cache-l2x0.c 20.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * arch/arm/mm/cache-l2x0.c - L210/L220 cache controller support
 *
 * Copyright (C) 2007 ARM Limited
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
 */
19
#include <linux/err.h>
20
#include <linux/init.h>
21
#include <linux/spinlock.h>
22
#include <linux/io.h>
23 24
#include <linux/of.h>
#include <linux/of_address.h>
25 26 27

#include <asm/cacheflush.h>
#include <asm/hardware/cache-l2x0.h>
28
#include "cache-aurora-l2.h"
29 30 31 32

#define CACHE_LINE_SIZE		32

static void __iomem *l2x0_base;
33
static DEFINE_RAW_SPINLOCK(l2x0_lock);
34 35
static u32 l2x0_way_mask;	/* Bitmask of active ways */
static u32 l2x0_size;
36
static unsigned long sync_reg_offset = L2X0_CACHE_SYNC;
37

38 39 40 41
/* Aurora don't have the cache ID register available, so we have to
 * pass it though the device tree */
static u32  cache_id_part_number_from_dt;

42 43 44
struct l2x0_regs l2x0_saved_regs;

struct l2x0_of_data {
45
	void (*setup)(const struct device_node *, u32 *, u32 *);
46
	void (*save)(void);
47
	struct outer_cache_fns outer_cache;
48 49
};

50 51
static bool of_init = false;

52
static inline void cache_wait_way(void __iomem *reg, unsigned long mask)
53
{
54
	/* wait for cache operation by line or way to complete */
55
	while (readl_relaxed(reg) & mask)
56
		cpu_relax();
57 58
}

59 60 61 62 63 64 65 66 67
#ifdef CONFIG_CACHE_PL310
static inline void cache_wait(void __iomem *reg, unsigned long mask)
{
	/* cache operations by line are atomic on PL310 */
}
#else
#define cache_wait	cache_wait_way
#endif

68 69
static inline void cache_sync(void)
{
70
	void __iomem *base = l2x0_base;
71

72
	writel_relaxed(0, base + sync_reg_offset);
73
	cache_wait(base + L2X0_CACHE_SYNC, 1);
74 75
}

76 77 78 79
static inline void l2x0_clean_line(unsigned long addr)
{
	void __iomem *base = l2x0_base;
	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
80
	writel_relaxed(addr, base + L2X0_CLEAN_LINE_PA);
81 82 83 84 85 86
}

static inline void l2x0_inv_line(unsigned long addr)
{
	void __iomem *base = l2x0_base;
	cache_wait(base + L2X0_INV_LINE_PA, 1);
87
	writel_relaxed(addr, base + L2X0_INV_LINE_PA);
88 89
}

90
#if defined(CONFIG_PL310_ERRATA_588369) || defined(CONFIG_PL310_ERRATA_727915)
91 92 93 94 95
static inline void debug_writel(unsigned long val)
{
	if (outer_cache.set_debug)
		outer_cache.set_debug(val);
}
96

97
static void pl310_set_debug(unsigned long val)
98 99
{
	writel_relaxed(val, l2x0_base + L2X0_DEBUG_CTRL);
100
}
101 102 103 104 105 106
#else
/* Optimised out for non-errata case */
static inline void debug_writel(unsigned long val)
{
}

107
#define pl310_set_debug	NULL
108
#endif
109

110
#ifdef CONFIG_PL310_ERRATA_588369
111 112 113 114 115 116
static inline void l2x0_flush_line(unsigned long addr)
{
	void __iomem *base = l2x0_base;

	/* Clean by PA followed by Invalidate by PA */
	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
117
	writel_relaxed(addr, base + L2X0_CLEAN_LINE_PA);
118
	cache_wait(base + L2X0_INV_LINE_PA, 1);
119
	writel_relaxed(addr, base + L2X0_INV_LINE_PA);
120 121 122
}
#else

123 124 125 126
static inline void l2x0_flush_line(unsigned long addr)
{
	void __iomem *base = l2x0_base;
	cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
127
	writel_relaxed(addr, base + L2X0_CLEAN_INV_LINE_PA);
128
}
129
#endif
130

131 132 133 134
static void l2x0_cache_sync(void)
{
	unsigned long flags;

135
	raw_spin_lock_irqsave(&l2x0_lock, flags);
136
	cache_sync();
137
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
138 139
}

