aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/pci/controller/dwc/pci-layerscape.c
blob: ee6f5256813374bdf656bef4f9b96e1b8760d1b5 (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
// SPDX-License-Identifier: GPL-2.0
/*
 * PCIe host controller driver for Freescale Layerscape SoCs
 *
 * Copyright (C) 2014 Freescale Semiconductor.
 * Copyright 2021 NXP
 *
 * Author: Minghuan Lian <Minghuan.Lian@freescale.com>
 */

#include <linux/delay.h>
#include <linux/kernel.h>
#include <linux/interrupt.h>
#include <linux/init.h>
#include <linux/iopoll.h>
#include <linux/of_pci.h>
#include <linux/of_platform.h>
#include <linux/of_address.h>
#include <linux/pci.h>
#include <linux/platform_device.h>
#include <linux/resource.h>
#include <linux/mfd/syscon.h>
#include <linux/regmap.h>

#include "../../pci.h"
#include "pcie-designware.h"

/* PEX Internal Configuration Registers */
#define PCIE_STRFMR1		0x71c /* Symbol Timer & Filter Mask Register1 */
#define PCIE_ABSERR		0x8d0 /* Bridge Slave Error Response Register */
#define PCIE_ABSERR_SETTING	0x9401 /* Forward error of non-posted request */

/* PF Message Command Register */
#define LS_PCIE_PF_MCR		0x2c
#define PF_MCR_PTOMR		BIT(0)
#define PF_MCR_EXL2S		BIT(1)

/* LS1021A PEXn PM Write Control Register */
#define SCFG_PEXPMWRCR(idx)	(0x5c + (idx) * 0x64)
#define PMXMTTURNOFF		BIT(31)
#define SCFG_PEXSFTRSTCR	0x190
#define PEXSR(idx)		BIT(idx)

/* LS1043A PEX PME control register */
#define SCFG_PEXPMECR		0x144
#define PEXPME(idx)		BIT(31 - (idx) * 4)

/* LS1043A PEX LUT debug register */
#define LS_PCIE_LDBG	0x7fc
#define LDBG_SR		BIT(30)
#define LDBG_WE		BIT(31)

#define PCIE_IATU_NUM		6

struct ls_pcie_drvdata {
	const u32 pf_lut_off;
	const struct dw_pcie_host_ops *ops;
	int (*exit_from_l2)(struct dw_pcie_rp *pp);
	bool scfg_support;
	bool pm_support;
};

struct ls_pcie {
	struct dw_pcie *pci;
	const struct ls_pcie_drvdata *drvdata;
	void __iomem *pf_lut_base;
	struct regmap *scfg;
	int index;
	bool big_endian;
};

#define ls_pcie_pf_lut_readl_addr(addr)	ls_pcie_pf_lut_readl(pcie, addr)
#define to_ls_pcie(x)	dev_get_drvdata((x)->dev)

static bool ls_pcie_is_bridge(struct ls_pcie *pcie)
{
	struct dw_pcie *pci = pcie->pci;
	u32 header_type;

	header_type = ioread8(pci->dbi_base + PCI_HEADER_TYPE);
	header_type &= PCI_HEADER_TYPE_MASK;

	return header_type == PCI_HEADER_TYPE_BRIDGE;
}

/* Clear multi-function bit */
static void ls_pcie_clear_multifunction(struct ls_pcie *pcie)
{
	struct dw_pcie *pci = pcie->pci;

	iowrite8(PCI_HEADER_TYPE_BRIDGE, pci->dbi_base + PCI_HEADER_TYPE);
}

/* Drop MSG TLP except for Vendor MSG */
static void ls_pcie_drop_msg_tlp(struct ls_pcie *pcie)
{
	u32 val;
	struct dw_pcie *pci = pcie->pci;

	val = ioread32(pci->dbi_base + PCIE_STRFMR1);
	val &= 0xDFFFFFFF;
	iowrite32(val, pci->dbi_base + PCIE_STRFMR1);
}

/* Forward error response of outbound non-posted requests */
static void ls_pcie_fix_error_response(struct ls_pcie *pcie)
{
	struct dw_pcie *pci = pcie->pci;

	iowrite32(PCIE_ABSERR_SETTING, pci->dbi_base + PCIE_ABSERR);
}

static u32 ls_pcie_pf_lut_readl(struct ls_pcie *pcie, u32 off)
{
	if (pcie->big_endian)
		return ioread32be(pcie->pf_lut_base + off);

	return ioread32(pcie->pf_lut_base + off);
}

static void ls_pcie_pf_lut_writel(struct ls_pcie *pcie, u32 off, u32 val)
{
	if (pcie->big_endian)
		iowrite32be(val, pcie->pf_lut_base + off);
	else
		iowrite32(val, pcie->pf_lut_base + off);
}

static void ls_pcie_send_turnoff_msg(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);
	u32 val;
	int ret;

	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_PF_MCR);
	val |= PF_MCR_PTOMR;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_PF_MCR, val);

	ret = readx_poll_timeout(ls_pcie_pf_lut_readl_addr, LS_PCIE_PF_MCR,
				 val, !(val & PF_MCR_PTOMR),
				 PCIE_PME_TO_L2_TIMEOUT_US/10,
				 PCIE_PME_TO_L2_TIMEOUT_US);
	if (ret)
		dev_err(pcie->pci->dev, "PME_Turn_off timeout\n");
}

static int ls_pcie_exit_from_l2(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);
	u32 val;
	int ret;

	/*
	 * Set PF_MCR_EXL2S bit in LS_PCIE_PF_MCR register for the link
	 * to exit L2 state.
	 */
	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_PF_MCR);
	val |= PF_MCR_EXL2S;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_PF_MCR, val);

	/*
	 * L2 exit timeout of 10ms is not defined in the specifications,
	 * it was chosen based on empirical observations.
	 */
	ret = readx_poll_timeout(ls_pcie_pf_lut_readl_addr, LS_PCIE_PF_MCR,
				 val, !(val & PF_MCR_EXL2S),
				 1000,
				 10000);
	if (ret)
		dev_err(pcie->pci->dev, "L2 exit timeout\n");

	return ret;
}

static int ls_pcie_host_init(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);

	ls_pcie_fix_error_response(pcie);

	dw_pcie_dbi_ro_wr_en(pci);
	ls_pcie_clear_multifunction(pcie);
	dw_pcie_dbi_ro_wr_dis(pci);

	ls_pcie_drop_msg_tlp(pcie);

	return 0;
}

static void scfg_pcie_send_turnoff_msg(struct regmap *scfg, u32 reg, u32 mask)
{
	/* Send PME_Turn_Off message */
	regmap_write_bits(scfg, reg, mask, mask);

	/*
	 * There is no specific register to check for PME_To_Ack from endpoint.
	 * So on the safe side, wait for PCIE_PME_TO_L2_TIMEOUT_US.
	 */
	mdelay(PCIE_PME_TO_L2_TIMEOUT_US/1000);

