_check_result =
				iop_desc_get_zero_result(iter);

		cookie = iop_adma_run_tx_complete_actions(
					iter, iop_chan, cookie);

		if (iop_adma_clean_slot(iter, iop_chan))
			break;
	}

	BUG_ON(!seen_current);

	iop_chan_idle(busy, iop_chan);

	if (cookie > 0) {
		iop_chan->completed_cookie = cookie;
		pr_debug("\tcompleted cookie %d\n", cookie);
	}
}

static void
iop_adma_slot_cleanup(struct iop_adma_chan *iop_chan)
{
	spin_lock_bh(&iop_chan->lock);
	__iop_adma_slot_cleanup(iop_chan);
	spin_unlock_bh(&iop_chan->lock);
}

static void iop_adma_tasklet(unsigned long data)
{
	struct iop_adma_chan *chan = (struct iop_adma_chan *) data;
	__iop_adma_slot_cleanup(chan);
}

static struct iop_adma_desc_slot *
iop_adma_alloc_slots(struct iop_adma_chan *iop_chan, int num_slots,
			int slots_per_op)
{
	struct iop_adma_desc_slot *iter, *_iter, *alloc_start = NULL;
287
	LIST_HEAD(chain);
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
	int slots_found, retry = 0;

	/* start search from the last allocated descrtiptor
	 * if a contiguous allocation can not be found start searching
	 * from the beginning of the list
	 */
retry:
	slots_found = 0;
	if (retry == 0)
		iter = iop_chan->last_used;
	else
		iter = list_entry(&iop_chan->all_slots,
			struct iop_adma_desc_slot,
			slot_node);

	list_for_each_entry_safe_continue(
		iter, _iter, &iop_chan->all_slots, slot_node) {
		prefetch(_iter);
		prefetch(&_iter->async_tx);
		if (iter->slots_per_op) {
			/* give up after finding the first busy slot
			 * on the second pass through the list
			 */
			if (retry)
				break;

			slots_found = 0;
			continue;
		}

		/* start the allocation if the slot is correctly aligned */
		if (!slots_found++) {
			if (iop_desc_is_aligned(iter, slots_per_op))
				alloc_start = iter;
			else {
				slots_found = 0;
				continue;
			}
		}

		if (slots_found == num_slots) {
			struct iop_adma_desc_slot *alloc_tail = NULL;
			struct iop_adma_desc_slot *last_used = NULL;
			iter = alloc_start;
			while (num_slots) {
				int i;
				dev_dbg(iop_chan->device->common.dev,
					"allocated slot: %d "
					"(desc %p phys: %#x) slots_per_op %d\n",
					iter->idx, iter->hw_desc,
					iter->async_tx.phys, slots_per_op);

				/* pre-ack all but the last descriptor */
				if (num_slots != slots_per_op)
					iter->async_tx.ack = 1;
				else
					iter->async_tx.ack = 0;

				list_add_tail(&iter->chain_node, &chain);
				alloc_tail = iter;
				iter->async_tx.cookie = 0;
				iter->slot_cnt = num_slots;
				iter->xor_check_result = NULL;
				for (i = 0; i < slots_per_op; i++) {
					iter->slots_per_op = slots_per_op - i;
					last_used = iter;
					iter = list_entry(iter->slot_node.next,
						struct iop_adma_desc_slot,
						slot_node);
				}
				num_slots -= slots_per_op;
			}
			alloc_tail->group_head = alloc_start;
			alloc_tail->async_tx.cookie = -EBUSY;
			list_splice(&chain, &alloc_tail->async_tx.tx_list);
			iop_chan->last_used = last_used;
			iop_desc_clear_next_desc(alloc_start);
			iop_desc_clear_next_desc(alloc_tail);
			return alloc_tail;
		}
	}
	if (!retry++)
		goto retry;

	/* try to free some slots if the allocation fails */
	tasklet_schedule(&iop_chan->irq_tasklet);

	return NULL;
}

static dma_cookie_t
iop_desc_assign_cookie(struct iop_adma_chan *iop_chan,
	struct iop_adma_desc_slot *desc)
{
	dma_cookie_t cookie = iop_chan->common.cookie;
	cookie++;
	if (cookie < 0)
		cookie = 1;
	iop_chan->common.cookie = desc->async_tx.cookie = cookie;
	return cookie;
}

static void iop_adma_check_threshold(struct iop_adma_chan *iop_chan)
{
	dev_dbg(iop_chan->device->common.dev, "pending: %d\n",
		iop_chan->pending);

	if (iop_chan->pending >= IOP_ADMA_THRESHOLD) {
		iop_chan->pending = 0;
		iop_chan_append(iop_chan);
	}
}

static dma_cookie_t
iop_adma_tx_submit(struct dma_async_tx_descriptor *tx)
{
	struct iop_adma_desc_slot *sw_desc = tx_to_iop_adma_slot(tx);
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(tx->chan);
	struct iop_adma_desc_slot *grp_start, *old_chain_tail;
	int slot_cnt;
	int slots_per_op;
	dma_cookie_t cookie;

	grp_start = sw_desc->group_head;
	slot_cnt = grp_start->slot_cnt;
	slots_per_op = grp_start->slots_per_op;

	spin_lock_bh(&iop_chan->lock);
	cookie = iop_desc_assign_cookie(iop_chan, sw_desc);

	old_chain_tail = list_entry(iop_chan->chain.prev,
		struct iop_adma_desc_slot, chain_node);
	list_splice_init(&sw_desc->async_tx.tx_list,
			 &old_chain_tail->chain_node);