140
static void __l2x0_flush_all(void)
141
{
142
	debug_writel(0x03);
143 144 145
	writel_relaxed(l2x0_way_mask, l2x0_base + L2X0_CLEAN_INV_WAY);
	cache_wait_way(l2x0_base + L2X0_CLEAN_INV_WAY, l2x0_way_mask);
	cache_sync();
146
	debug_writel(0x00);
147 148 149 150 151 152 153
}

static void l2x0_flush_all(void)
{
	unsigned long flags;

	/* clean all ways */
154
	raw_spin_lock_irqsave(&l2x0_lock, flags);
155
	__l2x0_flush_all();
156
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
157 158
}

159 160 161 162 163
static void l2x0_clean_all(void)
{
	unsigned long flags;

	/* clean all ways */
164
	raw_spin_lock_irqsave(&l2x0_lock, flags);
165 166 167
	writel_relaxed(l2x0_way_mask, l2x0_base + L2X0_CLEAN_WAY);
	cache_wait_way(l2x0_base + L2X0_CLEAN_WAY, l2x0_way_mask);
	cache_sync();
168
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
169 170
}

171
static void l2x0_inv_all(void)
172
{
173 174
	unsigned long flags;

175
	/* invalidate all ways */
176
	raw_spin_lock_irqsave(&l2x0_lock, flags);
177
	/* Invalidating when L2 is enabled is a nono */
178
	BUG_ON(readl(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN);
179
	writel_relaxed(l2x0_way_mask, l2x0_base + L2X0_INV_WAY);
180
	cache_wait_way(l2x0_base + L2X0_INV_WAY, l2x0_way_mask);
181
	cache_sync();
182
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
183 184 185 186
}

static void l2x0_inv_range(unsigned long start, unsigned long end)
{
187
	void __iomem *base = l2x0_base;
188
	unsigned long flags;
189

190
	raw_spin_lock_irqsave(&l2x0_lock, flags);
191 192
	if (start & (CACHE_LINE_SIZE - 1)) {
		start &= ~(CACHE_LINE_SIZE - 1);
193
		debug_writel(0x03);
194
		l2x0_flush_line(start);
195
		debug_writel(0x00);
196 197 198 199 200
		start += CACHE_LINE_SIZE;
	}

	if (end & (CACHE_LINE_SIZE - 1)) {
		end &= ~(CACHE_LINE_SIZE - 1);
201
		debug_writel(0x03);
202
		l2x0_flush_line(end);
203
		debug_writel(0x00);
204 205
	}

206 207 208 209
	while (start < end) {
		unsigned long blk_end = start + min(end - start, 4096UL);

		while (start < blk_end) {
210
			l2x0_inv_line(start);
211 212 213 214
			start += CACHE_LINE_SIZE;
		}

		if (blk_end < end) {
215 216
			raw_spin_unlock_irqrestore(&l2x0_lock, flags);
			raw_spin_lock_irqsave(&l2x0_lock, flags);
217 218
		}
	}
219
	cache_wait(base + L2X0_INV_LINE_PA, 1);
220
	cache_sync();
221
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
222 223 224 225
}

static void l2x0_clean_range(unsigned long start, unsigned long end)
{
226
	void __iomem *base = l2x0_base;
227
	unsigned long flags;
228

229 230 231 232 233
	if ((end - start) >= l2x0_size) {
		l2x0_clean_all();
		return;
	}

234
	raw_spin_lock_irqsave(&l2x0_lock, flags);
235
	start &= ~(CACHE_LINE_SIZE - 1);
236 237 238 239
	while (start < end) {
		unsigned long blk_end = start + min(end - start, 4096UL);

		while (start < blk_end) {
240
			l2x0_clean_line(start);
241 242 243 244
			start += CACHE_LINE_SIZE;
		}

		if (blk_end < end) {
245 246
			raw_spin_unlock_irqrestore(&l2x0_lock, flags);
			raw_spin_lock_irqsave(&l2x0_lock, flags);
247 248
		}
	}
249
	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
250
	cache_sync();
251
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
252 253 254 255
}

static void l2x0_flush_range(unsigned long start, unsigned long end)
{
256
	void __iomem *base = l2x0_base;
257
	unsigned long flags;
258

259 260 261 262 263
	if ((end - start) >= l2x0_size) {
		l2x0_flush_all();
		return;
	}