	/*
	 * Layerscape hardware reference manual recommends clearing the PMXMTTURNOFF bit
	 * to complete the PME_Turn_Off handshake.
	 */
	regmap_write_bits(scfg, reg, mask, 0);
}

static void ls1021a_pcie_send_turnoff_msg(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);

	scfg_pcie_send_turnoff_msg(pcie->scfg, SCFG_PEXPMWRCR(pcie->index), PMXMTTURNOFF);
}

static int scfg_pcie_exit_from_l2(struct regmap *scfg, u32 reg, u32 mask)
{
	/* Reset the PEX wrapper to bring the link out of L2 */
	regmap_write_bits(scfg, reg, mask, mask);
	regmap_write_bits(scfg, reg, mask, 0);

	return 0;
}

static int ls1021a_pcie_exit_from_l2(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);

	return scfg_pcie_exit_from_l2(pcie->scfg, SCFG_PEXSFTRSTCR, PEXSR(pcie->index));
}

static void ls1043a_pcie_send_turnoff_msg(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);

	scfg_pcie_send_turnoff_msg(pcie->scfg, SCFG_PEXPMECR, PEXPME(pcie->index));
}

static int ls1043a_pcie_exit_from_l2(struct dw_pcie_rp *pp)
{
	struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
	struct ls_pcie *pcie = to_ls_pcie(pci);
	u32 val;

	/*
	 * Reset the PEX wrapper to bring the link out of L2.
	 * LDBG_WE: allows the user to have write access to the PEXDBG[SR] for both setting and
	 *	    clearing the soft reset on the PEX module.
	 * LDBG_SR: When SR is set to 1, the PEX module enters soft reset.
	 */
	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_LDBG);
	val |= LDBG_WE;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_LDBG, val);

	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_LDBG);
	val |= LDBG_SR;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_LDBG, val);

	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_LDBG);
	val &= ~LDBG_SR;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_LDBG, val);

	val = ls_pcie_pf_lut_readl(pcie, LS_PCIE_LDBG);
	val &= ~LDBG_WE;
	ls_pcie_pf_lut_writel(pcie, LS_PCIE_LDBG, val);

	return 0;
}

static const struct dw_pcie_host_ops ls_pcie_host_ops = {
	.init = ls_pcie_host_init,
	.pme_turn_off = ls_pcie_send_turnoff_msg,
};

static const struct dw_pcie_host_ops ls1021a_pcie_host_ops = {
	.init = ls_pcie_host_init,
	.pme_turn_off = ls1021a_pcie_send_turnoff_msg,
};

static const struct ls_pcie_drvdata ls1021a_drvdata = {
	.pm_support = true,
	.scfg_support = true,
	.ops = &ls1021a_pcie_host_ops,
	.exit_from_l2 = ls1021a_pcie_exit_from_l2,
};

static const struct dw_pcie_host_ops ls1043a_pcie_host_ops = {
	.init = ls_pcie_host_init,
	.pme_turn_off = ls1043a_pcie_send_turnoff_msg,
};

static const struct ls_pcie_drvdata ls1043a_drvdata = {
	.pf_lut_off = 0x10000,
	.pm_support = true,
	.scfg_support = true,
	.ops = &ls1043a_pcie_host_ops,
	.exit_from_l2 = ls1043a_pcie_exit_from_l2,
};

static const struct ls_pcie_drvdata layerscape_drvdata = {
	.pf_lut_off = 0xc0000,
	.pm_support = true,
	.ops = &ls_pcie_host_ops,
	.exit_from_l2 = ls_pcie_exit_from_l2,
};

static const struct of_device_id ls_pcie_of_match[] = {
	{ .compatible = "fsl,ls1012a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls1021a-pcie", .data = &ls1021a_drvdata },
	{ .compatible = "fsl,ls1028a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls1043a-pcie", .data = &ls1043a_drvdata },
	{ .compatible = "fsl,ls1046a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls2080a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls2085a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls2088a-pcie", .data = &layerscape_drvdata },
	{ .compatible = "fsl,ls1088a-pcie", .data = &layerscape_drvdata },
	{ },
};

static int ls_pcie_probe(struct platform_device *pdev)
{
	struct device *dev = &pdev->dev;
	struct dw_pcie *pci;
	struct ls_pcie *pcie;
	struct resource *dbi_base;
	u32 index[2];
	int ret;

	pcie = devm_kzalloc(dev, sizeof(*pcie), GFP_KERNEL);
	if (!pcie)
		return -ENOMEM;

	pci = devm_kzalloc(dev, sizeof(*pci), GFP_KERNEL);
	if (!pci)
		return -ENOMEM;

	pcie->drvdata = of_device_get_match_data(dev);

	pci->dev = dev;
	pcie->pci = pci;
	pci->pp.ops = pcie->drvdata->ops;

	dbi_base = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs");
	pci->dbi_base = devm_pci_remap_cfg_resource(dev, dbi_base);
	if (IS_ERR(pci->dbi_base))
		return PTR_ERR(pci->dbi_base);

	pcie->big_endian = of_property_read_bool(dev->of_node, "big-endian");

	pcie->pf_lut_base = pci->dbi_base + pcie->drvdata->pf_lut_off;

	if (pcie->drvdata->scfg_support) {
		pcie->scfg = syscon_regmap_lookup_by_phandle(dev->of_node, "fsl,pcie-scfg");
		if (IS_ERR(pcie->scfg)) {
			dev_err(dev, "No syscfg phandle specified\n");
			return PTR_ERR(pcie->scfg);
		}

		ret = of_property_read_u32_array(dev->of_node, "fsl,pcie-scfg", index, 2);
		if (ret)
			return ret;

		pcie->index = index[1];
	}

	if (!ls_pcie_is_bridge(pcie))
		return -ENODEV;

	platform_set_drvdata(pdev, pcie);

	return dw_pcie_host_init(&pci->pp);
}

static int ls_pcie_suspend_noirq(struct device *dev)
{
	struct ls_pcie *pcie = dev_get_drvdata(dev);

	if (!pcie->drvdata->pm_support)
		return 0;

	return dw_pcie_suspend_noirq(pcie->pci);
}

static int ls_pcie_resume_noirq(struct device *dev)
{
	struct ls_pcie *pcie = dev_get_drvdata(dev);
	int ret;

	if (!pcie->drvdata->pm_support)
		return 0;

	ret = pcie->drvdata->exit_from_l2(&pcie->pci->pp);
	if (ret)
		return ret;

	return dw_pcie_resume_noirq(pcie->pci);
}

static const struct dev_pm_ops ls_pcie_pm_ops = {
	NOIRQ_SYSTEM_SLEEP_PM_OPS(ls_pcie_suspend_noirq, ls_pcie_resume_noirq)
};

static struct platform_driver ls_pcie_driver = {
	.probe = ls_pcie_probe,
	.driver = {
		.name = "layerscape-pcie",
		.of_match_table = ls_pcie_of_match,
		.suppress_bind_attrs = true,
		.pm = &ls_pcie_pm_ops,
	},
};
builtin_platform_driver(ls_pcie_driver);
1802' href='#n1802'>1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823
/*
 * Copyright (C) 2011 STRATO.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 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 021110-1307, USA.
 */

