summaryrefslogtreecommitdiffstats
path: root/fs/ext3/namei.c
blob: 3e5edc92aa0b088071d1711c4bcf620d585e0416 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
// SPDX-License-Identifier: GPL-2.0-or-later
/*
 *  sata_sil.c - Silicon Image SATA
 *
 *  Maintained by:  Tejun Heo <tj@kernel.org>
 *  		    Please ALWAYS copy linux-ide@vger.kernel.org
 *		    on emails.
 *
 *  Copyright 2003-2005 Red Hat, Inc.
 *  Copyright 2003 Benjamin Herrenschmidt
 *
 *  libata documentation is available via 'make {ps|pdf}docs',
 *  as Documentation/driver-api/libata.rst
 *
 *  Documentation for SiI 3112:
 *  http://gkernel.sourceforge.net/specs/sii/3112A_SiI-DS-0095-B2.pdf.bz2
 *
 *  Other errata and documentation available under NDA.
 */

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/blkdev.h>
#include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/device.h>
#include <scsi/scsi_host.h>
#include <linux/libata.h>
#include <linux/dmi.h>

#define DRV_NAME	"sata_sil"
#define DRV_VERSION	"2.4"

#define SIL_DMA_BOUNDARY	0x7fffffffUL

enum {
	SIL_MMIO_BAR		= 5,

	/*
	 * host flags
	 */
	SIL_FLAG_NO_SATA_IRQ	= (1 << 28),
	SIL_FLAG_RERR_ON_DMA_ACT = (1 << 29),
	SIL_FLAG_MOD15WRITE	= (1 << 30),

	SIL_DFL_PORT_FLAGS	= ATA_FLAG_SATA,

	/*
	 * Controller IDs
	 */
	sil_3112		= 0,
	sil_3112_no_sata_irq	= 1,
	sil_3512		= 2,
	sil_3114		= 3,

	/*
	 * Register offsets
	 */
	SIL_SYSCFG		= 0x48,

	/*
	 * Register bits
	 */
	/* SYSCFG */
	SIL_MASK_IDE0_INT	= (1 << 22),
	SIL_MASK_IDE1_INT	= (1 << 23),
	SIL_MASK_IDE2_INT	= (1 << 24),
	SIL_MASK_IDE3_INT	= (1 << 25),
	SIL_MASK_2PORT		= SIL_MASK_IDE0_INT | SIL_MASK_IDE1_INT,
	SIL_MASK_4PORT		= SIL_MASK_2PORT |
				  SIL_MASK_IDE2_INT | SIL_MASK_IDE3_INT,

	/* BMDMA/BMDMA2 */
	SIL_INTR_STEERING	= (1 << 1),

	SIL_DMA_ENABLE		= (1 << 0),  /* DMA run switch */
	SIL_DMA_RDWR		= (1 << 3),  /* DMA Rd-Wr */
	SIL_DMA_SATA_IRQ	= (1 << 4),  /* OR of all SATA IRQs */
	SIL_DMA_ACTIVE		= (1 << 16), /* DMA running */
	SIL_DMA_ERROR		= (1 << 17), /* PCI bus error */
	SIL_DMA_COMPLETE	= (1 << 18), /* cmd complete / IRQ pending */
	SIL_DMA_N_SATA_IRQ	= (1 << 6),  /* SATA_IRQ for the next channel */
	SIL_DMA_N_ACTIVE	= (1 << 24), /* ACTIVE for the next channel */
	SIL_DMA_N_ERROR		= (1 << 25), /* ERROR for the next channel */
	SIL_DMA_N_COMPLETE	= (1 << 26), /* COMPLETE for the next channel */

	/* SIEN */
	SIL_SIEN_N		= (1 << 16), /* triggered by SError.N */

	/*
	 * Others
	 */
	SIL_QUIRK_MOD15WRITE	= (1 << 0),
	SIL_QUIRK_UDMA5MAX	= (1 << 1),
};

static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
#ifdef CONFIG_PM_SLEEP
static int sil_pci_device_resume(struct pci_dev *pdev);
#endif
static void sil_dev_config(struct ata_device *dev);
static int sil_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
static int sil_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
static int sil_set_mode(struct ata_link *link, struct ata_device **r_failed);
static void sil_qc_prep(struct ata_queued_cmd *qc);
static void sil_bmdma_setup(struct ata_queued_cmd *qc);
static void sil_bmdma_start(struct ata_queued_cmd *qc);
static void sil_bmdma_stop(struct ata_queued_cmd *qc);
static void sil_freeze(struct ata_port *ap);
static void sil_thaw(struct ata_port *ap);


static const struct pci_device_id sil_pci_tbl[] = {
	{ PCI_VDEVICE(CMD, 0x3112), sil_3112 },
	{ PCI_VDEVICE(CMD, 0x0240), sil_3112 },
	{ PCI_VDEVICE(CMD, 0x3512), sil_3512 },
	{ PCI_VDEVICE(CMD, 0x3114), sil_3114 },
	{ PCI_VDEVICE(ATI, 0x436e), sil_3112 },
	{ PCI_VDEVICE(ATI, 0x4379), sil_3112_no_sata_irq },
	{ PCI_VDEVICE(ATI, 0x437a), sil_3112_no_sata_irq },

	{ }	/* terminate list */
};


/* TODO firmware versions should be added - eric */
static const struct sil_drivelist {
	const char *product;
	unsigned int quirk;
} sil_blacklist [] = {
	{ "ST320012AS",		SIL_QUIRK_MOD15WRITE },
	{ "ST330013AS",		SIL_QUIRK_MOD15WRITE },
	{ "ST340017AS",		SIL_QUIRK_MOD15WRITE },
	{ "ST360015AS",		SIL_QUIRK_MOD15WRITE },
	{ "ST380023AS",		SIL_QUIRK_MOD15WRITE },
	{ "ST3120023AS",	SIL_QUIRK_MOD15WRITE },
	{ "ST340014ASL",	SIL_QUIRK_MOD15WRITE },
	{ "ST360014ASL",	SIL_QUIRK_MOD15WRITE },
	{ "ST380011ASL",	SIL_QUIRK_MOD15WRITE },
	{ "ST3120022ASL",	SIL_QUIRK_MOD15WRITE },
	{ "ST3160021ASL",	SIL_QUIRK_MOD15WRITE },
	{ "TOSHIBA MK2561GSYN",	SIL_QUIRK_MOD15WRITE },
	{ "Maxtor 4D060H3",	SIL_QUIRK_UDMA5MAX },
	{ }
};

static struct pci_driver sil_pci_driver = {
	.name			= DRV_NAME,
	.id_table		= sil_pci_tbl,
	.probe			= sil_init_one,
	.remove			= ata_pci_remove_one,
#ifdef CONFIG_PM_SLEEP
	.suspend		= ata_pci_device_suspend,
	.resume			= sil_pci_device_resume,
#endif
};

static struct scsi_host_template sil_sht = {
	ATA_BASE_SHT(DRV_NAME),
	/** These controllers support Large Block Transfer which allows
	    transfer chunks up to 2GB and which cross 64KB boundaries,
	    therefore the DMA limits are more relaxed than standard ATA SFF. */
	.dma_boundary		= SIL_DMA_BOUNDARY,
	.sg_tablesize		= ATA_MAX_PRD
};

static struct ata_port_operations sil_ops = {
	.inherits		= &ata_bmdma32_port_ops,
	.dev_config		= sil_dev_config,
	.set_mode		= sil_set_mode,
	.bmdma_setup            = sil_bmdma_setup,
	.bmdma_start            = sil_bmdma_start,
	.bmdma_stop		= sil_bmdma_stop,
	.qc_prep		= sil_qc_prep,
	.freeze			= sil_freeze,
	.thaw			= sil_thaw,
	.scr_read		= sil_scr_read,
	.scr_write		= sil_scr_write,
};

static const struct ata_port_info sil_port_info[] = {
	/* sil_3112 */
	{
		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE,
		.pio_mask	= ATA_PIO4,
		.mwdma_mask	= ATA_MWDMA2,
		.udma_mask	= ATA_UDMA5,
		.port_ops	= &sil_ops,
	},
	/* sil_3112_no_sata_irq */
	{
		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_MOD15WRITE |
				  SIL_FLAG_NO_SATA_IRQ,
		.pio_mask	= ATA_PIO4,
		.mwdma_mask	= ATA_MWDMA2,
		.udma_mask	= ATA_UDMA5,
		.port_ops	= &sil_ops,
	},
	/* sil_3512 */
	{
		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
		.pio_mask	= ATA_PIO4,
		.mwdma_mask	= ATA_MWDMA2,
		.udma_mask	= ATA_UDMA5,
		.port_ops	= &sil_ops,
	},
	/* sil_3114 */
	{
		.flags		= SIL_DFL_PORT_FLAGS | SIL_FLAG_RERR_ON_DMA_ACT,
		.pio_mask	= ATA_PIO4,
		.mwdma_mask	= ATA_MWDMA2,
		.udma_mask	= ATA_UDMA5,
		.port_ops	= &sil_ops,
	},
};