264
	raw_spin_lock_irqsave(&l2x0_lock, flags);
265
	start &= ~(CACHE_LINE_SIZE - 1);
266 267 268
	while (start < end) {
		unsigned long blk_end = start + min(end - start, 4096UL);

269
		debug_writel(0x03);
270
		while (start < blk_end) {
271
			l2x0_flush_line(start);
272 273
			start += CACHE_LINE_SIZE;
		}
274
		debug_writel(0x00);
275 276

		if (blk_end < end) {
277 278
			raw_spin_unlock_irqrestore(&l2x0_lock, flags);
			raw_spin_lock_irqsave(&l2x0_lock, flags);
279 280
		}
	}
281
	cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
282
	cache_sync();
283
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
284 285
}

286 287 288 289
static void l2x0_disable(void)
{
	unsigned long flags;

290
	raw_spin_lock_irqsave(&l2x0_lock, flags);
291 292 293
	__l2x0_flush_all();
	writel_relaxed(0, l2x0_base + L2X0_CTRL);
	dsb();
294
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);
295 296
}

297
static void l2x0_unlock(u32 cache_id)
298 299 300 301
{
	int lockregs;
	int i;

302 303
	switch (cache_id) {
	case L2X0_CACHE_ID_PART_L310:
304
		lockregs = 8;
305 306 307 308 309
		break;
	case AURORA_CACHE_ID:
		lockregs = 4;
		break;
	default:
310 311
		/* L210 and unknown types */
		lockregs = 1;
312 313
		break;
	}
314 315 316 317 318 319 320 321 322

	for (i = 0; i < lockregs; i++) {
		writel_relaxed(0x0, l2x0_base + L2X0_LOCKDOWN_WAY_D_BASE +
			       i * L2X0_LOCKDOWN_STRIDE);
		writel_relaxed(0x0, l2x0_base + L2X0_LOCKDOWN_WAY_I_BASE +
			       i * L2X0_LOCKDOWN_STRIDE);
	}
}

323
void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask)
324
{
325 326 327
	u32 aux;
	u32 cache_id;
	u32 way_size = 0;
328
	int ways;
329
	int way_size_shift = L2X0_WAY_SIZE_SHIFT;
330
	const char *type;
331 332

	l2x0_base = base;
333 334 335 336 337
	if (cache_id_part_number_from_dt)
		cache_id = cache_id_part_number_from_dt;
	else
		cache_id = readl_relaxed(l2x0_base + L2X0_CACHE_ID)
			& L2X0_CACHE_ID_PART_MASK;
338
	aux = readl_relaxed(l2x0_base + L2X0_AUX_CTRL);
339

340 341 342
	aux &= aux_mask;
	aux |= aux_val;

343
	/* Determine the number of ways */
344
	switch (cache_id) {
345 346 347 348 349 350
	case L2X0_CACHE_ID_PART_L310:
		if (aux & (1 << 16))
			ways = 16;
		else
			ways = 8;
		type = "L310";
351 352 353 354
#ifdef CONFIG_PL310_ERRATA_753970
		/* Unmapped register. */
		sync_reg_offset = L2X0_DUMMY_REG;
#endif
355 356
		if ((cache_id & L2X0_CACHE_ID_RTL_MASK) <= L2X0_CACHE_ID_RTL_R3P0)
			outer_cache.set_debug = pl310_set_debug;
357 358 359 360 361
		break;
	case L2X0_CACHE_ID_PART_L210:
		ways = (aux >> 13) & 0xf;
		type = "L210";
		break;
362 363 364 365 366 367 368 369

	case AURORA_CACHE_ID:
		sync_reg_offset = AURORA_SYNC_REG;
		ways = (aux >> 13) & 0xf;
		ways = 2 << ((ways + 1) >> 2);
		way_size_shift = AURORA_WAY_SIZE_SHIFT;
		type = "Aurora";
		break;
370 371 372 373 374 375 376 377 378
	default:
		/* Assume unknown chips have 8 ways */
		ways = 8;
		type = "L2x0 series";
		break;
	}

	l2x0_way_mask = (1 << ways) - 1;

379 380 381 382
	/*
	 * L2 cache Size =  Way size * Number of ways
	 */
	way_size = (aux & L2X0_AUX_CTRL_WAY_SIZE_MASK) >> 17;
383 384
	way_size = 1 << (way_size + way_size_shift);

385 386
	l2x0_size = ways * way_size * SZ_1K;