#include <linux/vmalloc.h>
#include "ctree.h"
#include "disk-io.h"
#include "backref.h"
#include "ulist.h"
#include "transaction.h"
#include "delayed-ref.h"
#include "locking.h"

struct extent_inode_elem {
	u64 inum;
	u64 offset;
	struct extent_inode_elem *next;
};

static int check_extent_in_eb(struct btrfs_key *key, struct extent_buffer *eb,
				struct btrfs_file_extent_item *fi,
				u64 extent_item_pos,
				struct extent_inode_elem **eie)
{
	u64 offset = 0;
	struct extent_inode_elem *e;

	if (!btrfs_file_extent_compression(eb, fi) &&
	    !btrfs_file_extent_encryption(eb, fi) &&
	    !btrfs_file_extent_other_encoding(eb, fi)) {
		u64 data_offset;
		u64 data_len;

		data_offset = btrfs_file_extent_offset(eb, fi);
		data_len = btrfs_file_extent_num_bytes(eb, fi);

		if (extent_item_pos < data_offset ||
		    extent_item_pos >= data_offset + data_len)
			return 1;
		offset = extent_item_pos - data_offset;
	}

	e = kmalloc(sizeof(*e), GFP_NOFS);
	if (!e)
		return -ENOMEM;

	e->next = *eie;
	e->inum = key->objectid;
	e->offset = key->offset + offset;
	*eie = e;

	return 0;
}

static int find_extent_in_eb(struct extent_buffer *eb, u64 wanted_disk_byte,
				u64 extent_item_pos,
				struct extent_inode_elem **eie)
{
	u64 disk_byte;
	struct btrfs_key key;
	struct btrfs_file_extent_item *fi;
	int slot;
	int nritems;
	int extent_type;
	int ret;

	/*
	 * from the shared data ref, we only have the leaf but we need
	 * the key. thus, we must look into all items and see that we
	 * find one (some) with a reference to our extent item.
	 */
	nritems = btrfs_header_nritems(eb);
	for (slot = 0; slot < nritems; ++slot) {
		btrfs_item_key_to_cpu(eb, &key, slot);
		if (key.type != BTRFS_EXTENT_DATA_KEY)
			continue;
		fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
		extent_type = btrfs_file_extent_type(eb, fi);
		if (extent_type == BTRFS_FILE_EXTENT_INLINE)
			continue;
		/* don't skip BTRFS_FILE_EXTENT_PREALLOC, we can handle that */
		disk_byte = btrfs_file_extent_disk_bytenr(eb, fi);
		if (disk_byte != wanted_disk_byte)
			continue;

		ret = check_extent_in_eb(&key, eb, fi, extent_item_pos, eie);
		if (ret < 0)
			return ret;
	}

	return 0;
}

/*
 * this structure records all encountered refs on the way up to the root
 */
struct __prelim_ref {
	struct list_head list;
	u64 root_id;
	struct btrfs_key key_for_search;
	int level;
	int count;
	struct extent_inode_elem *inode_list;
	u64 parent;
	u64 wanted_disk_byte;
};

/*
 * the rules for all callers of this function are:
 * - obtaining the parent is the goal
 * - if you add a key, you must know that it is a correct key
 * - if you cannot add the parent or a correct key, then we will look into the
 *   block later to set a correct key
 *
 * delayed refs
 * ============
 *        backref type | shared | indirect | shared | indirect
 * information         |   tree |     tree |   data |     data
 * --------------------+--------+----------+--------+----------
 *      parent logical |    y   |     -    |    -   |     -
 *      key to resolve |    -   |     y    |    y   |     y
 *  tree block logical |    -   |     -    |    -   |     -
 *  root for resolving |    y   |     y    |    y   |     y
 *
 * - column 1:       we've the parent -> done
 * - column 2, 3, 4: we use the key to find the parent
 *
 * on disk refs (inline or keyed)
 * ==============================
 *        backref type | shared | indirect | shared | indirect
 * information         |   tree |     tree |   data |     data
 * --------------------+--------+----------+--------+----------
 *      parent logical |    y   |     -    |    y   |     -
 *      key to resolve |    -   |     -    |    -   |     y
 *  tree block logical |    y   |     y    |    y   |     y
 *  root for resolving |    -   |     y    |    y   |     y
 *
 * - column 1, 3: we've the parent -> done
 * - column 2:    we take the first key from the block to find the parent
 *                (see __add_missing_keys)
 * - column 4:    we use the key to find the parent
 *
 * additional information that's available but not required to find the parent
 * block might help in merging entries to gain some speed.
 */

static int __add_prelim_ref(struct list_head *head, u64 root_id,
			    struct btrfs_key *key, int level,
			    u64 parent, u64 wanted_disk_byte, int count)
{
	struct __prelim_ref *ref;

	/* in case we're adding delayed refs, we're holding the refs spinlock */
	ref = kmalloc(sizeof(*ref), GFP_ATOMIC);
	if (!ref)
		return -ENOMEM;

	ref->root_id = root_id;
	if (key)
		ref->key_for_search = *key;
	else
		memset(&ref->key_for_search, 0, sizeof(ref->key_for_search));

	ref->inode_list = NULL;
	ref->level = level;
	ref->count = count;
	ref->parent = parent;
	ref->wanted_disk_byte = wanted_disk_byte;
	list_add_tail(&ref->list, head);

	return 0;
}

static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
				struct ulist *parents, int level,
				struct btrfs_key *key_for_search, u64 time_seq,
				u64 wanted_disk_byte,
				const u64 *extent_item_pos)
{
	int ret = 0;
	int slot;
	struct extent_buffer *eb;
	struct btrfs_key key;
	struct btrfs_file_extent_item *fi;
	struct extent_inode_elem *eie = NULL, *old = NULL;
	u64 disk_byte;

	if (level != 0) {
		eb = path->nodes[level];
		ret = ulist_add(parents, eb->start, 0, GFP_NOFS);
		if (ret < 0)
			return ret;
		return 0;
	}

	/*
	 * We normally enter this function with the path already pointing to
	 * the first item to check. But sometimes, we may enter it with
	 * slot==nritems. In that case, go to the next leaf before we continue.
	 */
	if (path->slots[0] >= btrfs_header_nritems(path->nodes[0]))
		ret = btrfs_next_old_leaf(root, path, time_seq);

	while (!ret) {
		eb = path->nodes[0];
		slot = path->slots[0];

		btrfs_item_key_to_cpu(eb, &key, slot);

		if (key.objectid != key_for_search->objectid ||
		    key.type != BTRFS_EXTENT_DATA_KEY)
			break;

		fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
		disk_byte = btrfs_file_extent_disk_bytenr(eb, fi);

		if (disk_byte == wanted_disk_byte) {
			eie = NULL;
			old = NULL;
			if (extent_item_pos) {
				ret = check_extent_in_eb(&key, eb, fi,
						*extent_item_pos,
						&eie);
				if (ret < 0)
					break;
			}
			if (ret > 0)
				goto next;
			ret = ulist_add_merge(parents, eb->start,
					      (uintptr_t)eie,
					      (u64 *)&old, GFP_NOFS);
			if (ret < 0)
				break;
			if (!ret && extent_item_pos) {
				while (old->next)
					old = old->next;
				old->next = eie;
			}
		}
next:
		ret = btrfs_next_old_item(root, path, time_seq);
	}

	if (ret > 0)
		ret = 0;
	return ret;
}

/*
 * resolve an indirect backref in the form (root_id, key, level)
 * to a logical address
 */
static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
				  struct btrfs_path *path, u64 time_seq,
				  struct __prelim_ref *ref,
				  struct ulist *parents,
				  const u64 *extent_item_pos)
{
	struct btrfs_root *root;
	struct btrfs_key root_key;
	struct extent_buffer *eb;
	int ret = 0;
	int root_level;
	int level = ref->level;