/* per-port register offsets */
/* TODO: we can probably calculate rather than use a table */
static const struct {
	unsigned long tf;	/* ATA taskfile register block */
	unsigned long ctl;	/* ATA control/altstatus register block */
	unsigned long bmdma;	/* DMA register block */
	unsigned long bmdma2;	/* DMA register block #2 */
	unsigned long fifo_cfg;	/* FIFO Valid Byte Count and Control */
	unsigned long scr;	/* SATA control register block */
	unsigned long sien;	/* SATA Interrupt Enable register */
	unsigned long xfer_mode;/* data transfer mode register */
	unsigned long sfis_cfg;	/* SATA FIS reception config register */
} sil_port[] = {
	/* port 0 ... */
	/*   tf    ctl  bmdma  bmdma2  fifo    scr   sien   mode   sfis */
	{  0x80,  0x8A,   0x0,  0x10,  0x40, 0x100, 0x148,  0xb4, 0x14c },
	{  0xC0,  0xCA,   0x8,  0x18,  0x44, 0x180, 0x1c8,  0xf4, 0x1cc },
	{ 0x280, 0x28A, 0x200, 0x210, 0x240, 0x300, 0x348, 0x2b4, 0x34c },
	{ 0x2C0, 0x2CA, 0x208, 0x218, 0x244, 0x380, 0x3c8, 0x2f4, 0x3cc },
	/* ... port 3 */
};

MODULE_AUTHOR("Jeff Garzik");
MODULE_DESCRIPTION("low-level driver for Silicon Image SATA controller");
MODULE_LICENSE("GPL");
MODULE_DEVICE_TABLE(pci, sil_pci_tbl);
MODULE_VERSION(DRV_VERSION);

static int slow_down;
module_param(slow_down, int, 0444);
MODULE_PARM_DESC(slow_down, "Sledgehammer used to work around random problems, by limiting commands to 15 sectors (0=off, 1=on)");


static void sil_bmdma_stop(struct ata_queued_cmd *qc)
{
	struct ata_port *ap = qc->ap;
	void __iomem *mmio_base = ap->host->iomap[SIL_MMIO_BAR];
	void __iomem *bmdma2 = mmio_base + sil_port[ap->port_no].bmdma2;

	/* clear start/stop bit - can safely always write 0 */
	iowrite8(0, bmdma2);

	/* one-PIO-cycle guaranteed wait, per spec, for HDMA1:0 transition */
	ata_sff_dma_pause(ap);
}

static void sil_bmdma_setup(struct ata_queued_cmd *qc)
{
	struct ata_port *ap = qc->ap;
	void __iomem *bmdma = ap->ioaddr.bmdma_addr;

	/* load PRD table addr. */
	iowrite32(ap->bmdma_prd_dma, bmdma + ATA_DMA_TABLE_OFS);

	/* issue r/w command */
	ap->ops->sff_exec_command(ap, &qc->tf);
}

static void sil_bmdma_start(struct ata_queued_cmd *qc)
{
	unsigned int rw = (qc->tf.flags & ATA_TFLAG_WRITE);
	struct ata_port *ap = qc->ap;
	void __iomem *mmio_base = ap->host->iomap[SIL_MMIO_BAR];
	void __iomem *bmdma2 = mmio_base + sil_port[ap->port_no].bmdma2;
	u8 dmactl = ATA_DMA_START;

	/* set transfer direction, start host DMA transaction
	   Note: For Large Block Transfer to work, the DMA must be started
	   using the bmdma2 register. */
	if (!rw)
		dmactl |= ATA_DMA_WR;
	iowrite8(dmactl, bmdma2);
}

/* The way God intended PCI IDE scatter/gather lists to look and behave... */
static void sil_fill_sg(struct ata_queued_cmd *qc)
{
	struct scatterlist *sg;
	struct ata_port *ap = qc->ap;
	struct ata_bmdma_prd *prd, *last_prd = NULL;
	unsigned int si;

	prd = &ap->bmdma_prd[0];
	for_each_sg(qc->sg, sg, qc->n_elem, si) {
		/* Note h/w doesn't support 64-bit, so we unconditionally
		 * truncate dma_addr_t to u32.
		 */
		u32 addr = (u32) sg_dma_address(sg);
		u32 sg_len = sg_dma_len(sg);

		prd->addr = cpu_to_le32(addr);
		prd->flags_len = cpu_to_le32(sg_len);
		VPRINTK("PRD[%u] = (0x%X, 0x%X)\n", si, addr, sg_len);

		last_prd = prd;
		prd++;
	}

	if (likely(last_prd))
		last_prd->flags_len |= cpu_to_le32(ATA_PRD_EOT);
}

static void sil_qc_prep(struct ata_queued_cmd *qc)
{
	if (!(qc->flags & ATA_QCFLAG_DMAMAP))
		return;

	sil_fill_sg(qc);
}

static unsigned char sil_get_device_cache_line(struct pci_dev *pdev)
{
	u8 cache_line = 0;
	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &cache_line);
	return cache_line;
}

/**
 *	sil_set_mode		-	wrap set_mode functions
 *	@link: link to set up
 *	@r_failed: returned device when we fail
 *
 *	Wrap the libata method for device setup as after the setup we need
 *	to inspect the results and do some configuration work
 */

static int sil_set_mode(struct ata_link *link, struct ata_device **r_failed)
{
	struct ata_port *ap = link->ap;
	void __iomem *mmio_base = ap->host->iomap[SIL_MMIO_BAR];
	void __iomem *addr = mmio_base + sil_port[ap->port_no].xfer_mode;
	struct ata_device *dev;
	u32 tmp, dev_mode[2] = { };
	int rc;

	rc = ata_do_set_mode(link, r_failed);
	if (rc)
		return rc;

	ata_for_each_dev(dev, link, ALL) {
		if (!ata_dev_enabled(dev))
			dev_mode[dev->devno] = 0;	/* PIO0/1/2 */
		else if (dev->flags & ATA_DFLAG_PIO)
			dev_mode[dev->devno] = 1;	/* PIO3/4 */
		else
			dev_mode[dev->devno] = 3;	/* UDMA */
		/* value 2 indicates MDMA */
	}

	tmp = readl(addr);
	tmp &= ~((1<<5) | (1<<4) | (1<<1) | (1<<0));
	tmp |= dev_mode[0];
	tmp |= (dev_mode[1] << 4);
	writel(tmp, addr);
	readl(addr);	/* flush */
	return 0;
}

static inline void __iomem *sil_scr_addr(struct ata_port *ap,
					 unsigned int sc_reg)
{
	void __iomem *offset = ap->ioaddr.scr_addr;

	switch (sc_reg) {
	case SCR_STATUS:
		return offset + 4;
	case SCR_ERROR:
		return offset + 8;
	case SCR_CONTROL:
		return offset;
	default:
		/* do nothing */
		break;
	}

	return NULL;
}

static int sil_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
{
	void __iomem *mmio = sil_scr_addr(link->ap, sc_reg);

	if (mmio) {
		*val = readl(mmio);
		return 0;
	}
	return -EINVAL;
}

static int sil_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
{
	void __iomem *mmio = sil_scr_addr(link->ap, sc_reg);

	if (mmio) {
		writel(val, mmio);
		return 0;
	}
	return -EINVAL;
}

static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
{
	struct ata_eh_info *ehi = &ap->link.eh_info;
	struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->link.active_tag);
	u8 status;

	if (unlikely(bmdma2 & SIL_DMA_SATA_IRQ)) {
		u32 serror = 0xffffffff;

		/* SIEN doesn't mask SATA IRQs on some 3112s.  Those
		 * controllers continue to assert IRQ as long as
		 * SError bits are pending.  Clear SError immediately.
		 */
		sil_scr_read(&ap->link, SCR_ERROR, &serror);
		sil_scr_write(&ap->link, SCR_ERROR, serror);

		/* Sometimes spurious interrupts occur, double check
		 * it's PHYRDY CHG.
		 */
		if (serror & SERR_PHYRDY_CHG) {
			ap->link.eh_info.serror |= serror;
			goto freeze;
		}

		if (!(bmdma2 & SIL_DMA_COMPLETE))
			return;
	}

	if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) {
		/* this sometimes happens, just clear IRQ */
		ap->ops->sff_check_status(ap);
		return;
	}

	/* Check whether we are expecting interrupt in this state */
	switch (ap->hsm_task_state) {
	case HSM_ST_FIRST:
		/* Some pre-ATAPI-4 devices assert INTRQ
		 * at this state when ready to receive CDB.
		 */