387 388 389 390 391
	/*
	 * Check if l2x0 controller is already enabled.
	 * If you are booting from non-secure mode
	 * accessing the below registers will fault.
	 */
392
	if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) {
393 394
		/* Make sure that I&D is not locked down when starting */
		l2x0_unlock(cache_id);
395

396
		/* l2x0 controller is disabled */
397
		writel_relaxed(aux, l2x0_base + L2X0_AUX_CTRL);
398

399 400 401
		l2x0_inv_all();

		/* enable L2X0 */
402
		writel_relaxed(L2X0_CTRL_EN, l2x0_base + L2X0_CTRL);
403
	}
404

405 406 407 408 409 410
	/* Re-read it in case some bits are reserved. */
	aux = readl_relaxed(l2x0_base + L2X0_AUX_CTRL);

	/* Save the value for resuming. */
	l2x0_saved_regs.aux_ctrl = aux;

411 412 413 414 415 416 417 418 419
	if (!of_init) {
		outer_cache.inv_range = l2x0_inv_range;
		outer_cache.clean_range = l2x0_clean_range;
		outer_cache.flush_range = l2x0_flush_range;
		outer_cache.sync = l2x0_cache_sync;
		outer_cache.flush_all = l2x0_flush_all;
		outer_cache.inv_all = l2x0_inv_all;
		outer_cache.disable = l2x0_disable;
	}
420

421
	printk(KERN_INFO "%s cache controller enabled\n", type);
422 423
	printk(KERN_INFO "l2x0: %d ways, CACHE_ID 0x%08x, AUX_CTRL 0x%08x, Cache size: %d B\n",
			ways, cache_id, aux, l2x0_size);
424
}
425 426

#ifdef CONFIG_OF
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
static int l2_wt_override;

/*
 * Note that the end addresses passed to Linux primitives are
 * noninclusive, while the hardware cache range operations use
 * inclusive start and end addresses.
 */
static unsigned long calc_range_end(unsigned long start, unsigned long end)
{
	/*
	 * Limit the number of cache lines processed at once,
	 * since cache range operations stall the CPU pipeline
	 * until completion.
	 */
	if (end > start + MAX_RANGE_SIZE)
		end = start + MAX_RANGE_SIZE;

	/*
	 * Cache range operations can't straddle a page boundary.
	 */
	if (end > PAGE_ALIGN(start+1))
		end = PAGE_ALIGN(start+1);

	return end;
}

/*
 * Make sure 'start' and 'end' reference the same page, as L2 is PIPT
 * and range operations only do a TLB lookup on the start address.
 */
static void aurora_pa_range(unsigned long start, unsigned long end,
			unsigned long offset)
{
	unsigned long flags;

	raw_spin_lock_irqsave(&l2x0_lock, flags);
463 464
	writel_relaxed(start, l2x0_base + AURORA_RANGE_BASE_ADDR_REG);
	writel_relaxed(end, l2x0_base + offset);
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
	raw_spin_unlock_irqrestore(&l2x0_lock, flags);

	cache_sync();
}

static void aurora_inv_range(unsigned long start, unsigned long end)
{
	/*
	 * round start and end adresses up to cache line size
	 */
	start &= ~(CACHE_LINE_SIZE - 1);
	end = ALIGN(end, CACHE_LINE_SIZE);

	/*
	 * Invalidate all full cache lines between 'start' and 'end'.
	 */
	while (start < end) {
		unsigned long range_end = calc_range_end(start, end);
		aurora_pa_range(start, range_end - CACHE_LINE_SIZE,
				AURORA_INVAL_RANGE_REG);
		start = range_end;
	}
}

static void aurora_clean_range(unsigned long start, unsigned long end)
{
	/*
	 * If L2 is forced to WT, the L2 will always be clean and we
	 * don't need to do anything here.
	 */
	if (!l2_wt_override) {
		start &= ~(CACHE_LINE_SIZE - 1);
		end = ALIGN(end, CACHE_LINE_SIZE);
		while (start != end) {
			unsigned long range_end = calc_range_end(start, end);
			aurora_pa_range(start, range_end - CACHE_LINE_SIZE,
					AURORA_CLEAN_RANGE_REG);
			start = range_end;
		}
	}
}

static void aurora_flush_range(unsigned long start, unsigned long end)
{
509 510 511 512 513 514 515 516 517
	start &= ~(CACHE_LINE_SIZE - 1);
	end = ALIGN(end, CACHE_LINE_SIZE);
	while (start != end) {
		unsigned long range_end = calc_range_end(start, end);
		/*
		 * If L2 is forced to WT, the L2 will always be clean and we
		 * just need to invalidate.
		 */
		if (l2_wt_override)
518
			aurora_pa_range(start, range_end - CACHE_LINE_SIZE,
519 520 521 522 523
							AURORA_INVAL_RANGE_REG);
		else
			aurora_pa_range(start, range_end - CACHE_LINE_SIZE,
							AURORA_FLUSH_RANGE_REG);
		start = range_end;
524 525 526
	}
}