	root_key.objectid = ref->root_id;
	root_key.type = BTRFS_ROOT_ITEM_KEY;
	root_key.offset = (u64)-1;
	root = btrfs_read_fs_root_no_name(fs_info, &root_key);
	if (IS_ERR(root)) {
		ret = PTR_ERR(root);
		goto out;
	}

	root_level = btrfs_old_root_level(root, time_seq);

	if (root_level + 1 == level)
		goto out;

	path->lowest_level = level;
	ret = btrfs_search_old_slot(root, &ref->key_for_search, path, time_seq);
	pr_debug("search slot in root %llu (level %d, ref count %d) returned "
		 "%d for key (%llu %u %llu)\n",
		 ref->root_id, level, ref->count, ret,
		 ref->key_for_search.objectid, ref->key_for_search.type,
		 ref->key_for_search.offset);
	if (ret < 0)
		goto out;

	eb = path->nodes[level];
	while (!eb) {
		if (!level) {
			WARN_ON(1);
			ret = 1;
			goto out;
		}
		level--;
		eb = path->nodes[level];
	}

	ret = add_all_parents(root, path, parents, level, &ref->key_for_search,
				time_seq, ref->wanted_disk_byte,
				extent_item_pos);
out:
	path->lowest_level = 0;
	btrfs_release_path(path);
	return ret;
}

/*
 * resolve all indirect backrefs from the list
 */
static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info,
				   struct btrfs_path *path, u64 time_seq,
				   struct list_head *head,
				   const u64 *extent_item_pos)
{
	int err;
	int ret = 0;
	struct __prelim_ref *ref;
	struct __prelim_ref *ref_safe;
	struct __prelim_ref *new_ref;
	struct ulist *parents;
	struct ulist_node *node;
	struct ulist_iterator uiter;

	parents = ulist_alloc(GFP_NOFS);
	if (!parents)
		return -ENOMEM;

	/*
	 * _safe allows us to insert directly after the current item without
	 * iterating over the newly inserted items.
	 * we're also allowed to re-assign ref during iteration.
	 */
	list_for_each_entry_safe(ref, ref_safe, head, list) {
		if (ref->parent)	/* already direct */
			continue;
		if (ref->count == 0)
			continue;
		err = __resolve_indirect_ref(fs_info, path, time_seq, ref,
					     parents, extent_item_pos);
		if (err == -ENOMEM)
			goto out;
		if (err)
			continue;

		/* we put the first parent into the ref at hand */
		ULIST_ITER_INIT(&uiter);
		node = ulist_next(parents, &uiter);
		ref->parent = node ? node->val : 0;
		ref->inode_list = node ?
			(struct extent_inode_elem *)(uintptr_t)node->aux : NULL;

		/* additional parents require new refs being added here */
		while ((node = ulist_next(parents, &uiter))) {
			new_ref = kmalloc(sizeof(*new_ref), GFP_NOFS);
			if (!new_ref) {
				ret = -ENOMEM;
				goto out;
			}
			memcpy(new_ref, ref, sizeof(*ref));
			new_ref->parent = node->val;
			new_ref->inode_list = (struct extent_inode_elem *)
							(uintptr_t)node->aux;
			list_add(&new_ref->list, &ref->list);
		}
		ulist_reinit(parents);
	}
out:
	ulist_free(parents);
	return ret;
}

static inline int ref_for_same_block(struct __prelim_ref *ref1,
				     struct __prelim_ref *ref2)
{
	if (ref1->level != ref2->level)
		return 0;
	if (ref1->root_id != ref2->root_id)
		return 0;
	if (ref1->key_for_search.type != ref2->key_for_search.type)
		return 0;
	if (ref1->key_for_search.objectid != ref2->key_for_search.objectid)
		return 0;
	if (ref1->key_for_search.offset != ref2->key_for_search.offset)
		return 0;
	if (ref1->parent != ref2->parent)
		return 0;

	return 1;
}

/*
 * read tree blocks and add keys where required.
 */
static int __add_missing_keys(struct btrfs_fs_info *fs_info,
			      struct list_head *head)
{
	struct list_head *pos;
	struct extent_buffer *eb;

	list_for_each(pos, head) {
		struct __prelim_ref *ref;
		ref = list_entry(pos, struct __prelim_ref, list);

		if (ref->parent)
			continue;
		if (ref->key_for_search.type)
			continue;
		BUG_ON(!ref->wanted_disk_byte);
		eb = read_tree_block(fs_info->tree_root, ref->wanted_disk_byte,
				     fs_info->tree_root->leafsize, 0);
		if (!eb || !extent_buffer_uptodate(eb)) {
			free_extent_buffer(eb);
			return -EIO;
		}
		btrfs_tree_read_lock(eb);
		if (btrfs_header_level(eb) == 0)
			btrfs_item_key_to_cpu(eb, &ref->key_for_search, 0);
		else
			btrfs_node_key_to_cpu(eb, &ref->key_for_search, 0);
		btrfs_tree_read_unlock(eb);
		free_extent_buffer(eb);
	}
	return 0;
}

/*
 * merge two lists of backrefs and adjust counts accordingly
 *
 * mode = 1: merge identical keys, if key is set
 *    FIXME: if we add more keys in __add_prelim_ref, we can merge more here.
 *           additionally, we could even add a key range for the blocks we
 *           looked into to merge even more (-> replace unresolved refs by those
 *           having a parent).
 * mode = 2: merge identical parents
 */
static void __merge_refs(struct list_head *head, int mode)
{
	struct list_head *pos1;

	list_for_each(pos1, head) {
		struct list_head *n2;
		struct list_head *pos2;
		struct __prelim_ref *ref1;

		ref1 = list_entry(pos1, struct __prelim_ref, list);

		for (pos2 = pos1->next, n2 = pos2->next; pos2 != head;
		     pos2 = n2, n2 = pos2->next) {
			struct __prelim_ref *ref2;
			struct __prelim_ref *xchg;
			struct extent_inode_elem *eie;

			ref2 = list_entry(pos2, struct __prelim_ref, list);

			if (mode == 1) {
				if (!ref_for_same_block(ref1, ref2))
					continue;
				if (!ref1->parent && ref2->parent) {
					xchg = ref1;
					ref1 = ref2;
					ref2 = xchg;
				}
			} else {
				if (ref1->parent != ref2->parent)
					continue;
			}

			eie = ref1->inode_list;
			while (eie && eie->next)
				eie = eie->next;
			if (eie)
				eie->next = ref2->inode_list;
			else
				ref1->inode_list = ref2->inode_list;
			ref1->count += ref2->count;

			list_del(&ref2->list);
			kfree(ref2);
		}

	}
}

/*
 * add all currently queued delayed refs from this head whose seq nr is
 * smaller or equal that seq to the list
 */
static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq,
			      struct list_head *prefs)
{
	struct btrfs_delayed_extent_op *extent_op = head->extent_op;
	struct rb_node *n = &head->node.rb_node;
	struct btrfs_key key;
	struct btrfs_key op_key = {0};
	int sgn;
	int ret = 0;

	if (extent_op && extent_op->update_key)
		btrfs_disk_key_to_cpu(&op_key, &extent_op->key);

	while ((n = rb_prev(n))) {
		struct btrfs_delayed_ref_node *node;
		node = rb_entry(n, struct btrfs_delayed_ref_node,
				rb_node);
		if (node->bytenr != head->node.bytenr)
			break;
		WARN_ON(node->is_head);