		/* Check the ATA_DFLAG_CDB_INTR flag is enough here.
		 * The flag was turned on only for atapi devices.  No
		 * need to check ata_is_atapi(qc->tf.protocol) again.
		 */
		if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR))
			goto err_hsm;
		break;
	case HSM_ST_LAST:
		if (ata_is_dma(qc->tf.protocol)) {
			/* clear DMA-Start bit */
			ap->ops->bmdma_stop(qc);

			if (bmdma2 & SIL_DMA_ERROR) {
				qc->err_mask |= AC_ERR_HOST_BUS;
				ap->hsm_task_state = HSM_ST_ERR;
			}
		}
		break;
	case HSM_ST:
		break;
	default:
		goto err_hsm;
	}

	/* check main status, clearing INTRQ */
	status = ap->ops->sff_check_status(ap);
	if (unlikely(status & ATA_BUSY))
		goto err_hsm;

	/* ack bmdma irq events */
	ata_bmdma_irq_clear(ap);

	/* kick HSM in the ass */
	ata_sff_hsm_move(ap, qc, status, 0);

	if (unlikely(qc->err_mask) && ata_is_dma(qc->tf.protocol))
		ata_ehi_push_desc(ehi, "BMDMA2 stat 0x%x", bmdma2);

	return;

 err_hsm:
	qc->err_mask |= AC_ERR_HSM;
 freeze:
	ata_port_freeze(ap);
}

static irqreturn_t sil_interrupt(int irq, void *dev_instance)
{
	struct ata_host *host = dev_instance;
	void __iomem *mmio_base = host->iomap[SIL_MMIO_BAR];
	int handled = 0;
	int i;

	spin_lock(&host->lock);

	for (i = 0; i < host->n_ports; i++) {
		struct ata_port *ap = host->ports[i];
		u32 bmdma2 = readl(mmio_base + sil_port[ap->port_no].bmdma2);

		/* turn off SATA_IRQ if not supported */
		if (ap->flags & SIL_FLAG_NO_SATA_IRQ)
			bmdma2 &= ~SIL_DMA_SATA_IRQ;

		if (bmdma2 == 0xffffffff ||
		    !(bmdma2 & (SIL_DMA_COMPLETE | SIL_DMA_SATA_IRQ)))
			continue;

		sil_host_intr(ap, bmdma2);
		handled = 1;
	}

	spin_unlock(&host->lock);

	return IRQ_RETVAL(handled);
}

static void sil_freeze(struct ata_port *ap)
{
	void __iomem *mmio_base = ap->host->iomap[SIL_MMIO_BAR];
	u32 tmp;

	/* global IRQ mask doesn't block SATA IRQ, turn off explicitly */
	writel(0, mmio_base + sil_port[ap->port_no].sien);

	/* plug IRQ */
	tmp = readl(mmio_base + SIL_SYSCFG);
	tmp |= SIL_MASK_IDE0_INT << ap->port_no;
	writel(tmp, mmio_base + SIL_SYSCFG);
	readl(mmio_base + SIL_SYSCFG);	/* flush */

	/* Ensure DMA_ENABLE is off.
	 *
	 * This is because the controller will not give us access to the
	 * taskfile registers while a DMA is in progress
	 */
	iowrite8(ioread8(ap->ioaddr.bmdma_addr) & ~SIL_DMA_ENABLE,
		 ap->ioaddr.bmdma_addr);

	/* According to ata_bmdma_stop, an HDMA transition requires
	 * on PIO cycle. But we can't read a taskfile register.
	 */
	ioread8(ap->ioaddr.bmdma_addr);
}

static void sil_thaw(struct ata_port *ap)
{
	void __iomem *mmio_base = ap->host->iomap[SIL_MMIO_BAR];
	u32 tmp;

	/* clear IRQ */
	ap->ops->sff_check_status(ap);
	ata_bmdma_irq_clear(ap);

	/* turn on SATA IRQ if supported */
	if (!(ap->flags & SIL_FLAG_NO_SATA_IRQ))
		writel(SIL_SIEN_N, mmio_base + sil_port[ap->port_no].sien);

	/* turn on IRQ */
	tmp = readl(mmio_base + SIL_SYSCFG);
	tmp &= ~(SIL_MASK_IDE0_INT << ap->port_no);
	writel(tmp, mmio_base + SIL_SYSCFG);
}

/**
 *	sil_dev_config - Apply device/host-specific errata fixups
 *	@dev: Device to be examined
 *
 *	After the IDENTIFY [PACKET] DEVICE step is complete, and a
 *	device is known to be present, this function is called.
 *	We apply two errata fixups which are specific to Silicon Image,
 *	a Seagate and a Maxtor fixup.
 *
 *	For certain Seagate devices, we must limit the maximum sectors
 *	to under 8K.
 *
 *	For certain Maxtor devices, we must not program the drive
 *	beyond udma5.
 *
 *	Both fixups are unfairly pessimistic.  As soon as I get more
 *	information on these errata, I will create a more exhaustive
 *	list, and apply the fixups to only the specific
 *	devices/hosts/firmwares that need it.
 *
 *	20040111 - Seagate drives affected by the Mod15Write bug are blacklisted
 *	The Maxtor quirk is in the blacklist, but I'm keeping the original
 *	pessimistic fix for the following reasons...
 *	- There seems to be less info on it, only one device gleaned off the
 *	Windows	driver, maybe only one is affected.  More info would be greatly
 *	appreciated.
 *	- But then again UDMA5 is hardly anything to complain about
 */
static void sil_dev_config(struct ata_device *dev)
{
	struct ata_port *ap = dev->link->ap;
	int print_info = ap->link.eh_context.i.flags & ATA_EHI_PRINTINFO;
	unsigned int n, quirks = 0;
	unsigned char model_num[ATA_ID_PROD_LEN + 1];

	/* This controller doesn't support trim */
	dev->horkage |= ATA_HORKAGE_NOTRIM;

	ata_id_c_string(dev->id, model_num, ATA_ID_PROD, sizeof(model_num));

	for (n = 0; sil_blacklist[n].product; n++)
		if (!strcmp(sil_blacklist[n].product, model_num)) {
			quirks = sil_blacklist[n].quirk;
			break;
		}

	/* limit requests to 15 sectors */
	if (slow_down ||
	    ((ap->flags & SIL_FLAG_MOD15WRITE) &&
	     (quirks & SIL_QUIRK_MOD15WRITE))) {
		if (print_info)
			ata_dev_info(dev,
		"applying Seagate errata fix (mod15write workaround)\n");
		dev->max_sectors = 15;
		return;
	}

	/* limit to udma5 */
	if (quirks & SIL_QUIRK_UDMA5MAX) {
		if (print_info)
			ata_dev_info(dev, "applying Maxtor errata fix %s\n",
				     model_num);
		dev->udma_mask &= ATA_UDMA5;
		return;
	}
}

static void sil_init_controller(struct ata_host *host)
{
	struct pci_dev *pdev = to_pci_dev(host->dev);
	void __iomem *mmio_base = host->iomap[SIL_MMIO_BAR];
	u8 cls;
	u32 tmp;
	int i;

	/* Initialize FIFO PCI bus arbitration */
	cls = sil_get_device_cache_line(pdev);
	if (cls) {
		cls >>= 3;
		cls++;  /* cls = (line_size/8)+1 */
		for (i = 0; i < host->n_ports; i++)
			writew(cls << 8 | cls,
			       mmio_base + sil_port[i].fifo_cfg);
	} else
		dev_warn(&pdev->dev,
			 "cache line size not set.  Driver may not function\n");

	/* Apply R_ERR on DMA activate FIS errata workaround */
	if (host->ports[0]->flags & SIL_FLAG_RERR_ON_DMA_ACT) {
		int cnt;

		for (i = 0, cnt = 0; i < host->n_ports; i++) {
			tmp = readl(mmio_base + sil_port[i].sfis_cfg);
			if ((tmp & 0x3) != 0x01)
				continue;
			if (!cnt)
				dev_info(&pdev->dev,
					 "Applying R_ERR on DMA activate FIS errata fix\n");
			writel(tmp & ~0x3, mmio_base + sil_port[i].sfis_cfg);
			cnt++;
		}
	}

	if (host->n_ports == 4) {
		/* flip the magic "make 4 ports work" bit */
		tmp = readl(mmio_base + sil_port[2].bmdma);
		if ((tmp & SIL_INTR_STEERING) == 0)
			writel(tmp | SIL_INTR_STEERING,
			       mmio_base + sil_port[2].bmdma);
	}
}

static bool sil_broken_system_poweroff(struct pci_dev *pdev)
{
	static const struct dmi_system_id broken_systems[] = {
		{
			.ident = "HP Compaq nx6325",
			.matches = {
				DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
				DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq nx6325"),
			},
			/* PCI slot number of the controller */
			.driver_data = (void *)0x12UL,
		},