	/* fix up the hardware chain */
	iop_desc_set_next_desc(old_chain_tail, grp_start->async_tx.phys);

	/* 1/ don't add pre-chained descriptors
	 * 2/ dummy read to flush next_desc write
	 */
	BUG_ON(iop_desc_get_next_desc(sw_desc));

	/* increment the pending count by the number of slots
	 * memcpy operations have a 1:1 (slot:operation) relation
	 * other operations are heavier and will pop the threshold
	 * more often.
	 */
	iop_chan->pending += slot_cnt;
	iop_adma_check_threshold(iop_chan);
	spin_unlock_bh(&iop_chan->lock);

	dev_dbg(iop_chan->device->common.dev, "%s cookie: %d slot: %d\n",
441
		__func__, sw_desc->async_tx.cookie, sw_desc->idx);
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

	return cookie;
}

static void iop_chan_start_null_memcpy(struct iop_adma_chan *iop_chan);
static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan);

/* returns the number of allocated descriptors */
static int iop_adma_alloc_chan_resources(struct dma_chan *chan)
{
	char *hw_desc;
	int idx;
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *slot = NULL;
	int init = iop_chan->slots_allocated ? 0 : 1;
	struct iop_adma_platform_data *plat_data =
		iop_chan->device->pdev->dev.platform_data;
	int num_descs_in_pool = plat_data->pool_size/IOP_ADMA_SLOT_SIZE;

	/* Allocate descriptor slots */
	do {
		idx = iop_chan->slots_allocated;
		if (idx == num_descs_in_pool)
			break;

		slot = kzalloc(sizeof(*slot), GFP_KERNEL);
		if (!slot) {
			printk(KERN_INFO "IOP ADMA Channel only initialized"
				" %d descriptor slots", idx);
			break;
		}
		hw_desc = (char *) iop_chan->device->dma_desc_pool_virt;
		slot->hw_desc = (void *) &hw_desc[idx * IOP_ADMA_SLOT_SIZE];

		dma_async_tx_descriptor_init(&slot->async_tx, chan);
		slot->async_tx.tx_submit = iop_adma_tx_submit;
		INIT_LIST_HEAD(&slot->chain_node);
		INIT_LIST_HEAD(&slot->slot_node);
		INIT_LIST_HEAD(&slot->async_tx.tx_list);
		hw_desc = (char *) iop_chan->device->dma_desc_pool;
		slot->async_tx.phys =
			(dma_addr_t) &hw_desc[idx * IOP_ADMA_SLOT_SIZE];
		slot->idx = idx;

		spin_lock_bh(&iop_chan->lock);
		iop_chan->slots_allocated++;
		list_add_tail(&slot->slot_node, &iop_chan->all_slots);
		spin_unlock_bh(&iop_chan->lock);
	} while (iop_chan->slots_allocated < num_descs_in_pool);

	if (idx && !iop_chan->last_used)
		iop_chan->last_used = list_entry(iop_chan->all_slots.next,
					struct iop_adma_desc_slot,
					slot_node);

	dev_dbg(iop_chan->device->common.dev,
		"allocated %d descriptor slots last_used: %p\n",
		iop_chan->slots_allocated, iop_chan->last_used);

	/* initialize the channel and the chain with a null operation */
	if (init) {
		if (dma_has_cap(DMA_MEMCPY,
			iop_chan->device->common.cap_mask))
			iop_chan_start_null_memcpy(iop_chan);
		else if (dma_has_cap(DMA_XOR,
			iop_chan->device->common.cap_mask))
			iop_chan_start_null_xor(iop_chan);
		else
			BUG();
	}

	return (idx > 0) ? idx : -ENOMEM;
}

static struct dma_async_tx_descriptor *
iop_adma_prep_dma_interrupt(struct dma_chan *chan)
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	int slot_cnt, slots_per_op;

523
	dev_dbg(iop_chan->device->common.dev, "%s\n", __func__);
524 525 526 527 528 529 530 531 532 533 534 535 536 537 538

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_interrupt_slot_count(&slots_per_op, iop_chan);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
		iop_desc_init_interrupt(grp_start, iop_chan);
		grp_start->unmap_len = 0;
	}
	spin_unlock_bh(&iop_chan->lock);

	return sw_desc ? &sw_desc->async_tx : NULL;
}

static struct dma_async_tx_descriptor *
539
iop_adma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dma_dest,
540
			 dma_addr_t dma_src, size_t len, unsigned long flags)
541 542 543 544 545 546 547 548 549 550
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	int slot_cnt, slots_per_op;

	if (unlikely(!len))
		return NULL;
	BUG_ON(unlikely(len > IOP_ADMA_MAX_BYTE_COUNT));

	dev_dbg(iop_chan->device->common.dev, "%s len: %u\n",
551
		__func__, len);
552 553 554 555 556 557