		if (node->seq > seq)
			continue;

		switch (node->action) {
		case BTRFS_ADD_DELAYED_EXTENT:
		case BTRFS_UPDATE_DELAYED_HEAD:
			WARN_ON(1);
			continue;
		case BTRFS_ADD_DELAYED_REF:
			sgn = 1;
			break;
		case BTRFS_DROP_DELAYED_REF:
			sgn = -1;
			break;
		default:
			BUG_ON(1);
		}
		switch (node->type) {
		case BTRFS_TREE_BLOCK_REF_KEY: {
			struct btrfs_delayed_tree_ref *ref;

			ref = btrfs_delayed_node_to_tree_ref(node);
			ret = __add_prelim_ref(prefs, ref->root, &op_key,
					       ref->level + 1, 0, node->bytenr,
					       node->ref_mod * sgn);
			break;
		}
		case BTRFS_SHARED_BLOCK_REF_KEY: {
			struct btrfs_delayed_tree_ref *ref;

			ref = btrfs_delayed_node_to_tree_ref(node);
			ret = __add_prelim_ref(prefs, ref->root, NULL,
					       ref->level + 1, ref->parent,
					       node->bytenr,
					       node->ref_mod * sgn);
			break;
		}
		case BTRFS_EXTENT_DATA_REF_KEY: {
			struct btrfs_delayed_data_ref *ref;
			ref = btrfs_delayed_node_to_data_ref(node);

			key.objectid = ref->objectid;
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = ref->offset;
			ret = __add_prelim_ref(prefs, ref->root, &key, 0, 0,
					       node->bytenr,
					       node->ref_mod * sgn);
			break;
		}
		case BTRFS_SHARED_DATA_REF_KEY: {
			struct btrfs_delayed_data_ref *ref;

			ref = btrfs_delayed_node_to_data_ref(node);

			key.objectid = ref->objectid;
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = ref->offset;
			ret = __add_prelim_ref(prefs, ref->root, &key, 0,
					       ref->parent, node->bytenr,
					       node->ref_mod * sgn);
			break;
		}
		default:
			WARN_ON(1);
		}
		if (ret)
			return ret;
	}

	return 0;
}

/*
 * add all inline backrefs for bytenr to the list
 */
static int __add_inline_refs(struct btrfs_fs_info *fs_info,
			     struct btrfs_path *path, u64 bytenr,
			     int *info_level, struct list_head *prefs)
{
	int ret = 0;
	int slot;
	struct extent_buffer *leaf;
	struct btrfs_key key;
	struct btrfs_key found_key;
	unsigned long ptr;
	unsigned long end;
	struct btrfs_extent_item *ei;
	u64 flags;
	u64 item_size;

	/*
	 * enumerate all inline refs
	 */
	leaf = path->nodes[0];
	slot = path->slots[0];

	item_size = btrfs_item_size_nr(leaf, slot);
	BUG_ON(item_size < sizeof(*ei));

	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
	flags = btrfs_extent_flags(leaf, ei);
	btrfs_item_key_to_cpu(leaf, &found_key, slot);

	ptr = (unsigned long)(ei + 1);
	end = (unsigned long)ei + item_size;

	if (found_key.type == BTRFS_EXTENT_ITEM_KEY &&
	    flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
		struct btrfs_tree_block_info *info;

		info = (struct btrfs_tree_block_info *)ptr;
		*info_level = btrfs_tree_block_level(leaf, info);
		ptr += sizeof(struct btrfs_tree_block_info);
		BUG_ON(ptr > end);
	} else if (found_key.type == BTRFS_METADATA_ITEM_KEY) {
		*info_level = found_key.offset;
	} else {
		BUG_ON(!(flags & BTRFS_EXTENT_FLAG_DATA));
	}

	while (ptr < end) {
		struct btrfs_extent_inline_ref *iref;
		u64 offset;
		int type;

		iref = (struct btrfs_extent_inline_ref *)ptr;
		type = btrfs_extent_inline_ref_type(leaf, iref);
		offset = btrfs_extent_inline_ref_offset(leaf, iref);

		switch (type) {
		case BTRFS_SHARED_BLOCK_REF_KEY:
			ret = __add_prelim_ref(prefs, 0, NULL,
						*info_level + 1, offset,
						bytenr, 1);
			break;
		case BTRFS_SHARED_DATA_REF_KEY: {
			struct btrfs_shared_data_ref *sdref;
			int count;

			sdref = (struct btrfs_shared_data_ref *)(iref + 1);
			count = btrfs_shared_data_ref_count(leaf, sdref);
			ret = __add_prelim_ref(prefs, 0, NULL, 0, offset,
					       bytenr, count);
			break;
		}
		case BTRFS_TREE_BLOCK_REF_KEY:
			ret = __add_prelim_ref(prefs, offset, NULL,
					       *info_level + 1, 0,
					       bytenr, 1);
			break;
		case BTRFS_EXTENT_DATA_REF_KEY: {
			struct btrfs_extent_data_ref *dref;
			int count;
			u64 root;

			dref = (struct btrfs_extent_data_ref *)(&iref->offset);
			count = btrfs_extent_data_ref_count(leaf, dref);
			key.objectid = btrfs_extent_data_ref_objectid(leaf,
								      dref);
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = btrfs_extent_data_ref_offset(leaf, dref);
			root = btrfs_extent_data_ref_root(leaf, dref);
			ret = __add_prelim_ref(prefs, root, &key, 0, 0,
					       bytenr, count);
			break;
		}
		default:
			WARN_ON(1);
		}
		if (ret)
			return ret;
		ptr += btrfs_extent_inline_ref_size(type);
	}

	return 0;
}

/*
 * add all non-inline backrefs for bytenr to the list
 */
static int __add_keyed_refs(struct btrfs_fs_info *fs_info,
			    struct btrfs_path *path, u64 bytenr,
			    int info_level, struct list_head *prefs)
{
	struct btrfs_root *extent_root = fs_info->extent_root;
	int ret;
	int slot;
	struct extent_buffer *leaf;
	struct btrfs_key key;

	while (1) {
		ret = btrfs_next_item(extent_root, path);
		if (ret < 0)
			break;
		if (ret) {
			ret = 0;
			break;
		}

		slot = path->slots[0];
		leaf = path->nodes[0];
		btrfs_item_key_to_cpu(leaf, &key, slot);

		if (key.objectid != bytenr)
			break;
		if (key.type < BTRFS_TREE_BLOCK_REF_KEY)
			continue;
		if (key.type > BTRFS_SHARED_DATA_REF_KEY)
			break;

		switch (key.type) {
		case BTRFS_SHARED_BLOCK_REF_KEY:
			ret = __add_prelim_ref(prefs, 0, NULL,
						info_level + 1, key.offset,
						bytenr, 1);
			break;
		case BTRFS_SHARED_DATA_REF_KEY: {
			struct btrfs_shared_data_ref *sdref;
			int count;

			sdref = btrfs_item_ptr(leaf, slot,
					      struct btrfs_shared_data_ref);
			count = btrfs_shared_data_ref_count(leaf, sdref);
			ret = __add_prelim_ref(prefs, 0, NULL, 0, key.offset,
						bytenr, count);
			break;
		}
		case BTRFS_TREE_BLOCK_REF_KEY:
			ret = __add_prelim_ref(prefs, key.offset, NULL,
					       info_level + 1, 0,
					       bytenr, 1);
			break;
		case BTRFS_EXTENT_DATA_REF_KEY: {
			struct btrfs_extent_data_ref *dref;
			int count;
			u64 root;

			dref = btrfs_item_ptr(leaf, slot,
					      struct btrfs_extent_data_ref);
			count = btrfs_extent_data_ref_count(leaf, dref);
			key.objectid = btrfs_extent_data_ref_objectid(leaf,
								      dref);
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = btrfs_extent_data_ref_offset(leaf, dref);
			root = btrfs_extent_data_ref_root(leaf, dref);
			ret = __add_prelim_ref(prefs, root, &key, 0, 0,
					       bytenr, count);
			break;
		}
		default:
			WARN_ON(1);
		}
		if (ret)
			return ret;