		{ }	/* terminate list */
	};
	const struct dmi_system_id *dmi = dmi_first_match(broken_systems);

	if (dmi) {
		unsigned long slot = (unsigned long)dmi->driver_data;
		/* apply the quirk only to on-board controllers */
		return slot == PCI_SLOT(pdev->devfn);
	}

	return false;
}

static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
{
	int board_id = ent->driver_data;
	struct ata_port_info pi = sil_port_info[board_id];
	const struct ata_port_info *ppi[] = { &pi, NULL };
	struct ata_host *host;
	void __iomem *mmio_base;
	int n_ports, rc;
	unsigned int i;

	ata_print_version_once(&pdev->dev, DRV_VERSION);

	/* allocate host */
	n_ports = 2;
	if (board_id == sil_3114)
		n_ports = 4;

	if (sil_broken_system_poweroff(pdev)) {
		pi.flags |= ATA_FLAG_NO_POWEROFF_SPINDOWN |
					ATA_FLAG_NO_HIBERNATE_SPINDOWN;
		dev_info(&pdev->dev, "quirky BIOS, skipping spindown "
				"on poweroff and hibernation\n");
	}

	host = ata_host_alloc_pinfo(&pdev->dev, ppi, n_ports);
	if (!host)
		return -ENOMEM;

	/* acquire resources and fill host */
	rc = pcim_enable_device(pdev);
	if (rc)
		return rc;

	rc = pcim_iomap_regions(pdev, 1 << SIL_MMIO_BAR, DRV_NAME);
	if (rc == -EBUSY)
		pcim_pin_device(pdev);
	if (rc)
		return rc;
	host->iomap = pcim_iomap_table(pdev);

	rc = dma_set_mask(&pdev->dev, ATA_DMA_MASK);
	if (rc)
		return rc;
	rc = dma_set_coherent_mask(&pdev->dev, ATA_DMA_MASK);
	if (rc)
		return rc;

	mmio_base = host->iomap[SIL_MMIO_BAR];

	for (i = 0; i < host->n_ports; i++) {
		struct ata_port *ap = host->ports[i];
		struct ata_ioports *ioaddr = &ap->ioaddr;

		ioaddr->cmd_addr = mmio_base + sil_port[i].tf;
		ioaddr->altstatus_addr =
		ioaddr->ctl_addr = mmio_base + sil_port[i].ctl;
		ioaddr->bmdma_addr = mmio_base + sil_port[i].bmdma;
		ioaddr->scr_addr = mmio_base + sil_port[i].scr;
		ata_sff_std_ports(ioaddr);

		ata_port_pbar_desc(ap, SIL_MMIO_BAR, -1, "mmio");
		ata_port_pbar_desc(ap, SIL_MMIO_BAR, sil_port[i].tf, "tf");
	}

	/* initialize and activate */
	sil_init_controller(host);

	pci_set_master(pdev);
	return ata_host_activate(host, pdev->irq, sil_interrupt, IRQF_SHARED,
				 &sil_sht);
}

#ifdef CONFIG_PM_SLEEP
static int sil_pci_device_resume(struct pci_dev *pdev)
{
	struct ata_host *host = pci_get_drvdata(pdev);
	int rc;

	rc = ata_pci_device_do_resume(pdev);
	if (rc)
		return rc;

	sil_init_controller(host);
	ata_host_resume(host);