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_memcpy_slot_count(len, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
558
		iop_desc_init_memcpy(grp_start, flags);
559
		iop_desc_set_byte_count(grp_start, iop_chan, len);
560 561
		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
		iop_desc_set_memcpy_src_addr(grp_start, dma_src);
562 563 564 565 566 567 568 569 570
		sw_desc->unmap_src_cnt = 1;
		sw_desc->unmap_len = len;
	}
	spin_unlock_bh(&iop_chan->lock);

	return sw_desc ? &sw_desc->async_tx : NULL;
}

static struct dma_async_tx_descriptor *
571
iop_adma_prep_dma_memset(struct dma_chan *chan, dma_addr_t dma_dest,
572
			 int value, size_t len, unsigned long flags)
573 574 575 576 577 578 579 580 581 582
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	int slot_cnt, slots_per_op;

	if (unlikely(!len))
		return NULL;
	BUG_ON(unlikely(len > IOP_ADMA_MAX_BYTE_COUNT));

	dev_dbg(iop_chan->device->common.dev, "%s len: %u\n",
583
		__func__, len);
584 585 586 587 588 589

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_memset_slot_count(len, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
590
		iop_desc_init_memset(grp_start, flags);
591 592
		iop_desc_set_byte_count(grp_start, iop_chan, len);
		iop_desc_set_block_fill_val(grp_start, value);
593
		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
594 595 596 597 598 599 600 601 602
		sw_desc->unmap_src_cnt = 1;
		sw_desc->unmap_len = len;
	}
	spin_unlock_bh(&iop_chan->lock);

	return sw_desc ? &sw_desc->async_tx : NULL;
}

static struct dma_async_tx_descriptor *
603 604
iop_adma_prep_dma_xor(struct dma_chan *chan, dma_addr_t dma_dest,
		      dma_addr_t *dma_src, unsigned int src_cnt, size_t len,
605
		      unsigned long flags)
606 607 608 609 610 611 612 613 614 615
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	int slot_cnt, slots_per_op;

	if (unlikely(!len))
		return NULL;
	BUG_ON(unlikely(len > IOP_ADMA_XOR_MAX_BYTE_COUNT));

	dev_dbg(iop_chan->device->common.dev,
616
		"%s src_cnt: %d len: %u flags: %lx\n",
617
		__func__, src_cnt, len, flags);
618 619 620 621 622 623

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_xor_slot_count(len, src_cnt, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
624
		iop_desc_init_xor(grp_start, src_cnt, flags);
625
		iop_desc_set_byte_count(grp_start, iop_chan, len);
626
		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
627 628
		sw_desc->unmap_src_cnt = src_cnt;
		sw_desc->unmap_len = len;
629 630 631
		while (src_cnt--)
			iop_desc_set_xor_src_addr(grp_start, src_cnt,
						  dma_src[src_cnt]);
632 633 634 635 636 637 638
	}
	spin_unlock_bh(&iop_chan->lock);

	return sw_desc ? &sw_desc->async_tx : NULL;
}

static struct dma_async_tx_descriptor *
639 640
iop_adma_prep_dma_zero_sum(struct dma_chan *chan, dma_addr_t *dma_src,
			   unsigned int src_cnt, size_t len, u32 *result,
641
			   unsigned long flags)
642 643 644 645 646 647 648 649 650
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	int slot_cnt, slots_per_op;

	if (unlikely(!len))
		return NULL;

	dev_dbg(iop_chan->device->common.dev, "%s src_cnt: %d len: %u\n",
651
		__func__, src_cnt, len);
652 653 654 655 656 657

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_zero_sum_slot_count(len, src_cnt, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
658
		iop_desc_init_zero_sum(grp_start, src_cnt, flags);
659 660 661
		iop_desc_set_zero_sum_byte_count(grp_start, len);
		grp_start->xor_check_result = result;
		pr_debug("\t%s: grp_start->xor_check_result: %p\n",
662
			__func__, grp_start->xor_check_result);
663 664
		sw_desc->unmap_src_cnt = src_cnt;
		sw_desc->unmap_len = len;
665 666 667
		while (src_cnt--)
			iop_desc_set_zero_sum_src_addr(grp_start, src_cnt,
						       dma_src[src_cnt]);
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
	}
	spin_unlock_bh(&iop_chan->lock);

	return sw_desc ? &sw_desc->async_tx : NULL;
}

static void iop_adma_dependency_added(struct dma_chan *chan)
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	tasklet_schedule(&iop_chan->irq_tasklet);
}

static void iop_adma_free_chan_resources(struct dma_chan *chan)
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	struct iop_adma_desc_slot *iter, *_iter;
	int in_use_descs = 0;

	iop_adma_slot_cleanup(iop_chan);

	spin_lock_bh(&iop_chan->lock);
	list_for_each_entry_safe(iter, _iter, &iop_chan->chain,
					chain_node) {
		in_use_descs++;
		list_del(&iter->chain_node);
	}
	list_for_each_entry_safe_reverse(
		iter, _iter, &iop_chan->all_slots, slot_node) {
		list_del(&iter->slot_node);
		kfree(iter);
		iop_chan->slots_allocated--;
	}
	iop_chan->last_used = NULL;

	dev_dbg(iop_chan->device->common.dev, "%s slots_allocated %d\n",
703
		__func__, iop_chan->slots_allocated);
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
	spin_unlock_bh(&iop_chan->lock);