	}

	return ret;
}

/*
 * this adds all existing backrefs (inline backrefs, backrefs and delayed
 * refs) for the given bytenr to the refs list, merges duplicates and resolves
 * indirect refs to their parent bytenr.
 * When roots are found, they're added to the roots list
 *
 * FIXME some caching might speed things up
 */
static int find_parent_nodes(struct btrfs_trans_handle *trans,
			     struct btrfs_fs_info *fs_info, u64 bytenr,
			     u64 time_seq, struct ulist *refs,
			     struct ulist *roots, const u64 *extent_item_pos)
{
	struct btrfs_key key;
	struct btrfs_path *path;
	struct btrfs_delayed_ref_root *delayed_refs = NULL;
	struct btrfs_delayed_ref_head *head;
	int info_level = 0;
	int ret;
	struct list_head prefs_delayed;
	struct list_head prefs;
	struct __prelim_ref *ref;

	INIT_LIST_HEAD(&prefs);
	INIT_LIST_HEAD(&prefs_delayed);

	key.objectid = bytenr;
	key.offset = (u64)-1;
	if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
		key.type = BTRFS_METADATA_ITEM_KEY;
	else
		key.type = BTRFS_EXTENT_ITEM_KEY;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;
	if (!trans)
		path->search_commit_root = 1;

	/*
	 * grab both a lock on the path and a lock on the delayed ref head.
	 * We need both to get a consistent picture of how the refs look
	 * at a specified point in time
	 */
again:
	head = NULL;

	ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 0);
	if (ret < 0)
		goto out;
	BUG_ON(ret == 0);

	if (trans) {
		/*
		 * look if there are updates for this ref queued and lock the
		 * head
		 */
		delayed_refs = &trans->transaction->delayed_refs;
		spin_lock(&delayed_refs->lock);
		head = btrfs_find_delayed_ref_head(trans, bytenr);
		if (head) {
			if (!mutex_trylock(&head->mutex)) {
				atomic_inc(&head->node.refs);
				spin_unlock(&delayed_refs->lock);

				btrfs_release_path(path);

				/*
				 * Mutex was contended, block until it's
				 * released and try again
				 */
				mutex_lock(&head->mutex);
				mutex_unlock(&head->mutex);
				btrfs_put_delayed_ref(&head->node);
				goto again;
			}
			ret = __add_delayed_refs(head, time_seq,
						 &prefs_delayed);
			mutex_unlock(&head->mutex);
			if (ret) {
				spin_unlock(&delayed_refs->lock);
				goto out;
			}
		}
		spin_unlock(&delayed_refs->lock);
	}

	if (path->slots[0]) {
		struct extent_buffer *leaf;
		int slot;

		path->slots[0]--;
		leaf = path->nodes[0];
		slot = path->slots[0];
		btrfs_item_key_to_cpu(leaf, &key, slot);
		if (key.objectid == bytenr &&
		    (key.type == BTRFS_EXTENT_ITEM_KEY ||
		     key.type == BTRFS_METADATA_ITEM_KEY)) {
			ret = __add_inline_refs(fs_info, path, bytenr,
						&info_level, &prefs);
			if (ret)
				goto out;
			ret = __add_keyed_refs(fs_info, path, bytenr,
					       info_level, &prefs);
			if (ret)
				goto out;
		}
	}
	btrfs_release_path(path);

	list_splice_init(&prefs_delayed, &prefs);

	ret = __add_missing_keys(fs_info, &prefs);
	if (ret)
		goto out;

	__merge_refs(&prefs, 1);

	ret = __resolve_indirect_refs(fs_info, path, time_seq, &prefs,
				      extent_item_pos);
	if (ret)
		goto out;

	__merge_refs(&prefs, 2);

	while (!list_empty(&prefs)) {
		ref = list_first_entry(&prefs, struct __prelim_ref, list);
		WARN_ON(ref->count < 0);
		if (ref->count && ref->root_id && ref->parent == 0) {
			/* no parent == root of tree */
			ret = ulist_add(roots, ref->root_id, 0, GFP_NOFS);
			if (ret < 0)
				goto out;
		}
		if (ref->count && ref->parent) {
			struct extent_inode_elem *eie = NULL;
			if (extent_item_pos && !ref->inode_list) {
				u32 bsz;
				struct extent_buffer *eb;
				bsz = btrfs_level_size(fs_info->extent_root,
							info_level);
				eb = read_tree_block(fs_info->extent_root,
							   ref->parent, bsz, 0);
				if (!eb || !extent_buffer_uptodate(eb)) {
					free_extent_buffer(eb);
					ret = -EIO;
					goto out;
				}
				ret = find_extent_in_eb(eb, bytenr,
							*extent_item_pos, &eie);
				free_extent_buffer(eb);
				if (ret < 0)
					goto out;
				ref->inode_list = eie;
			}
			ret = ulist_add_merge(refs, ref->parent,
					      (uintptr_t)ref->inode_list,
					      (u64 *)&eie, GFP_NOFS);
			if (ret < 0)
				goto out;
			if (!ret && extent_item_pos) {
				/*
				 * we've recorded that parent, so we must extend
				 * its inode list here
				 */
				BUG_ON(!eie);
				while (eie->next)
					eie = eie->next;
				eie->next = ref->inode_list;
			}
		}
		list_del(&ref->list);
		kfree(ref);
	}

out:
	btrfs_free_path(path);
	while (!list_empty(&prefs)) {
		ref = list_first_entry(&prefs, struct __prelim_ref, list);
		list_del(&ref->list);
		kfree(ref);
	}
	while (!list_empty(&prefs_delayed)) {
		ref = list_first_entry(&prefs_delayed, struct __prelim_ref,
				       list);
		list_del(&ref->list);
		kfree(ref);
	}

	return ret;
}

static void free_leaf_list(struct ulist *blocks)
{
	struct ulist_node *node = NULL;
	struct extent_inode_elem *eie;
	struct extent_inode_elem *eie_next;
	struct ulist_iterator uiter;