527
static void __init l2x0_of_setup(const struct device_node *np,
528
				 u32 *aux_val, u32 *aux_mask)
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
{
	u32 data[2] = { 0, 0 };
	u32 tag = 0;
	u32 dirty = 0;
	u32 val = 0, mask = 0;

	of_property_read_u32(np, "arm,tag-latency", &tag);
	if (tag) {
		mask |= L2X0_AUX_CTRL_TAG_LATENCY_MASK;
		val |= (tag - 1) << L2X0_AUX_CTRL_TAG_LATENCY_SHIFT;
	}

	of_property_read_u32_array(np, "arm,data-latency",
				   data, ARRAY_SIZE(data));
	if (data[0] && data[1]) {
		mask |= L2X0_AUX_CTRL_DATA_RD_LATENCY_MASK |
			L2X0_AUX_CTRL_DATA_WR_LATENCY_MASK;
		val |= ((data[0] - 1) << L2X0_AUX_CTRL_DATA_RD_LATENCY_SHIFT) |
		       ((data[1] - 1) << L2X0_AUX_CTRL_DATA_WR_LATENCY_SHIFT);
	}

	of_property_read_u32(np, "arm,dirty-latency", &dirty);
	if (dirty) {
		mask |= L2X0_AUX_CTRL_DIRTY_LATENCY_MASK;
		val |= (dirty - 1) << L2X0_AUX_CTRL_DIRTY_LATENCY_SHIFT;
	}

	*aux_val &= ~mask;
	*aux_val |= val;
	*aux_mask &= ~mask;
}

static void __init pl310_of_setup(const struct device_node *np,
562
				  u32 *aux_val, u32 *aux_mask)
563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586
{
	u32 data[3] = { 0, 0, 0 };
	u32 tag[3] = { 0, 0, 0 };
	u32 filter[2] = { 0, 0 };

	of_property_read_u32_array(np, "arm,tag-latency", tag, ARRAY_SIZE(tag));
	if (tag[0] && tag[1] && tag[2])
		writel_relaxed(
			((tag[0] - 1) << L2X0_LATENCY_CTRL_RD_SHIFT) |
			((tag[1] - 1) << L2X0_LATENCY_CTRL_WR_SHIFT) |
			((tag[2] - 1) << L2X0_LATENCY_CTRL_SETUP_SHIFT),
			l2x0_base + L2X0_TAG_LATENCY_CTRL);

	of_property_read_u32_array(np, "arm,data-latency",
				   data, ARRAY_SIZE(data));
	if (data[0] && data[1] && data[2])
		writel_relaxed(
			((data[0] - 1) << L2X0_LATENCY_CTRL_RD_SHIFT) |
			((data[1] - 1) << L2X0_LATENCY_CTRL_WR_SHIFT) |
			((data[2] - 1) << L2X0_LATENCY_CTRL_SETUP_SHIFT),
			l2x0_base + L2X0_DATA_LATENCY_CTRL);

	of_property_read_u32_array(np, "arm,filter-ranges",
				   filter, ARRAY_SIZE(filter));
587
	if (filter[1]) {
588 589 590 591 592 593 594
		writel_relaxed(ALIGN(filter[0] + filter[1], SZ_1M),
			       l2x0_base + L2X0_ADDR_FILTER_END);
		writel_relaxed((filter[0] & ~(SZ_1M - 1)) | L2X0_ADDR_FILTER_EN,
			       l2x0_base + L2X0_ADDR_FILTER_START);
	}
}