	/* one is ok since we left it on there on purpose */
	if (in_use_descs > 1)
		printk(KERN_ERR "IOP: Freeing %d in use descriptors!\n",
			in_use_descs - 1);
}

/**
 * iop_adma_is_complete - poll the status of an ADMA transaction
 * @chan: ADMA channel handle
 * @cookie: ADMA transaction identifier
 */
static enum dma_status iop_adma_is_complete(struct dma_chan *chan,
					dma_cookie_t cookie,
					dma_cookie_t *done,
					dma_cookie_t *used)
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);
	dma_cookie_t last_used;
	dma_cookie_t last_complete;
	enum dma_status ret;

	last_used = chan->cookie;
	last_complete = iop_chan->completed_cookie;

	if (done)
		*done = last_complete;
	if (used)
		*used = last_used;

	ret = dma_async_is_complete(cookie, last_complete, last_used);
	if (ret == DMA_SUCCESS)
		return ret;

	iop_adma_slot_cleanup(iop_chan);

	last_used = chan->cookie;
	last_complete = iop_chan->completed_cookie;

	if (done)
		*done = last_complete;
	if (used)
		*used = last_used;

	return dma_async_is_complete(cookie, last_complete, last_used);
}

static irqreturn_t iop_adma_eot_handler(int irq, void *data)
{
	struct iop_adma_chan *chan = data;

756
	dev_dbg(chan->device->common.dev, "%s\n", __func__);
757 758 759 760 761 762 763 764 765 766 767 768

	tasklet_schedule(&chan->irq_tasklet);

	iop_adma_device_clear_eot_status(chan);

	return IRQ_HANDLED;
}

static irqreturn_t iop_adma_eoc_handler(int irq, void *data)
{
	struct iop_adma_chan *chan = data;

769
	dev_dbg(chan->device->common.dev, "%s\n", __func__);
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

	tasklet_schedule(&chan->irq_tasklet);

	iop_adma_device_clear_eoc_status(chan);

	return IRQ_HANDLED;
}

static irqreturn_t iop_adma_err_handler(int irq, void *data)
{
	struct iop_adma_chan *chan = data;
	unsigned long status = iop_chan_get_status(chan);

	dev_printk(KERN_ERR, chan->device->common.dev,
		"error ( %s%s%s%s%s%s%s)\n",
		iop_is_err_int_parity(status, chan) ? "int_parity " : "",
		iop_is_err_mcu_abort(status, chan) ? "mcu_abort " : "",
		iop_is_err_int_tabort(status, chan) ? "int_tabort " : "",
		iop_is_err_int_mabort(status, chan) ? "int_mabort " : "",
		iop_is_err_pci_tabort(status, chan) ? "pci_tabort " : "",
		iop_is_err_pci_mabort(status, chan) ? "pci_mabort " : "",
		iop_is_err_split_tx(status, chan) ? "split_tx " : "");

	iop_adma_device_clear_err_status(chan);

	BUG();

	return IRQ_HANDLED;
}

static void iop_adma_issue_pending(struct dma_chan *chan)
{
	struct iop_adma_chan *iop_chan = to_iop_adma_chan(chan);

	if (iop_chan->pending) {
		iop_chan->pending = 0;
		iop_chan_append(iop_chan);
	}
}

/*
 * Perform a transaction to verify the HW works.
 */
#define IOP_ADMA_TEST_SIZE 2000

static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device)
{
	int i;
	void *src, *dest;
	dma_addr_t src_dma, dest_dma;
	struct dma_chan *dma_chan;
	dma_cookie_t cookie;
	struct dma_async_tx_descriptor *tx;
	int err = 0;
	struct iop_adma_chan *iop_chan;

826
	dev_dbg(device->common.dev, "%s\n", __func__);
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

	src = kzalloc(sizeof(u8) * IOP_ADMA_TEST_SIZE, GFP_KERNEL);
	if (!src)
		return -ENOMEM;
	dest = kzalloc(sizeof(u8) * IOP_ADMA_TEST_SIZE, GFP_KERNEL);
	if (!dest) {
		kfree(src);
		return -ENOMEM;
	}

	/* Fill in src buffer */
	for (i = 0; i < IOP_ADMA_TEST_SIZE; i++)
		((u8 *) src)[i] = (u8)i;

	memset(dest, 0, IOP_ADMA_TEST_SIZE);

	/* Start copy, using first DMA channel */
	dma_chan = container_of(device->common.channels.next,
				struct dma_chan,
				device_node);
	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
		err = -ENODEV;
		goto out;
	}

	dest_dma = dma_map_single(dma_chan->device->dev, dest,
				IOP_ADMA_TEST_SIZE, DMA_FROM_DEVICE);
	src_dma = dma_map_single(dma_chan->device->dev, src,
				IOP_ADMA_TEST_SIZE, DMA_TO_DEVICE);
856 857
	tx = iop_adma_prep_dma_memcpy(dma_chan, dest_dma, src_dma,
				      IOP_ADMA_TEST_SIZE, 1);
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

	cookie = iop_adma_tx_submit(tx);
	iop_adma_issue_pending(dma_chan);
	async_tx_ack(tx);
	msleep(1);