	ULIST_ITER_INIT(&uiter);
	while ((node = ulist_next(blocks, &uiter))) {
		if (!node->aux)
			continue;
		eie = (struct extent_inode_elem *)(uintptr_t)node->aux;
		for (; eie; eie = eie_next) {
			eie_next = eie->next;
			kfree(eie);
		}
		node->aux = 0;
	}

	ulist_free(blocks);
}

/*
 * Finds all leafs with a reference to the specified combination of bytenr and
 * offset. key_list_head will point to a list of corresponding keys (caller must
 * free each list element). The leafs will be stored in the leafs ulist, which
 * must be freed with ulist_free.
 *
 * returns 0 on success, <0 on error
 */
static int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
				struct btrfs_fs_info *fs_info, u64 bytenr,
				u64 time_seq, struct ulist **leafs,
				const u64 *extent_item_pos)
{
	struct ulist *tmp;
	int ret;

	tmp = ulist_alloc(GFP_NOFS);
	if (!tmp)
		return -ENOMEM;
	*leafs = ulist_alloc(GFP_NOFS);
	if (!*leafs) {
		ulist_free(tmp);
		return -ENOMEM;
	}

	ret = find_parent_nodes(trans, fs_info, bytenr,
				time_seq, *leafs, tmp, extent_item_pos);
	ulist_free(tmp);

	if (ret < 0 && ret != -ENOENT) {
		free_leaf_list(*leafs);
		return ret;
	}

	return 0;
}

/*
 * walk all backrefs for a given extent to find all roots that reference this
 * extent. Walking a backref means finding all extents that reference this
 * extent and in turn walk the backrefs of those, too. Naturally this is a
 * recursive process, but here it is implemented in an iterative fashion: We
 * find all referencing extents for the extent in question and put them on a
 * list. In turn, we find all referencing extents for those, further appending
 * to the list. The way we iterate the list allows adding more elements after
 * the current while iterating. The process stops when we reach the end of the
 * list. Found roots are added to the roots list.
 *
 * returns 0 on success, < 0 on error.
 */
int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
				struct btrfs_fs_info *fs_info, u64 bytenr,
				u64 time_seq, struct ulist **roots)
{
	struct ulist *tmp;
	struct ulist_node *node = NULL;
	struct ulist_iterator uiter;
	int ret;

	tmp = ulist_alloc(GFP_NOFS);
	if (!tmp)
		return -ENOMEM;
	*roots = ulist_alloc(GFP_NOFS);
	if (!*roots) {
		ulist_free(tmp);
		return -ENOMEM;
	}

	ULIST_ITER_INIT(&uiter);
	while (1) {
		ret = find_parent_nodes(trans, fs_info, bytenr,
					time_seq, tmp, *roots, NULL);
		if (ret < 0 && ret != -ENOENT) {
			ulist_free(tmp);
			ulist_free(*roots);
			return ret;
		}
		node = ulist_next(tmp, &uiter);
		if (!node)
			break;
		bytenr = node->val;
	}

	ulist_free(tmp);
	return 0;
}


static int __inode_info(u64 inum, u64 ioff, u8 key_type,
			struct btrfs_root *fs_root, struct btrfs_path *path,
			struct btrfs_key *found_key)
{
	int ret;
	struct btrfs_key key;
	struct extent_buffer *eb;

	key.type = key_type;
	key.objectid = inum;
	key.offset = ioff;

	ret = btrfs_search_slot(NULL, fs_root, &key, path, 0, 0);
	if (ret < 0)
		return ret;

	eb = path->nodes[0];
	if (ret && path->slots[0] >= btrfs_header_nritems(eb)) {
		ret = btrfs_next_leaf(fs_root, path);
		if (ret)
			return ret;
		eb = path->nodes[0];
	}

	btrfs_item_key_to_cpu(eb, found_key, path->slots[0]);
	if (found_key->type != key.type || found_key->objectid != key.objectid)
		return 1;

	return 0;
}

/*
 * this makes the path point to (inum INODE_ITEM ioff)
 */
int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
			struct btrfs_path *path)
{
	struct btrfs_key key;
	return __inode_info(inum, ioff, BTRFS_INODE_ITEM_KEY, fs_root, path,
				&key);
}

static int inode_ref_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
				struct btrfs_path *path,
				struct btrfs_key *found_key)
{
	return __inode_info(inum, ioff, BTRFS_INODE_REF_KEY, fs_root, path,
				found_key);
}

int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
			  u64 start_off, struct btrfs_path *path,
			  struct btrfs_inode_extref **ret_extref,
			  u64 *found_off)
{
	int ret, slot;
	struct btrfs_key key;
	struct btrfs_key found_key;
	struct btrfs_inode_extref *extref;
	struct extent_buffer *leaf;
	unsigned long ptr;

	key.objectid = inode_objectid;
	btrfs_set_key_type(&key, BTRFS_INODE_EXTREF_KEY);
	key.offset = start_off;

	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
	if (ret < 0)
		return ret;

	while (1) {
		leaf = path->nodes[0];
		slot = path->slots[0];
		if (slot >= btrfs_header_nritems(leaf)) {
			/*
			 * If the item at offset is not found,
			 * btrfs_search_slot will point us to the slot
			 * where it should be inserted. In our case
			 * that will be the slot directly before the
			 * next INODE_REF_KEY_V2 item. In the case
			 * that we're pointing to the last slot in a
			 * leaf, we must move one leaf over.
			 */
			ret = btrfs_next_leaf(root, path);
			if (ret) {
				if (ret >= 1)
					ret = -ENOENT;
				break;
			}
			continue;
		}

		btrfs_item_key_to_cpu(leaf, &found_key, slot);

		/*
		 * Check that we're still looking at an extended ref key for
		 * this particular objectid. If we have different
		 * objectid or type then there are no more to be found
		 * in the tree and we can exit.
		 */
		ret = -ENOENT;
		if (found_key.objectid != inode_objectid)
			break;
		if (btrfs_key_type(&found_key) != BTRFS_INODE_EXTREF_KEY)
			break;

		ret = 0;
		ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
		extref = (struct btrfs_inode_extref *)ptr;
		*ret_extref = extref;
		if (found_off)
			*found_off = found_key.offset;
		break;
	}

	return ret;
}

/*
 * this iterates to turn a name (from iref/extref) into a full filesystem path.
 * Elements of the path are separated by '/' and the path is guaranteed to be
 * 0-terminated. the path is only given within the current file system.
 * Therefore, it never starts with a '/'. the caller is responsible to provide
 * "size" bytes in "dest". the dest buffer will be filled backwards. finally,
 * the start point of the resulting string is returned. this pointer is within
 * dest, normally.
 * in case the path buffer would overflow, the pointer is decremented further
 * as if output was written to the buffer, though no more output is actually
 * generated. that way, the caller can determine how much space would be
 * required for the path to fit into the buffer. in that case, the returned
 * value will be smaller than dest. callers must check this!
 */
char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
			u32 name_len, unsigned long name_off,
			struct extent_buffer *eb_in, u64 parent,
			char *dest, u32 size)
{
	int slot;
	u64 next_inum;
	int ret;
	s64 bytes_left = ((s64)size) - 1;
	struct extent_buffer *eb = eb_in;
	struct btrfs_key found_key;
	int leave_spinning = path->leave_spinning;
	struct btrfs_inode_ref *iref;

	if (bytes_left >= 0)
		dest[bytes_left] = '\0';

	path->leave_spinning = 1;
	while (1) {
		bytes_left -= name_len;
		if (bytes_left >= 0)
			read_extent_buffer(eb, dest + bytes_left,
					   name_off, name_len);
		if (eb != eb_in) {
			btrfs_tree_read_unlock_blocking(eb);
			free_extent_buffer(eb);
		}
		ret = inode_ref_info(parent, 0, fs_root, path, &found_key);
		if (ret > 0)
			ret = -ENOENT;
		if (ret)
			break;

		next_inum = found_key.offset;