	return 0;
}
#endif

module_pci_driver(sil_pci_driver);
um">0, space = 0; char *base = (char *) de; struct dx_hash_info h = *hinfo; printk("names: "); while ((char *) de < base + size) { if (de->inode) { if (show_names) { int len = de->name_len; char *name = de->name; while (len--) printk("%c", *name++); ext3fs_dirhash(de->name, de->name_len, &h); printk(":%x.%u ", h.hash, ((char *) de - base)); } space += EXT3_DIR_REC_LEN(de->name_len); names++; } de = ext3_next_entry(de); } printk("(%i)\n", names); return (struct stats) { names, space, 1 }; } struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, struct dx_entry *entries, int levels) { unsigned blocksize = dir->i_sb->s_blocksize; unsigned count = dx_get_count (entries), names = 0, space = 0, i; unsigned bcount = 0; struct buffer_head *bh; int err; printk("%i indexed blocks...\n", count); for (i = 0; i < count; i++, entries++) { u32 block = dx_get_block(entries), hash = i? dx_get_hash(entries): 0; u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash; struct stats stats; printk("%s%3u:%03u hash %8x/%8x ",levels?"":" ", i, block, hash, range); if (!(bh = ext3_bread (NULL,dir, block, 0,&err))) continue; stats = levels? dx_show_entries(hinfo, dir, ((struct dx_node *) bh->b_data)->entries, levels - 1): dx_show_leaf(hinfo, (struct ext3_dir_entry_2 *) bh->b_data, blocksize, 0); names += stats.names; space += stats.space; bcount += stats.bcount; brelse (bh); } if (bcount) printk("%snames %u, fullness %u (%u%%)\n", levels?"":" ", names, space/bcount,(space/bcount)*100/blocksize); return (struct stats) { names, space, bcount}; } #endif /* DX_DEBUG */ /* * Probe for a directory leaf block to search. * * dx_probe can return ERR_BAD_DX_DIR, which means there was a format * error in the directory index, and the caller should fall back to * searching the directory normally. The callers of dx_probe **MUST** * check for this error code, and make sure it never gets reflected * back to userspace. */ static struct dx_frame * dx_probe(struct qstr *entry, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err) { unsigned count, indirect; struct dx_entry *at, *entries, *p, *q, *m; struct dx_root *root; struct buffer_head *bh; struct dx_frame *frame = frame_in; u32 hash; frame->bh = NULL; if (!(bh = ext3_bread (NULL,dir, 0, 0, err))) goto fail; root = (struct dx_root *) bh->b_data; if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && root->info.hash_version != DX_HASH_LEGACY) { ext3_warning(dir->i_sb, __func__, "Unrecognised inode hash code %d", root->info.hash_version); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail; } hinfo->hash_version = root->info.hash_version; hinfo->seed = EXT3_SB(dir->i_sb)->s_hash_seed; if (entry) ext3fs_dirhash(entry->name, entry->len, hinfo); hash = hinfo->hash; if (root->info.unused_flags & 1) { ext3_warning(dir->i_sb, __func__, "Unimplemented inode hash flags: %#06x", root->info.unused_flags); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail; } if ((indirect = root->info.indirect_levels) > 1) { ext3_warning(dir->i_sb, __func__, "Unimplemented inode hash depth: %#06x", root->info.indirect_levels); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail; } entries = (struct dx_entry *) (((char *)&root->info) + root->info.info_length); if (dx_get_limit(entries) != dx_root_limit(dir, root->info.info_length)) { ext3_warning(dir->i_sb, __func__, "dx entry: limit != root limit"); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail; } dxtrace (printk("Look up %x", hash)); while (1) { count = dx_get_count(entries); if (!count || count > dx_get_limit(entries)) { ext3_warning(dir->i_sb, __func__, "dx entry: no count or count > limit"); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail2; } p = entries + 1; q = entries + count - 1; while (p <= q) { m = p + (q - p)/2; dxtrace(printk(".")); if (dx_get_hash(m) > hash) q = m - 1; else p = m + 1; } if (0) // linear search cross check { unsigned n = count - 1; at = entries; while (n--) { dxtrace(printk(",")); if (dx_get_hash(++at) > hash) { at--; break; } } assert (at == p - 1); } at = p - 1; dxtrace(printk(" %x->%u\n", at == entries? 0: dx_get_hash(at), dx_get_block(at))); frame->bh = bh; frame->entries = entries; frame->at = at; if (!indirect--) return frame; if (!(bh = ext3_bread (NULL,dir, dx_get_block(at), 0, err))) goto fail2; at = entries = ((struct dx_node *) bh->b_data)->entries; if (dx_get_limit(entries) != dx_node_limit (dir)) { ext3_warning(dir->i_sb, __func__, "dx entry: limit != node limit"); brelse(bh); *err = ERR_BAD_DX_DIR; goto fail2; } frame++; frame->bh = NULL; } fail2: while (frame >= frame_in) { brelse(frame->bh); frame--; } fail: if (*err == ERR_BAD_DX_DIR) ext3_warning(dir->i_sb, __func__, "Corrupt dir inode %ld, running e2fsck is " "recommended.", dir->i_ino); return NULL; } static void dx_release (struct dx_frame *frames) { if (frames[0].bh == NULL) return; if (((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels) brelse(frames[1].bh); brelse(frames[0].bh); } /* * This function increments the frame pointer to search the next leaf * block, and reads in the necessary intervening nodes if the search * should be necessary. Whether or not the search is necessary is * controlled by the hash parameter. If the hash value is even, then * the search is only continued if the next block starts with that * hash value. This is used if we are searching for a specific file. * * If the hash value is HASH_NB_ALWAYS, then always go to the next block. * * This function returns 1 if the caller should continue to search, * or 0 if it should not. If there is an error reading one of the * index blocks, it will a negative error code. * * If start_hash is non-null, it will be filled in with the starting * hash of the next page. */ static int ext3_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, __u32 *start_hash) { struct dx_frame *p; struct buffer_head *bh; int err, num_frames = 0; __u32 bhash; p = frame; /* * Find the next leaf page by incrementing the frame pointer. * If we run out of entries in the interior node, loop around and * increment pointer in the parent node. When we break out of * this loop, num_frames indicates the number of interior * nodes need to be read. */ while (1) { if (++(p->at) < p->entries + dx_get_count(p->entries)) break; if (p == frames) return 0; num_frames++; p--; } /* * If the hash is 1, then continue only if the next page has a * continuation hash of any value. This is used for readdir * handling. Otherwise, check to see if the hash matches the * desired contiuation hash. If it doesn't, return since * there's no point to read in the successive index pages. */ bhash = dx_get_hash(p->at); if (start_hash) *start_hash = bhash; if ((hash & 1) == 0) { if ((bhash & ~1) != hash) return 0; } /* * If the hash is HASH_NB_ALWAYS, we always go to the next * block so no check is necessary */ while (num_frames--) { if (!(bh = ext3_bread(NULL, dir, dx_get_block(p->at), 0, &err))) return err; /* Failure */ p++; brelse (p->bh); p->bh = bh; p->at = p->entries = ((struct dx_node *) bh->b_data)->entries; } return 1; } /* * This function fills a red-black tree with information from a * directory block. It returns the number directory entries loaded * into the tree. If there is an error it is returned in err. */ static int htree_dirblock_to_tree(struct file *dir_file, struct inode *dir, int block, struct dx_hash_info *hinfo, __u32 start_hash, __u32 start_minor_hash) { struct buffer_head *bh; struct ext3_dir_entry_2 *de, *top; int err, count = 0; dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); if (!(bh = ext3_bread (NULL, dir, block, 0, &err))) return err; de = (struct ext3_dir_entry_2 *) bh->b_data; top = (struct ext3_dir_entry_2 *) ((char *) de + dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(0)); for (; de < top; de = ext3_next_entry(de)) { if (!ext3_check_dir_entry("htree_dirblock_to_tree", dir, de, bh, (block<<EXT3_BLOCK_SIZE_BITS(dir->i_sb)) +((char *)de - bh->b_data))) { /* On error, skip the f_pos to the next block. */ dir_file->f_pos = (dir_file->f_pos | (dir->i_sb->s_blocksize - 1)) + 1; brelse (bh); return count; } ext3fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || ((hinfo->hash == start_hash) && (hinfo->minor_hash < start_minor_hash))) continue; if (de->inode == 0) continue; if ((err = ext3_htree_store_dirent(dir_file, hinfo->hash, hinfo->minor_hash, de)) != 0) { brelse(bh); return err; } count++; } brelse(bh); return count; } /* * This function fills a red-black tree with information from a * directory. We start scanning the directory in hash order, starting * at start_hash and start_minor_hash. * * This function returns the number of entries inserted into the tree, * or a negative error code. */ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, __u32 start_minor_hash, __u32 *next_hash) { struct dx_hash_info hinfo; struct ext3_dir_entry_2 *de; struct dx_frame frames[2], *frame; struct inode *dir; int block, err; int count = 0; int ret; __u32 hashval; dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, start_minor_hash)); dir = dir_file->f_path.dentry->d_inode; if (!(EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) { hinfo.