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
static void __init pl310_save(void)
{
	u32 l2x0_revision = readl_relaxed(l2x0_base + L2X0_CACHE_ID) &
		L2X0_CACHE_ID_RTL_MASK;

	l2x0_saved_regs.tag_latency = readl_relaxed(l2x0_base +
		L2X0_TAG_LATENCY_CTRL);
	l2x0_saved_regs.data_latency = readl_relaxed(l2x0_base +
		L2X0_DATA_LATENCY_CTRL);
	l2x0_saved_regs.filter_end = readl_relaxed(l2x0_base +
		L2X0_ADDR_FILTER_END);
	l2x0_saved_regs.filter_start = readl_relaxed(l2x0_base +
		L2X0_ADDR_FILTER_START);

	if (l2x0_revision >= L2X0_CACHE_ID_RTL_R2P0) {
		/*
		 * From r2p0, there is Prefetch offset/control register
		 */
		l2x0_saved_regs.prefetch_ctrl = readl_relaxed(l2x0_base +
			L2X0_PREFETCH_CTRL);
		/*
		 * From r3p0, there is Power control register
		 */
		if (l2x0_revision >= L2X0_CACHE_ID_RTL_R3P0)
			l2x0_saved_regs.pwr_ctrl = readl_relaxed(l2x0_base +
				L2X0_POWER_CTRL);
	}
}

624 625 626 627 628 629
static void aurora_save(void)
{
	l2x0_saved_regs.ctrl = readl_relaxed(l2x0_base + L2X0_CTRL);
	l2x0_saved_regs.aux_ctrl = readl_relaxed(l2x0_base + L2X0_AUX_CTRL);
}

630 631
static void l2x0_resume(void)
{
632
	if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) {
633 634 635 636 637 638 639 640
		/* restore aux ctrl and enable l2 */
		l2x0_unlock(readl_relaxed(l2x0_base + L2X0_CACHE_ID));

		writel_relaxed(l2x0_saved_regs.aux_ctrl, l2x0_base +
			L2X0_AUX_CTRL);

		l2x0_inv_all();

641
		writel_relaxed(L2X0_CTRL_EN, l2x0_base + L2X0_CTRL);
642 643 644 645 646 647 648
	}
}

static void pl310_resume(void)
{
	u32 l2x0_revision;

649
	if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) {
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
		/* restore pl310 setup */
		writel_relaxed(l2x0_saved_regs.tag_latency,
			l2x0_base + L2X0_TAG_LATENCY_CTRL);
		writel_relaxed(l2x0_saved_regs.data_latency,
			l2x0_base + L2X0_DATA_LATENCY_CTRL);
		writel_relaxed(l2x0_saved_regs.filter_end,
			l2x0_base + L2X0_ADDR_FILTER_END);
		writel_relaxed(l2x0_saved_regs.filter_start,
			l2x0_base + L2X0_ADDR_FILTER_START);

		l2x0_revision = readl_relaxed(l2x0_base + L2X0_CACHE_ID) &
			L2X0_CACHE_ID_RTL_MASK;

		if (l2x0_revision >= L2X0_CACHE_ID_RTL_R2P0) {
			writel_relaxed(l2x0_saved_regs.prefetch_ctrl,
				l2x0_base + L2X0_PREFETCH_CTRL);
			if (l2x0_revision >= L2X0_CACHE_ID_RTL_R3P0)
				writel_relaxed(l2x0_saved_regs.pwr_ctrl,
					l2x0_base + L2X0_POWER_CTRL);
		}
	}

	l2x0_resume();
}

675 676 677
static void aurora_resume(void)
{
	if (!(readl(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) {
678 679 680
		writel_relaxed(l2x0_saved_regs.aux_ctrl,
				l2x0_base + L2X0_AUX_CTRL);
		writel_relaxed(l2x0_saved_regs.ctrl, l2x0_base + L2X0_CTRL);
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
	}
}

static void __init aurora_broadcast_l2_commands(void)
{
	__u32 u;
	/* Enable Broadcasting of cache commands to L2*/
	__asm__ __volatile__("mrc p15, 1, %0, c15, c2, 0" : "=r"(u));
	u |= AURORA_CTRL_FW;		/* Set the FW bit */
	__asm__ __volatile__("mcr p15, 1, %0, c15, c2, 0\n" : : "r"(u));
	isb();
}

static void __init aurora_of_setup(const struct device_node *np,
				u32 *aux_val, u32 *aux_mask)
{
	u32 val = AURORA_ACR_REPLACEMENT_TYPE_SEMIPLRU;
	u32 mask =  AURORA_ACR_REPLACEMENT_MASK;

	of_property_read_u32(np, "cache-id-part",
			&cache_id_part_number_from_dt);