		/* regular exit ahead */
		if (parent == next_inum)
			break;

		slot = path->slots[0];
		eb = path->nodes[0];
		/* make sure we can use eb after releasing the path */
		if (eb != eb_in) {
			atomic_inc(&eb->refs);
			btrfs_tree_read_lock(eb);
			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
		}
		btrfs_release_path(path);
		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);

		name_len = btrfs_inode_ref_name_len(eb, iref);
		name_off = (unsigned long)(iref + 1);

		parent = next_inum;
		--bytes_left;
		if (bytes_left >= 0)
			dest[bytes_left] = '/';
	}

	btrfs_release_path(path);
	path->leave_spinning = leave_spinning;

	if (ret)
		return ERR_PTR(ret);

	return dest + bytes_left;
}

/*
 * this makes the path point to (logical EXTENT_ITEM *)
 * returns BTRFS_EXTENT_FLAG_DATA for data, BTRFS_EXTENT_FLAG_TREE_BLOCK for
 * tree blocks and <0 on error.
 */
int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
			struct btrfs_path *path, struct btrfs_key *found_key,
			u64 *flags_ret)
{
	int ret;
	u64 flags;
	u64 size = 0;
	u32 item_size;
	struct extent_buffer *eb;
	struct btrfs_extent_item *ei;
	struct btrfs_key key;

	if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
		key.type = BTRFS_METADATA_ITEM_KEY;
	else
		key.type = BTRFS_EXTENT_ITEM_KEY;
	key.objectid = logical;
	key.offset = (u64)-1;

	ret = btrfs_search_slot(NULL, fs_info->extent_root, &key, path, 0, 0);
	if (ret < 0)
		return ret;
	ret = btrfs_previous_item(fs_info->extent_root, path,
					0, BTRFS_EXTENT_ITEM_KEY);
	if (ret < 0)
		return ret;

	btrfs_item_key_to_cpu(path->nodes[0], found_key, path->slots[0]);
	if (found_key->type == BTRFS_METADATA_ITEM_KEY)
		size = fs_info->extent_root->leafsize;
	else if (found_key->type == BTRFS_EXTENT_ITEM_KEY)
		size = found_key->offset;

	if ((found_key->type != BTRFS_EXTENT_ITEM_KEY &&
	     found_key->type != BTRFS_METADATA_ITEM_KEY) ||
	    found_key->objectid > logical ||
	    found_key->objectid + size <= logical) {
		pr_debug("logical %llu is not within any extent\n", logical);
		return -ENOENT;
	}

	eb = path->nodes[0];
	item_size = btrfs_item_size_nr(eb, path->slots[0]);
	BUG_ON(item_size < sizeof(*ei));

	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
	flags = btrfs_extent_flags(eb, ei);

	pr_debug("logical %llu is at position %llu within the extent (%llu "
		 "EXTENT_ITEM %llu) flags %#llx size %u\n",
		 logical, logical - found_key->objectid, found_key->objectid,
		 found_key->offset, flags, item_size);

	WARN_ON(!flags_ret);
	if (flags_ret) {
		if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)
			*flags_ret = BTRFS_EXTENT_FLAG_TREE_BLOCK;
		else if (flags & BTRFS_EXTENT_FLAG_DATA)
			*flags_ret = BTRFS_EXTENT_FLAG_DATA;
		else
			BUG_ON(1);
		return 0;
	}

	return -EIO;
}

/*
 * helper function to iterate extent inline refs. ptr must point to a 0 value
 * for the first call and may be modified. it is used to track state.
 * if more refs exist, 0 is returned and the next call to
 * __get_extent_inline_ref must pass the modified ptr parameter to get the
 * next ref. after the last ref was processed, 1 is returned.
 * returns <0 on error
 */
static int __get_extent_inline_ref(unsigned long *ptr, struct extent_buffer *eb,
				struct btrfs_extent_item *ei, u32 item_size,
				struct btrfs_extent_inline_ref **out_eiref,
				int *out_type)
{
	unsigned long end;
	u64 flags;
	struct btrfs_tree_block_info *info;

	if (!*ptr) {
		/* first call */
		flags = btrfs_extent_flags(eb, ei);
		if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
			info = (struct btrfs_tree_block_info *)(ei + 1);
			*out_eiref =
				(struct btrfs_extent_inline_ref *)(info + 1);
		} else {
			*out_eiref = (struct btrfs_extent_inline_ref *)(ei + 1);
		}
		*ptr = (unsigned long)*out_eiref;
		if ((void *)*ptr >= (void *)ei + item_size)
			return -ENOENT;
	}

	end = (unsigned long)ei + item_size;
	*out_eiref = (struct btrfs_extent_inline_ref *)*ptr;
	*out_type = btrfs_extent_inline_ref_type(eb, *out_eiref);

	*ptr += btrfs_extent_inline_ref_size(*out_type);
	WARN_ON(*ptr > end);
	if (*ptr == end)
		return 1; /* last */

	return 0;
}

/*
 * reads the tree block backref for an extent. tree level and root are returned
 * through out_level and out_root. ptr must point to a 0 value for the first
 * call and may be modified (see __get_extent_inline_ref comment).
 * returns 0 if data was provided, 1 if there was no more data to provide or
 * <0 on error.
 */
int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb,
				struct btrfs_extent_item *ei, u32 item_size,
				u64 *out_root, u8 *out_level)
{
	int ret;
	int type;
	struct btrfs_tree_block_info *info;
	struct btrfs_extent_inline_ref *eiref;

	if (*ptr == (unsigned long)-1)
		return 1;

	while (1) {
		ret = __get_extent_inline_ref(ptr, eb, ei, item_size,
						&eiref, &type);
		if (ret < 0)
			return ret;

		if (type == BTRFS_TREE_BLOCK_REF_KEY ||
		    type == BTRFS_SHARED_BLOCK_REF_KEY)
			break;

		if (ret == 1)
			return 1;
	}

	/* we can treat both ref types equally here */
	info = (struct btrfs_tree_block_info *)(ei + 1);
	*out_root = btrfs_extent_inline_ref_offset(eb, eiref);
	*out_level = btrfs_tree_block_level(eb, info);

	if (ret == 1)
		*ptr = (unsigned long)-1;

	return 0;
}

static int iterate_leaf_refs(struct extent_inode_elem *inode_list,
				u64 root, u64 extent_item_objectid,
				iterate_extent_inodes_t *iterate, void *ctx)
{
	struct extent_inode_elem *eie;
	int ret = 0;

	for (eie = inode_list; eie; eie = eie->next) {
		pr_debug("ref for %llu resolved, key (%llu EXTEND_DATA %llu), "
			 "root %llu\n", extent_item_objectid,
			 eie->inum, eie->offset, root);
		ret = iterate(eie->inum, eie->offset, root, ctx);
		if (ret) {
			pr_debug("stopping iteration for %llu due to ret=%d\n",
				 extent_item_objectid, ret);
			break;
		}
	}

	return ret;
}

/*
 * calls iterate() for every inode that references the extent identified by
 * the given parameters.
 * when the iterator function returns a non-zero value, iteration stops.
 */
int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
				u64 extent_item_objectid, u64 extent_item_pos,
				int search_commit_root,