	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) !=
			DMA_SUCCESS) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test copy timed out, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	iop_chan = to_iop_adma_chan(dma_chan);
	dma_sync_single_for_cpu(&iop_chan->device->pdev->dev, dest_dma,
		IOP_ADMA_TEST_SIZE, DMA_FROM_DEVICE);
	if (memcmp(src, dest, IOP_ADMA_TEST_SIZE)) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test copy failed compare, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

free_resources:
	iop_adma_free_chan_resources(dma_chan);
out:
	kfree(src);
	kfree(dest);
	return err;
}

#define IOP_ADMA_NUM_SRC_TEST 4 /* must be <= 15 */
static int __devinit
iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
{
	int i, src_idx;
	struct page *dest;
	struct page *xor_srcs[IOP_ADMA_NUM_SRC_TEST];
	struct page *zero_sum_srcs[IOP_ADMA_NUM_SRC_TEST + 1];
898
	dma_addr_t dma_srcs[IOP_ADMA_NUM_SRC_TEST + 1];
899 900 901 902 903 904 905 906 907 908
	dma_addr_t dma_addr, dest_dma;
	struct dma_async_tx_descriptor *tx;
	struct dma_chan *dma_chan;
	dma_cookie_t cookie;
	u8 cmp_byte = 0;
	u32 cmp_word;
	u32 zero_sum_result;
	int err = 0;
	struct iop_adma_chan *iop_chan;

909
	dev_dbg(device->common.dev, "%s\n", __func__);
910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952

	for (src_idx = 0; src_idx < IOP_ADMA_NUM_SRC_TEST; src_idx++) {
		xor_srcs[src_idx] = alloc_page(GFP_KERNEL);
		if (!xor_srcs[src_idx])
			while (src_idx--) {
				__free_page(xor_srcs[src_idx]);
				return -ENOMEM;
			}
	}

	dest = alloc_page(GFP_KERNEL);
	if (!dest)
		while (src_idx--) {
			__free_page(xor_srcs[src_idx]);
			return -ENOMEM;
		}

	/* Fill in src buffers */
	for (src_idx = 0; src_idx < IOP_ADMA_NUM_SRC_TEST; src_idx++) {
		u8 *ptr = page_address(xor_srcs[src_idx]);
		for (i = 0; i < PAGE_SIZE; i++)
			ptr[i] = (1 << src_idx);
	}

	for (src_idx = 0; src_idx < IOP_ADMA_NUM_SRC_TEST; src_idx++)
		cmp_byte ^= (u8) (1 << src_idx);

	cmp_word = (cmp_byte << 24) | (cmp_byte << 16) |
			(cmp_byte << 8) | cmp_byte;

	memset(page_address(dest), 0, PAGE_SIZE);

	dma_chan = container_of(device->common.channels.next,
				struct dma_chan,
				device_node);
	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
		err = -ENODEV;
		goto out;
	}

	/* test xor */
	dest_dma = dma_map_page(dma_chan->device->dev, dest, 0,
				PAGE_SIZE, DMA_FROM_DEVICE);
953 954 955 956 957
	for (i = 0; i < IOP_ADMA_NUM_SRC_TEST; i++)
		dma_srcs[i] = dma_map_page(dma_chan->device->dev, xor_srcs[i],
					   0, PAGE_SIZE, DMA_TO_DEVICE);
	tx = iop_adma_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
				   IOP_ADMA_NUM_SRC_TEST, PAGE_SIZE, 1);
958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997

	cookie = iop_adma_tx_submit(tx);
	iop_adma_issue_pending(dma_chan);
	async_tx_ack(tx);
	msleep(8);

	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) !=
		DMA_SUCCESS) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test xor timed out, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	iop_chan = to_iop_adma_chan(dma_chan);
	dma_sync_single_for_cpu(&iop_chan->device->pdev->dev, dest_dma,
		PAGE_SIZE, DMA_FROM_DEVICE);
	for (i = 0; i < (PAGE_SIZE / sizeof(u32)); i++) {
		u32 *ptr = page_address(dest);
		if (ptr[i] != cmp_word) {
			dev_printk(KERN_ERR, dma_chan->device->dev,
				"Self-test xor failed compare, disabling\n");
			err = -ENODEV;
			goto free_resources;
		}
	}
	dma_sync_single_for_device(&iop_chan->device->pdev->dev, dest_dma,
		PAGE_SIZE, DMA_TO_DEVICE);

	/* skip zero sum if the capability is not present */
	if (!dma_has_cap(DMA_ZERO_SUM, dma_chan->device->cap_mask))
		goto free_resources;

	/* zero sum the sources with the destintation page */
	for (i = 0; i < IOP_ADMA_NUM_SRC_TEST; i++)
		zero_sum_srcs[i] = xor_srcs[i];
	zero_sum_srcs[i] = dest;

	zero_sum_result = 1;