	/* Determine and save the write policy */
	l2_wt_override = of_property_read_bool(np, "wt-override");

	if (l2_wt_override) {
		val |= AURORA_ACR_FORCE_WRITE_THRO_POLICY;
		mask |= AURORA_ACR_FORCE_WRITE_POLICY_MASK;
	}

	*aux_val &= ~mask;
	*aux_val |= val;
	*aux_mask &= ~mask;
}

716
static const struct l2x0_of_data pl310_data = {
717 718 719 720 721 722 723 724 725 726 727 728 729
	.setup = pl310_of_setup,
	.save  = pl310_save,
	.outer_cache = {
		.resume      = pl310_resume,
		.inv_range   = l2x0_inv_range,
		.clean_range = l2x0_clean_range,
		.flush_range = l2x0_flush_range,
		.sync        = l2x0_cache_sync,
		.flush_all   = l2x0_flush_all,
		.inv_all     = l2x0_inv_all,
		.disable     = l2x0_disable,
		.set_debug   = pl310_set_debug,
	},
730 731 732
};

static const struct l2x0_of_data l2x0_data = {
733 734 735 736 737 738 739 740 741 742 743 744
	.setup = l2x0_of_setup,
	.save  = NULL,
	.outer_cache = {
		.resume      = l2x0_resume,
		.inv_range   = l2x0_inv_range,
		.clean_range = l2x0_clean_range,
		.flush_range = l2x0_flush_range,
		.sync        = l2x0_cache_sync,
		.flush_all   = l2x0_flush_all,
		.inv_all     = l2x0_inv_all,
		.disable     = l2x0_disable,
	},
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
static const struct l2x0_of_data aurora_with_outer_data = {
	.setup = aurora_of_setup,
	.save  = aurora_save,
	.outer_cache = {
		.resume      = aurora_resume,
		.inv_range   = aurora_inv_range,
		.clean_range = aurora_clean_range,
		.flush_range = aurora_flush_range,
		.sync        = l2x0_cache_sync,
		.flush_all   = l2x0_flush_all,
		.inv_all     = l2x0_inv_all,
		.disable     = l2x0_disable,
	},
};

static const struct l2x0_of_data aurora_no_outer_data = {
	.setup = aurora_of_setup,
	.save  = aurora_save,
	.outer_cache = {
		.resume      = aurora_resume,
	},
};

770
static const struct of_device_id l2x0_ids[] __initconst = {
771 772 773
	{ .compatible = "arm,pl310-cache", .data = (void *)&pl310_data },
	{ .compatible = "arm,l220-cache", .data = (void *)&l2x0_data },
	{ .compatible = "arm,l210-cache", .data = (void *)&l2x0_data },
774 775 776 777
	{ .compatible = "marvell,aurora-system-cache",
	  .data = (void *)&aurora_no_outer_data},
	{ .compatible = "marvell,aurora-outer-cache",
	  .data = (void *)&aurora_with_outer_data},
778 779 780
	{}
};

781
int __init l2x0_of_init(u32 aux_val, u32 aux_mask)
782 783
{
	struct device_node *np;
784
	const struct l2x0_of_data *data;
785
	struct resource res;
786 787 788 789

	np = of_find_matching_node(NULL, l2x0_ids);
	if (!np)
		return -ENODEV;
790 791 792 793 794

	if (of_address_to_resource(np, 0, &res))
		return -ENODEV;

	l2x0_base = ioremap(res.start, resource_size(&res));
795 796 797
	if (!l2x0_base)
		return -ENOMEM;

798 799 800 801
	l2x0_saved_regs.phy_base = res.start;

	data = of_match_node(l2x0_ids, np)->data;

802
	/* L2 configuration can only be changed if the cache is disabled */
803
	if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) {
804 805
		if (data->setup)
			data->setup(np, &aux_val, &aux_mask);
806 807 808 809 810

		/* For aurora cache in no outer mode select the
		 * correct mode using the coprocessor*/
		if (data == &aurora_no_outer_data)
			aurora_broadcast_l2_commands();
811
	}
812 813 814 815

	if (data->save)
		data->save();

816
	of_init = true;
817
	l2x0_init(l2x0_base, aux_val, aux_mask);
818

819 820
	memcpy(&outer_cache, &data->outer_cache, sizeof(outer_cache));

821 822 823
	return 0;
}
#endif