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo, start_hash, start_minor_hash); *next_hash = ~0; return count; } hinfo.hash = start_hash; hinfo.minor_hash = 0; frame = dx_probe(NULL, dir_file->f_path.dentry->d_inode, &hinfo, frames, &err); if (!frame) return err; /* Add '.' and '..' from the htree header */ if (!start_hash && !start_minor_hash) { de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; if ((err = ext3_htree_store_dirent(dir_file, 0, 0, de)) != 0) goto errout; count++; } if (start_hash < 2 || (start_hash ==2 && start_minor_hash==0)) { de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; de = ext3_next_entry(de); if ((err = ext3_htree_store_dirent(dir_file, 2, 0, de)) != 0) goto errout; count++; } while (1) { block = dx_get_block(frame->at); ret = htree_dirblock_to_tree(dir_file, dir, block, &hinfo, start_hash, start_minor_hash); if (ret < 0) { err = ret; goto errout; } count += ret; hashval = ~0; ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS, frame, frames, &hashval); *next_hash = hashval; if (ret < 0) { err = ret; goto errout; } /* * Stop if: (a) there are no more entries, or * (b) we have inserted at least one entry and the * next hash value is not a continuation */ if ((ret == 0) || (count && ((hashval & 1) == 0))) break; } dx_release(frames); dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", count, *next_hash)); return count; errout: dx_release(frames); return (err); } /* * Directory block splitting, compacting */ /* * Create map of hash values, offsets, and sizes, stored at end of block. * Returns number of entries mapped. */ static int dx_make_map (struct ext3_dir_entry_2 *de, int size, struct dx_hash_info *hinfo, struct dx_map_entry *map_tail) { int count = 0; char *base = (char *) de; struct dx_hash_info h = *hinfo; while ((char *) de < base + size) { if (de->name_len && de->inode) { ext3fs_dirhash(de->name, de->name_len, &h); map_tail--; map_tail->hash = h.hash; map_tail->offs = (u16) ((char *) de - base); map_tail->size = le16_to_cpu(de->rec_len); count++; cond_resched(); } /* XXX: do we need to check rec_len == 0 case? -Chris */ de = ext3_next_entry(de); } return count; } /* Sort map by hash value */ static void dx_sort_map (struct dx_map_entry *map, unsigned count) { struct dx_map_entry *p, *q, *top = map + count - 1; int more; /* Combsort until bubble sort doesn't suck */ while (count > 2) { count = count*10/13; if (count - 9 < 2) /* 9, 10 -> 11 */ count = 11; for (p = top, q = p - count; q >= map; p--, q--) if (p->hash < q->hash) swap(*p, *q); } /* Garden variety bubble sort */ do { more = 0; q = top; while (q-- > map) { if (q[1].hash >= q[0].hash) continue; swap(*(q+1), *q); more = 1; } } while(more); } static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block) { struct dx_entry *entries = frame->entries; struct dx_entry *old = frame->at, *new = old + 1; int count = dx_get_count(entries); assert(count < dx_get_limit(entries)); assert(old < entries + count); memmove(new + 1, new, (char *)(entries + count) - (char *)(new)); dx_set_hash(new, hash); dx_set_block(new, block); dx_set_count(entries, count + 1); } static void ext3_update_dx_flag(struct inode *inode) { if (!EXT3_HAS_COMPAT_FEATURE(inode->i_sb, EXT3_FEATURE_COMPAT_DIR_INDEX)) EXT3_I(inode)->i_flags &= ~EXT3_INDEX_FL; } /* * NOTE! unlike strncmp, ext3_match returns 1 for success, 0 for failure. * * `len <= EXT3_NAME_LEN' is guaranteed by caller. * `de != NULL' is guaranteed by caller. */ static inline int ext3_match (int len, const char * const name, struct ext3_dir_entry_2 * de) { if (len != de->name_len) return 0; if (!de->inode) return 0; return !memcmp(name, de->name, len); } /* * Returns 0 if not found, -1 on failure, and 1 on success */ static inline int search_dirblock(struct buffer_head * bh, struct inode *dir, struct qstr *child, unsigned long offset, struct ext3_dir_entry_2 ** res_dir) { struct ext3_dir_entry_2 * de; char * dlimit; int de_len; const char *name = child->name; int namelen = child->len; de = (struct ext3_dir_entry_2 *) bh->b_data; dlimit = bh->b_data + dir->i_sb->s_blocksize; while ((char *) de < dlimit) { /* this code is executed quadratically often */ /* do minimal checking `by hand' */ if ((char *) de + namelen <= dlimit && ext3_match (namelen, name, de)) { /* found a match - just to be sure, do a full check */ if (!ext3_check_dir_entry("ext3_find_entry", dir, de, bh, offset)) return -1; *res_dir = de; return 1; } /* prevent looping on a bad block */ de_len = ext3_rec_len_from_disk(de->rec_len); if (de_len <= 0) return -1; offset += de_len; de = (struct ext3_dir_entry_2 *) ((char *) de + de_len); } return 0; } /* * ext3_find_entry() * * finds an entry in the specified directory with the wanted name. It * returns the cache buffer in which the entry was found, and the entry * itself (as a parameter - res_dir). It does NOT read the inode of the * entry - you'll have to do that yourself if you want to. * * The returned buffer_head has ->b_count elevated. The caller is expected * to brelse() it when appropriate. */ static struct buffer_head *ext3_find_entry(struct inode *dir, struct qstr *entry, struct ext3_dir_entry_2 **res_dir) { struct super_block * sb; struct buffer_head * bh_use[NAMEI_RA_SIZE]; struct buffer_head * bh, *ret = NULL; unsigned long start, block, b; int ra_max = 0; /* Number of bh's in the readahead buffer, bh_use[] */ int ra_ptr = 0; /* Current index into readahead buffer */ int num = 0; int nblocks, i, err; int namelen; *res_dir = NULL; sb = dir->i_sb; namelen = entry->len; if (namelen > EXT3_NAME_LEN) return NULL; if (is_dx(dir)) { bh = ext3_dx_find_entry(dir, entry, res_dir, &err); /* * On success, or if the error was file not found, * return. Otherwise, fall back to doing a search the * old fashioned way. */ if (bh || (err != ERR_BAD_DX_DIR)) return bh; dxtrace(printk("ext3_find_entry: dx failed, falling back\n")); } nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); start = EXT3_I(dir)->i_dir_start_lookup; if (start >= nblocks) start = 0; block = start; restart: do { /* * We deal with the read-ahead logic here. */ if (ra_ptr >= ra_max) { /* Refill the readahead buffer */ ra_ptr = 0; b = block; for (ra_max = 0; ra_max < NAMEI_RA_SIZE; ra_max++) { /* * Terminate if we reach the end of the * directory and must wrap, or if our * search has finished at this block. */ if (b >= nblocks || (num && block == start)) { bh_use[ra_max] = NULL; break; } num++; bh = ext3_getblk(NULL, dir, b++, 0, &err); bh_use[ra_max] = bh; if (bh) ll_rw_block(READ_META, 1, &bh); } } if ((bh = bh_use[ra_ptr++]) == NULL) goto next; wait_on_buffer(bh); if (!buffer_uptodate(bh)) { /* read error, skip block & hope for the best */ ext3_error(sb, __func__, "reading directory #%lu " "offset %lu", dir->i_ino, block); brelse(bh); goto next; } i = search_dirblock(bh, dir, entry, block << EXT3_BLOCK_SIZE_BITS(sb), res_dir); if (i == 1) { EXT3_I(dir)->i_dir_start_lookup = block; ret = bh; goto cleanup_and_exit; } else { brelse(bh); if (i < 0) goto cleanup_and_exit; } next: if (++block >= nblocks) block = 0; } while (block != start); /* * If the directory has grown while we were searching, then * search the last part of the directory before giving up. */ block = nblocks; nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); if (block < nblocks) { start = 0; goto restart; } cleanup_and_exit: /* Clean up the read-ahead blocks */ for (; ra_ptr < ra_max; ra_ptr++) brelse (bh_use[ra_ptr]); return ret; } static struct buffer_head * ext3_dx_find_entry(struct inode *dir, struct qstr *entry, struct ext3_dir_entry_2 **res_dir, int *err) { struct super_block * sb; struct dx_hash_info hinfo; u32 hash; struct dx_frame frames[2], *frame; struct ext3_dir_entry_2 *de, *top; struct buffer_head *bh; unsigned long block; int retval; int namelen = entry->len; const u8 *name = entry->name; sb = dir->i_sb; /* NFS may look up ".." - look at dx_root directory block */ if (namelen > 2 || name[0] != '.'|| (namelen == 2 && name[1] != '.')) { if (!(frame = dx_probe(entry, dir, &hinfo, frames, err))) return NULL; } else { frame = frames; frame->bh = NULL; /* for dx_release() */ frame->at = (struct dx_entry *)frames; /* hack for zero entry*/ dx_set_block(frame->at, 0); /* dx_root block is 0 */ } hash = hinfo.hash; do { block = dx_get_block(frame->at); if (!(bh = ext3_bread (NULL,dir, block, 0, err))) goto errout; de = (struct ext3_dir_entry_2 *) bh->b_data; top = (struct ext3_dir_entry_2 *) ((char *) de + sb->s_blocksize - EXT3_DIR_REC_LEN(0)); for (; de < top; de = ext3_next_entry(de)) { int off = (block << EXT3_BLOCK_SIZE_BITS(sb)) + ((char *) de - bh->b_data); if (!ext3_check_dir_entry(__func__, dir, de, bh, off)) { brelse(bh); *err = ERR_BAD_DX_DIR; goto errout; } if (ext3_match(namelen, name, de)) { *res_dir = de; dx_release(frames); return bh; } } brelse (bh); /* Check to see if we should continue to search */ retval = ext3_htree_next_block(dir, hash, frame, frames, NULL); if (retval < 0) { ext3_warning(sb, __func__, "error reading index page in directory #%lu", dir->i_ino); *err = retval; goto errout; } } while (retval == 1); *err = -ENOENT; errout: dxtrace(printk("%s not found\n", name)); dx_release (frames); return NULL; } static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) { struct inode * inode; struct ext3_dir_entry_2 * de; struct buffer_head * bh; if (dentry->d_name.len > EXT3_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); bh = ext3_find_entry(dir, &dentry->d_name, &de); inode = NULL; if (bh) { unsigned long ino = le32_to_cpu(de->inode); brelse (bh); if (!ext3_valid_inum(dir->i_sb, ino)) { ext3_error(dir->i_sb, "ext3_lookup", "bad inode number: %lu", ino); return ERR_PTR(-EIO); } inode = ext3_iget(dir->i_sb, ino); if (IS_ERR(inode)) return ERR_CAST(inode); } return d_splice_alias(inode, dentry); } struct dentry *ext3_get_parent(struct dentry *child) { unsigned long ino; struct qstr dotdot = {.name = "..", .len = 2}; struct ext3_dir_entry_2 * de; struct buffer_head *bh; bh = ext3_find_entry(child->d_inode, &dotdot, &de); if (!bh) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); brelse(bh); if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { ext3_error(child->d_inode->i_sb, "ext3_get_parent", "bad inode number: %lu", ino); return ERR_PTR(-EIO); } return d_obtain_alias(ext3_iget(child->d_inode->i_sb, ino)); } #define