998 999 1000 1001 1002 1003 1004
	for (i = 0; i < IOP_ADMA_NUM_SRC_TEST + 1; i++)
		dma_srcs[i] = dma_map_page(dma_chan->device->dev,
					   zero_sum_srcs[i], 0, PAGE_SIZE,
					   DMA_TO_DEVICE);
	tx = iop_adma_prep_dma_zero_sum(dma_chan, dma_srcs,
					IOP_ADMA_NUM_SRC_TEST + 1, PAGE_SIZE,
					&zero_sum_result, 1);
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027

	cookie = iop_adma_tx_submit(tx);
	iop_adma_issue_pending(dma_chan);
	async_tx_ack(tx);
	msleep(8);

	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test zero sum timed out, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	if (zero_sum_result != 0) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test zero sum failed compare, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	/* test memset */
	dma_addr = dma_map_page(dma_chan->device->dev, dest, 0,
			PAGE_SIZE, DMA_FROM_DEVICE);
1028
	tx = iop_adma_prep_dma_memset(dma_chan, dma_addr, 0, PAGE_SIZE, 1);
1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053

	cookie = iop_adma_tx_submit(tx);
	iop_adma_issue_pending(dma_chan);
	async_tx_ack(tx);
	msleep(8);

	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test memset timed out, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	for (i = 0; i < PAGE_SIZE/sizeof(u32); i++) {
		u32 *ptr = page_address(dest);
		if (ptr[i]) {
			dev_printk(KERN_ERR, dma_chan->device->dev,
				"Self-test memset failed compare, disabling\n");
			err = -ENODEV;
			goto free_resources;
		}
	}

	/* test for non-zero parity sum */
	zero_sum_result = 0;
1054 1055 1056 1057 1058 1059 1060
	for (i = 0; i < IOP_ADMA_NUM_SRC_TEST + 1; i++)
		dma_srcs[i] = dma_map_page(dma_chan->device->dev,
					   zero_sum_srcs[i], 0, PAGE_SIZE,
					   DMA_TO_DEVICE);
	tx = iop_adma_prep_dma_zero_sum(dma_chan, dma_srcs,
					IOP_ADMA_NUM_SRC_TEST + 1, PAGE_SIZE,
					&zero_sum_result, 1);
1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161

	cookie = iop_adma_tx_submit(tx);
	iop_adma_issue_pending(dma_chan);
	async_tx_ack(tx);
	msleep(8);

	if (iop_adma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test non-zero sum timed out, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

	if (zero_sum_result != 1) {
		dev_printk(KERN_ERR, dma_chan->device->dev,
			"Self-test non-zero sum failed compare, disabling\n");
		err = -ENODEV;
		goto free_resources;
	}

free_resources:
	iop_adma_free_chan_resources(dma_chan);
out:
	src_idx = IOP_ADMA_NUM_SRC_TEST;
	while (src_idx--)
		__free_page(xor_srcs[src_idx]);
	__free_page(dest);
	return err;
}

static int __devexit iop_adma_remove(struct platform_device *dev)
{
	struct iop_adma_device *device = platform_get_drvdata(dev);
	struct dma_chan *chan, *_chan;
	struct iop_adma_chan *iop_chan;
	int i;
	struct iop_adma_platform_data *plat_data = dev->dev.platform_data;

	dma_async_device_unregister(&device->common);

	for (i = 0; i < 3; i++) {
		unsigned int irq;
		irq = platform_get_irq(dev, i);
		free_irq(irq, device);
	}

	dma_free_coherent(&dev->dev, plat_data->pool_size,
			device->dma_desc_pool_virt, device->dma_desc_pool);

	do {
		struct resource *res;
		res = platform_get_resource(dev, IORESOURCE_MEM, 0);
		release_mem_region(res->start, res->end - res->start);
	} while (0);

	list_for_each_entry_safe(chan, _chan, &device->common.channels,
				device_node) {
		iop_chan = to_iop_adma_chan(chan);
		list_del(&chan->device_node);
		kfree(iop_chan);
	}
	kfree(device);

	return 0;
}

static int __devinit iop_adma_probe(struct platform_device *pdev)
{
	struct resource *res;
	int ret = 0, i;
	struct iop_adma_device *adev;
	struct iop_adma_chan *iop_chan;
	struct dma_device *dma_dev;
	struct iop_adma_platform_data *plat_data = pdev->dev.platform_data;

	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
	if (!res)
		return -ENODEV;

	if (!devm_request_mem_region(&pdev->dev, res->start,
				res->end - res->start, pdev->name))
		return -EBUSY;

	adev = kzalloc(sizeof(*adev), GFP_KERNEL);
	if (!adev)
		return -ENOMEM;
	dma_dev = &adev->common;

	/* allocate coherent memory for hardware descriptors
	 * note: writecombine gives slightly better performance, but
	 * requires that we explicitly flush the writes
	 */
	if ((adev->dma_desc_pool_virt = dma_alloc_writecombine(&pdev->dev,
					plat_data->pool_size,
					&adev->dma_desc_pool,
					GFP_KERNEL)) == NULL) {
		ret = -ENOMEM;
		goto err_free_adev;
	}

	dev_dbg(&pdev->dev, "%s: allocted descriptor pool virt %p phys %p\n",
1162
		__func__, adev->dma_desc_pool_virt,
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291
		(void *) adev->dma_desc_pool);

	adev->id = plat_data->hw_id;

	/* discover transaction capabilites from the platform data */
	dma_dev->cap_mask = plat_data->cap_mask;

	adev->pdev = pdev;
	platform_set_drvdata(pdev, adev);

	INIT_LIST_HEAD(&dma_dev->channels);

	/* set base routines */
	dma_dev->device_alloc_chan_resources = iop_adma_alloc_chan_resources;
	dma_dev->device_free_chan_resources = iop_adma_free_chan_resources;
	dma_dev->device_is_tx_complete = iop_adma_is_complete;
	dma_dev->device_issue_pending = iop_adma_issue_pending;
	dma_dev->device_dependency_added = iop_adma_dependency_added;
	dma_dev->dev = &pdev->dev;

	/* set prep routines based on capability */
	if (dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask))
		dma_dev->device_prep_dma_memcpy = iop_adma_prep_dma_memcpy;
	if (dma_has_cap(DMA_MEMSET, dma_dev->cap_mask))
		dma_dev->device_prep_dma_memset = iop_adma_prep_dma_memset;
	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask)) {
		dma_dev->max_xor = iop_adma_get_max_xor();
		dma_dev->device_prep_dma_xor = iop_adma_prep_dma_xor;
	}
	if (dma_has_cap(DMA_ZERO_SUM, dma_dev->cap_mask))
		dma_dev->device_prep_dma_zero_sum =
			iop_adma_prep_dma_zero_sum;
	if (dma_has_cap(DMA_INTERRUPT, dma_dev->cap_mask))
		dma_dev->device_prep_dma_interrupt =
			iop_adma_prep_dma_interrupt;

	iop_chan = kzalloc(sizeof(*iop_chan), GFP_KERNEL);
	if (!iop_chan) {
		ret = -ENOMEM;
		goto err_free_dma;
	}
	iop_chan->device = adev;

	iop_chan->mmr_base = devm_ioremap(&pdev->dev, res->start,
					res->end - res->start);
	if (!iop_chan->mmr_base) {
		ret = -ENOMEM;
		goto err_free_iop_chan;
	}
	tasklet_init(&iop_chan->irq_tasklet, iop_adma_tasklet, (unsigned long)
		iop_chan);

	/* clear errors before enabling interrupts */
	iop_adma_device_clear_err_status(iop_chan);

	for (i = 0; i < 3; i++) {
		irq_handler_t handler[] = { iop_adma_eot_handler,
					iop_adma_eoc_handler,
					iop_adma_err_handler };
		int irq = platform_get_irq(pdev, i);
		if (irq < 0) {
			ret = -ENXIO;
			goto err_free_iop_chan;
		} else {
			ret = devm_request_irq(&pdev->dev, irq,
					handler[i], 0, pdev->name, iop_chan);
			if (ret)
				goto err_free_iop_chan;
		}
	}

	spin_lock_init(&iop_chan->lock);
	init_timer(&iop_chan->cleanup_watchdog);
	iop_chan->cleanup_watchdog.data = (unsigned long) iop_chan;
	iop_chan->cleanup_watchdog.function = iop_adma_tasklet;
	INIT_LIST_HEAD(&iop_chan->chain);
	INIT_LIST_HEAD(&iop_chan->all_slots);
	INIT_RCU_HEAD(&iop_chan->common.rcu);
	iop_chan->common.device = dma_dev;
	list_add_tail(&iop_chan->common.device_node, &dma_dev->channels);

	if (dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask)) {
		ret = iop_adma_memcpy_self_test(adev);
		dev_dbg(&pdev->dev, "memcpy self test returned %d\n", ret);
		if (ret)
			goto err_free_iop_chan;
	}

	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask) ||
		dma_has_cap(DMA_MEMSET, dma_dev->cap_mask)) {
		ret = iop_adma_xor_zero_sum_self_test(adev);
		dev_dbg(&pdev->dev, "xor self test returned %d\n", ret);
		if (ret)
			goto err_free_iop_chan;
	}

	dev_printk(KERN_INFO, &pdev->dev, "Intel(R) IOP: "
	  "( %s%s%s%s%s%s%s%s%s%s)\n",
	  dma_has_cap(DMA_PQ_XOR, dma_dev->cap_mask) ? "pq_xor " : "",
	  dma_has_cap(DMA_PQ_UPDATE, dma_dev->cap_mask) ? "pq_update " : "",
	  dma_has_cap(DMA_PQ_ZERO_SUM, dma_dev->cap_mask) ? "pq_zero_sum " : "",
	  dma_has_cap(DMA_XOR, dma_dev->cap_mask) ? "xor " : "",
	  dma_has_cap(DMA_DUAL_XOR, dma_dev->cap_mask) ? "dual_xor " : "",
	  dma_has_cap(DMA_ZERO_SUM, dma_dev->cap_mask) ? "xor_zero_sum " : "",
	  dma_has_cap(DMA_MEMSET, dma_dev->cap_mask)  ? "fill " : "",
	  dma_has_cap(DMA_MEMCPY_CRC32C, dma_dev->cap_mask) ? "cpy+crc " : "",
	  dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask) ? "cpy " : "",
	  dma_has_cap(DMA_INTERRUPT, dma_dev->cap_mask) ? "intr " : "");

	dma_async_device_register(dma_dev);
	goto out;

 err_free_iop_chan:
	kfree(iop_chan);
 err_free_dma:
	dma_free_coherent(&adev->pdev->dev, plat_data->pool_size,
			adev->dma_desc_pool_virt, adev->dma_desc_pool);
 err_free_adev:
	kfree(adev);
 out:
	return ret;
}

static void iop_chan_start_null_memcpy(struct iop_adma_chan *iop_chan)
{
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	dma_cookie_t cookie;
	int slot_cnt, slots_per_op;