S_SHIFT 12 static unsigned char ext3_type_by_mode[S_IFMT >> S_SHIFT] = { [S_IFREG >> S_SHIFT] = EXT3_FT_REG_FILE, [S_IFDIR >> S_SHIFT] = EXT3_FT_DIR, [S_IFCHR >> S_SHIFT] = EXT3_FT_CHRDEV, [S_IFBLK >> S_SHIFT] = EXT3_FT_BLKDEV, [S_IFIFO >> S_SHIFT] = EXT3_FT_FIFO, [S_IFSOCK >> S_SHIFT] = EXT3_FT_SOCK, [S_IFLNK >> S_SHIFT] = EXT3_FT_SYMLINK, }; static inline void ext3_set_de_type(struct super_block *sb, struct ext3_dir_entry_2 *de, umode_t mode) { if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_FILETYPE)) de->file_type = ext3_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; } /* * Move count entries from end of map between two memory locations. * Returns pointer to last entry moved. */ static struct ext3_dir_entry_2 * dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count) { unsigned rec_len = 0; while (count--) { struct ext3_dir_entry_2 *de = (struct ext3_dir_entry_2 *) (from + map->offs); rec_len = EXT3_DIR_REC_LEN(de->name_len); memcpy (to, de, rec_len); ((struct ext3_dir_entry_2 *) to)->rec_len = ext3_rec_len_to_disk(rec_len); de->inode = 0; map++; to += rec_len; } return (struct ext3_dir_entry_2 *) (to - rec_len); } /* * Compact each dir entry in the range to the minimal rec_len. * Returns pointer to last entry in range. */ static struct ext3_dir_entry_2* dx_pack_dirents(char *base, int size) { struct ext3_dir_entry_2 *next, *to, *prev, *de = (struct ext3_dir_entry_2 *) base; unsigned rec_len = 0; prev = to = de; while ((char*)de < base + size) { next = ext3_next_entry(de); if (de->inode && de->name_len) { rec_len = EXT3_DIR_REC_LEN(de->name_len); if (de > to) memmove(to, de, rec_len); to->rec_len = ext3_rec_len_to_disk(rec_len); prev = to; to = (struct ext3_dir_entry_2 *) (((char *) to) + rec_len); } de = next; } return prev; } /* * Split a full leaf block to make room for a new dir entry. * Allocate a new block, and move entries so that they are approx. equally full. * Returns pointer to de in block into which the new entry will be inserted. */ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct buffer_head **bh,struct dx_frame *frame, struct dx_hash_info *hinfo, int *error) { unsigned blocksize = dir->i_sb->s_blocksize; unsigned count, continued; struct buffer_head *bh2; u32 newblock; u32 hash2; struct dx_map_entry *map; char *data1 = (*bh)->b_data, *data2; unsigned split, move, size, i; struct ext3_dir_entry_2 *de = NULL, *de2; int err = 0; bh2 = ext3_append (handle, dir, &newblock, &err); if (!(bh2)) { brelse(*bh); *bh = NULL; goto errout; } BUFFER_TRACE(*bh, "get_write_access"); err = ext3_journal_get_write_access(handle, *bh); if (err) goto journal_error; BUFFER_TRACE(frame->bh, "get_write_access"); err = ext3_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; data2 = bh2->b_data; /* create map in the end of data2 block */ map = (struct dx_map_entry *) (data2 + blocksize); count = dx_make_map ((struct ext3_dir_entry_2 *) data1, blocksize, hinfo, map); map -= count; dx_sort_map (map, count); /* Split the existing block in the middle, size-wise */ size = 0; move = 0; for (i = count-1; i >= 0; i--) { /* is more than half of this entry in 2nd half of the block? */ if (size + map[i].size/2 > blocksize/2) break; size += map[i].size; move++; } /* map index at which we will split */ split = count - move; hash2 = map[split].hash; continued = hash2 == map[split - 1].hash; dxtrace(printk("Split block %i at %x, %i/%i\n", dx_get_block(frame->at), hash2, split, count-split)); /* Fancy dance to stay within two buffers */ de2 = dx_move_dirents(data1, data2, map + split, count - split); de = dx_pack_dirents(data1,blocksize); de->rec_len = ext3_rec_len_to_disk(data1 + blocksize - (char *) de); de2->rec_len = ext3_rec_len_to_disk(data2 + blocksize - (char *) de2); dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data1, blocksize, 1)); dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data2, blocksize, 1)); /* Which block gets the new entry? */ if (hinfo->hash >= hash2) { swap(*bh, bh2); de = de2; } dx_insert_block (frame, hash2 + continued, newblock); err = ext3_journal_dirty_metadata (handle, bh2); if (err) goto journal_error; err = ext3_journal_dirty_metadata (handle, frame->bh); if (err) goto journal_error; brelse (bh2); dxtrace(dx_show_index ("frame", frame->entries)); return de; journal_error: brelse(*bh); brelse(bh2); *bh = NULL; ext3_std_error(dir->i_sb, err); errout: *error = err; return NULL; } /* * Add a new entry into a directory (leaf) block. If de is non-NULL, * it points to a directory entry which is guaranteed to be large * enough for new directory entry. If de is NULL, then * add_dirent_to_buf will attempt search the directory block for * space. It will return -ENOSPC if no space is available, and -EIO * and -EEXIST if directory entry already exists. * * NOTE! bh is NOT released in the case where ENOSPC is returned. In * all other cases bh is released. */ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, struct inode *inode, struct ext3_dir_entry_2 *de, struct buffer_head * bh) { struct inode *dir = dentry->d_parent->d_inode; const char *name = dentry->d_name.name; int namelen = dentry->d_name.len; unsigned long offset = 0; unsigned short reclen; int nlen, rlen, err; char *top; reclen = EXT3_DIR_REC_LEN(namelen); if (!de) { de = (struct ext3_dir_entry_2 *)bh->b_data; top = bh->b_data + dir->i_sb->s_blocksize - reclen; while ((char *) de <= top) { if (!ext3_check_dir_entry("ext3_add_entry", dir, de, bh, offset)) { brelse (bh); return -EIO; } if (ext3_match (namelen, name, de)) { brelse (bh); return -EEXIST; } nlen = EXT3_DIR_REC_LEN(de->name_len); rlen = ext3_rec_len_from_disk(de->rec_len); if ((de->inode? rlen - nlen: rlen) >= reclen) break; de = (struct ext3_dir_entry_2 *)((char *)de + rlen); offset += rlen; } if ((char *) de > top) return -ENOSPC; } BUFFER_TRACE(bh, "get_write_access"); err = ext3_journal_get_write_access(handle, bh); if (err) { ext3_std_error(dir->i_sb, err); brelse(bh); return err; } /* By now the buffer is marked for journaling */ nlen = EXT3_DIR_REC_LEN(de->name_len); rlen = ext3_rec_len_from_disk(de->rec_len); if (de->inode) { struct ext3_dir_entry_2 *de1 = (struct ext3_dir_entry_2 *)((char *)de + nlen); de1->rec_len = ext3_rec_len_to_disk(rlen - nlen); de->rec_len = ext3_rec_len_to_disk(nlen); de = de1; } de->file_type = EXT3_FT_UNKNOWN; if (inode) { de->inode = cpu_to_le32(inode->i_ino); ext3_set_de_type(dir->i_sb, de, inode->i_mode); } else de->inode = 0; de->name_len = namelen; memcpy (de->name, name, namelen); /* * XXX shouldn't update any times until successful * completion of syscall, but too many callers depend * on this. * * XXX similarly, too many callers depend on * ext3_new_inode() setting the times, but error * recovery deletes the inode, so the worst that can * happen is that the times are slightly out of date * and/or different from the directory change time. */ dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC; ext3_update_dx_flag(dir); dir->i_version++; ext3_mark_inode_dirty(handle, dir); BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); err = ext3_journal_dirty_metadata(handle, bh); if (err) ext3_std_error(dir->i_sb, err); brelse(bh); return 0; } /* * This converts a one block unindexed directory to a 3 block indexed * directory, and adds the dentry to the indexed directory. */ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, struct inode *inode, struct buffer_head *bh) { struct inode *dir = dentry->d_parent->d_inode; const char *name = dentry->d_name.name; int namelen = dentry->d_name.len; struct buffer_head *bh2; struct dx_root *root; struct dx_frame frames[2], *frame; struct dx_entry *entries; struct ext3_dir_entry_2 *de, *de2; char *data1, *top; unsigned len; int retval; unsigned blocksize; struct dx_hash_info hinfo; u32 block; struct fake_dirent *fde; blocksize = dir->i_sb->s_blocksize; dxtrace(printk("Creating index\n")); retval = ext3_journal_get_write_access(handle, bh); if (retval) { ext3_std_error(dir->i_sb, retval); brelse(bh); return retval; } root = (struct dx_root *) bh->b_data; bh2 = ext3_append (handle, dir, &block, &retval); if (!(bh2)) { brelse(bh); return retval; } EXT3_I(dir)->i_flags |= EXT3_INDEX_FL; data1 = bh2->b_data; /* The 0th block becomes the root, move the dirents out */ fde = &root->dotdot; de = (struct ext3_dir_entry_2 *)((char *)fde + ext3_rec_len_from_disk(fde->rec_len)); len = ((char *) root) + blocksize - (char *) de; memcpy (data1, de, len); de = (struct ext3_dir_entry_2 *) data1; top = data1 + len; while ((char *)(de2 = ext3_next_entry(de)) < top) de = de2; de->rec_len = ext3_rec_len_to_disk(data1 + blocksize - (char *) de); /* Initialize the root; the dot dirents already exist */ de = (struct ext3_dir_entry_2 *) (&root->dotdot); de->rec_len = ext3_rec_len_to_disk(blocksize - EXT3_DIR_REC_LEN(2)); memset (&root->info, 0, sizeof(root->info)); root->info.info_length = sizeof(root->info); root->info.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; entries = root->entries; dx_set_block (entries, 1); dx_set_count (entries, 1); dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info))); /* Initialize as for dx_probe */ hinfo.hash_version = root->info.hash_version; hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; ext3fs_dirhash(name, namelen, &hinfo); frame = frames; frame->entries = entries; frame->at = entries; frame->bh = bh; bh = bh2; de = do_split(handle,dir, &bh, frame, &hinfo, &retval); dx_release (frames); if (!