1292
	dev_dbg(iop_chan->device->common.dev, "%s\n", __func__);
1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_memcpy_slot_count(0, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;

		list_splice_init(&sw_desc->async_tx.tx_list, &iop_chan->chain);
		sw_desc->async_tx.ack = 1;
		iop_desc_init_memcpy(grp_start, 0);
		iop_desc_set_byte_count(grp_start, iop_chan, 0);
		iop_desc_set_dest_addr(grp_start, iop_chan, 0);
		iop_desc_set_memcpy_src_addr(grp_start, 0);

		cookie = iop_chan->common.cookie;
		cookie++;
		if (cookie <= 1)
			cookie = 2;

		/* initialize the completed cookie to be less than
		 * the most recently used cookie
		 */
		iop_chan->completed_cookie = cookie - 1;
		iop_chan->common.cookie = sw_desc->async_tx.cookie = cookie;

		/* channel should not be busy */
		BUG_ON(iop_chan_is_busy(iop_chan));

		/* clear any prior error-status bits */
		iop_adma_device_clear_err_status(iop_chan);

		/* disable operation */
		iop_chan_disable(iop_chan);

		/* set the descriptor address */
		iop_chan_set_next_descriptor(iop_chan, sw_desc->async_tx.phys);

		/* 1/ don't add pre-chained descriptors
		 * 2/ dummy read to flush next_desc write
		 */
		BUG_ON(iop_desc_get_next_desc(sw_desc));

		/* run the descriptor */
		iop_chan_enable(iop_chan);
	} else
		dev_printk(KERN_ERR, iop_chan->device->common.dev,
			 "failed to allocate null descriptor\n");
	spin_unlock_bh(&iop_chan->lock);
}

static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan)
{
	struct iop_adma_desc_slot *sw_desc, *grp_start;
	dma_cookie_t cookie;
	int slot_cnt, slots_per_op;

1349
	dev_dbg(iop_chan->device->common.dev, "%s\n", __func__);
1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413

	spin_lock_bh(&iop_chan->lock);
	slot_cnt = iop_chan_xor_slot_count(0, 2, &slots_per_op);
	sw_desc = iop_adma_alloc_slots(iop_chan, slot_cnt, slots_per_op);
	if (sw_desc) {
		grp_start = sw_desc->group_head;
		list_splice_init(&sw_desc->async_tx.tx_list, &iop_chan->chain);
		sw_desc->async_tx.ack = 1;
		iop_desc_init_null_xor(grp_start, 2, 0);
		iop_desc_set_byte_count(grp_start, iop_chan, 0);
		iop_desc_set_dest_addr(grp_start, iop_chan, 0);
		iop_desc_set_xor_src_addr(grp_start, 0, 0);
		iop_desc_set_xor_src_addr(grp_start, 1, 0);

		cookie = iop_chan->common.cookie;
		cookie++;
		if (cookie <= 1)
			cookie = 2;

		/* initialize the completed cookie to be less than
		 * the most recently used cookie
		 */
		iop_chan->completed_cookie = cookie - 1;
		iop_chan->common.cookie = sw_desc->async_tx.cookie = cookie;

		/* channel should not be busy */
		BUG_ON(iop_chan_is_busy(iop_chan));

		/* clear any prior error-status bits */
		iop_adma_device_clear_err_status(iop_chan);

		/* disable operation */
		iop_chan_disable(iop_chan);

		/* set the descriptor address */
		iop_chan_set_next_descriptor(iop_chan, sw_desc->async_tx.phys);

		/* 1/ don't add pre-chained descriptors
		 * 2/ dummy read to flush next_desc write
		 */
		BUG_ON(iop_desc_get_next_desc(sw_desc));

		/* run the descriptor */
		iop_chan_enable(iop_chan);
	} else
		dev_printk(KERN_ERR, iop_chan->device->common.dev,
			"failed to allocate null descriptor\n");
	spin_unlock_bh(&iop_chan->lock);
}

static struct platform_driver iop_adma_driver = {
	.probe		= iop_adma_probe,
	.remove		= iop_adma_remove,
	.driver		= {
		.owner	= THIS_MODULE,
		.name	= "iop-adma",
	},
};

static int __init iop_adma_init (void)
{
	return platform_driver_register(&iop_adma_driver);
}

R
Rusty Russell 已提交
1414 1415
/* it's currently unsafe to unload this module */
#if 0
1416 1417 1418 1419 1420
static void __exit iop_adma_exit (void)
{
	platform_driver_unregister(&iop_adma_driver);
	return;
}
R
Rusty Russell 已提交
1421 1422
module_exit(iop_adma_exit);
#endif
1423 1424 1425 1426 1427 1428

module_init(iop_adma_init);

MODULE_AUTHOR("Intel Corporation");
MODULE_DESCRIPTION("IOP ADMA Engine Driver");
MODULE_LICENSE("GPL");