(de)) return retval; return add_dirent_to_buf(handle, dentry, inode, de, bh); } /* * ext3_add_entry() * * adds a file entry to the specified directory, using the same * semantics as ext3_find_entry(). It returns NULL if it failed. * * NOTE!! The inode part of 'de' is left at 0 - which means you * may not sleep between calling this and putting something into * the entry, as someone else might have used it while you slept. */ static int ext3_add_entry (handle_t *handle, struct dentry *dentry, struct inode *inode) { struct inode *dir = dentry->d_parent->d_inode; unsigned long offset; struct buffer_head * bh; struct ext3_dir_entry_2 *de; struct super_block * sb; int retval; int dx_fallback=0; unsigned blocksize; u32 block, blocks; sb = dir->i_sb; blocksize = sb->s_blocksize; if (!dentry->d_name.len) return -EINVAL; if (is_dx(dir)) { retval = ext3_dx_add_entry(handle, dentry, inode); if (!retval || (retval != ERR_BAD_DX_DIR)) return retval; EXT3_I(dir)->i_flags &= ~EXT3_INDEX_FL; dx_fallback++; ext3_mark_inode_dirty(handle, dir); } blocks = dir->i_size >> sb->s_blocksize_bits; for (block = 0, offset = 0; block < blocks; block++) { bh = ext3_bread(handle, dir, block, 0, &retval); if(!bh) return retval; retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh); if (retval != -ENOSPC) return retval; if (blocks == 1 && !dx_fallback && EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_DIR_INDEX)) return make_indexed_dir(handle, dentry, inode, bh); brelse(bh); } bh = ext3_append(handle, dir, &block, &retval); if (!bh) return retval; de = (struct ext3_dir_entry_2 *) bh->b_data; de->inode = 0; de->rec_len = ext3_rec_len_to_disk(blocksize); return add_dirent_to_buf(handle, dentry, inode, de, bh); } /* * Returns 0 for success, or a negative error value */ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode) { struct dx_frame frames[2], *frame; struct dx_entry *entries, *at; struct dx_hash_info hinfo; struct buffer_head * bh; struct inode *dir = dentry->d_parent->d_inode; struct super_block * sb = dir->i_sb; struct ext3_dir_entry_2 *de; int err; frame = dx_probe(&dentry->d_name, dir, &hinfo, frames, &err); if (!frame) return err; entries = frame->entries; at = frame->at; if (!(bh = ext3_bread(handle,dir, dx_get_block(frame->at), 0, &err))) goto cleanup; BUFFER_TRACE(bh, "get_write_access"); err = ext3_journal_get_write_access(handle, bh); if (err) goto journal_error; err = add_dirent_to_buf(handle, dentry, inode, NULL, bh); if (err != -ENOSPC) { bh = NULL; goto cleanup; } /* Block full, should compress but for now just split */ dxtrace(printk("using %u of %u node entries\n", dx_get_count(entries), dx_get_limit(entries))); /* Need to split index? */ if (dx_get_count(entries) == dx_get_limit(entries)) { u32 newblock; unsigned icount = dx_get_count(entries); int levels = frame - frames; struct dx_entry *entries2; struct dx_node *node2; struct buffer_head *bh2; if (levels && (dx_get_count(frames->entries) == dx_get_limit(frames->entries))) { ext3_warning(sb, __func__, "Directory index full!"); err = -ENOSPC; goto cleanup; } bh2 = ext3_append (handle, dir, &newblock, &err); if (!(bh2)) goto cleanup; node2 = (struct dx_node *)(bh2->b_data); entries2 = node2->entries; node2->fake.rec_len = ext3_rec_len_to_disk(sb->s_blocksize); node2->fake.inode = 0; BUFFER_TRACE(frame->bh, "get_write_access"); err = ext3_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; if (levels) { unsigned icount1 = icount/2, icount2 = icount - icount1; unsigned hash2 = dx_get_hash(entries + icount1); dxtrace(printk("Split index %i/%i\n", icount1, icount2)); BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ err = ext3_journal_get_write_access(handle, frames[0].bh); if (err) goto journal_error; memcpy ((char *) entries2, (char *) (entries + icount1), icount2 * sizeof(struct dx_entry)); dx_set_count (entries, icount1); dx_set_count (entries2, icount2); dx_set_limit (entries2, dx_node_limit(dir)); /* Which index block gets the new entry? */ if (at - entries >= icount1) { frame->at = at = at - entries - icount1 + entries2; frame->entries = entries = entries2; swap(frame->bh, bh2); } dx_insert_block (frames + 0, hash2, newblock); dxtrace(dx_show_index ("node", frames[1].entries)); dxtrace(dx_show_index ("node", ((struct dx_node *) bh2->b_data)->entries)); err = ext3_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; brelse (bh2); } else { dxtrace(printk("Creating second level index...\n")); memcpy((char *) entries2, (char *) entries, icount * sizeof(struct dx_entry)); dx_set_limit(entries2, dx_node_limit(dir)); /* Set up root */ dx_set_count(entries, 1); dx_set_block(entries + 0, newblock); ((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels = 1; /* Add new access path frame */ frame = frames + 1; frame->at = at = at - entries + entries2; frame->entries = entries = entries2; frame->bh = bh2; err = ext3_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; } ext3_journal_dirty_metadata(handle, frames[0].bh); } de = do_split(handle, dir, &bh, frame, &hinfo, &err); if (!de) goto cleanup; err = add_dirent_to_buf(handle, dentry, inode, de, bh); bh = NULL; goto cleanup; journal_error: ext3_std_error(dir->i_sb, err); cleanup: if (bh) brelse(bh); dx_release(frames); return err; } /* * ext3_delete_entry deletes a directory entry by merging it with the * previous entry */ static int ext3_delete_entry (handle_t *handle, struct inode * dir, struct ext3_dir_entry_2 * de_del, struct buffer_head * bh) { struct ext3_dir_entry_2 * de, * pde; int i; i = 0; pde = NULL; de = (struct ext3_dir_entry_2 *) bh->b_data; while (i < bh->b_size) { if (!ext3_check_dir_entry("ext3_delete_entry", dir, de, bh, i)) return -EIO; if (de == de_del) { BUFFER_TRACE(bh, "get_write_access"); ext3_journal_get_write_access(handle, bh); if (pde) pde->rec_len = ext3_rec_len_to_disk( ext3_rec_len_from_disk(pde->rec_len) + ext3_rec_len_from_disk(de->rec_len)); else de->inode = 0; dir->i_version++; BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, bh); return 0; } i += ext3_rec_len_from_disk(de->rec_len); pde = de; de = ext3_next_entry(de); } return -ENOENT; } static int ext3_add_nondir(handle_t *handle, struct dentry *dentry, struct inode *inode) { int err = ext3_add_entry(handle, dentry, inode); if (!err) { ext3_mark_inode_dirty(handle, inode); d_instantiate(dentry, inode); return 0; } drop_nlink(inode); iput(inode); return err; } /* * By the time this is called, we already have created * the directory cache entry for the new file, but it * is so far negative - it has no inode. * * If the create succeeds, we fill in the inode information * with d_instantiate(). */ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, struct nameidata *nd) { handle_t *handle; struct inode * inode; int err, retries = 0; retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode = ext3_new_inode (handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { inode->i_op = &ext3_file_inode_operations; inode->i_fop = &ext3_file_operations; ext3_set_aops(inode); err = ext3_add_nondir(handle, dentry, inode); } ext3_journal_stop(handle); if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } static int ext3_mknod (struct inode * dir, struct dentry *dentry, int mode, dev_t rdev) { handle_t *handle; struct inode *inode; int err, retries = 0; if (!new_valid_dev(rdev)) return -EINVAL; retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode = ext3_new_inode (handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { init_special_inode(inode, inode->i_mode, rdev); #ifdef CONFIG_EXT3_FS_XATTR inode->i_op = &ext3_special_inode_operations; #endif err = ext3_add_nondir(handle, dentry, inode); } ext3_journal_stop(handle); if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) { handle_t *handle; struct inode * inode; struct buffer_head * dir_block; struct ext3_dir_entry_2 * de; int err, retries = 0; if (dir->i_nlink >= EXT3_LINK_MAX) return -EMLINK; retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode = ext3_new_inode (handle, dir, S_IFDIR | mode); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; inode->i_op = &ext3_dir_inode_operations; inode->i_fop = &ext3_dir_operations; inode->i_size = EXT3_I(inode)->i_disksize = inode->i_sb->s_blocksize; dir_block = ext3_bread (handle, inode, 0, 1, &err); if (!dir_block) { drop_nlink(inode); /* is this nlink == 0? */ ext3_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } BUFFER_TRACE(dir_block, "get_write_access"); ext3_journal_get_write_access(handle, dir_block); de = (struct ext3_dir_entry_2 *) dir_block->b_data; de->inode = cpu_to_le32(inode->i_ino); de->name_len = 1; de->rec_len = ext3_rec_len_to_disk(EXT3_DIR_REC_LEN(de->name_len)); strcpy (de->name, "."); ext3_set_de_type(dir->i_sb, de, S_IFDIR); de = ext3_next_entry(de); de->inode = cpu_to_le32(dir->i_ino); de->rec_len = ext3_rec_len_to_disk(inode->i_sb->s_blocksize - EXT3_DIR_REC_LEN(1)); de->name_len = 2; strcpy (de->name, ".."); ext3_set_de_type(dir->i_sb, de, S_IFDIR); inode->i_nlink = 2; BUFFER_TRACE(dir_block, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, dir_block); brelse (dir_block); ext3_mark_inode_dirty(handle, inode); err = ext3_add_entry (handle, dentry, inode); if (err) { inode->i_nlink = 0; ext3_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } inc_nlink(dir); ext3_update_dx_flag(dir); ext3_mark_inode_dirty(handle, dir); d_instantiate(dentry, inode); out_stop: ext3_journal_stop(handle); if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } /* * routine to check that the specified directory is empty (for rmdir) */ static int empty_dir (struct inode * inode) { unsigned long offset; struct buffer_head * bh; struct ext3_dir_entry_2 * de, * de1; struct super_block * sb; int err = 0; sb = inode->i_sb; if (inode->i_size < EXT3_DIR_REC_LEN(1) + EXT3_DIR_REC_LEN(2) || !(bh = ext3_bread (NULL, inode, 0, 0, &err))) { if (err) ext3_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", err, inode->i_ino); else ext3_warning(inode->i_sb, __func__, "bad directory (dir #%lu) - no data block", inode->i_ino); return 1; } de = (struct ext3_dir_entry_2 *) bh->b_data; de1 = ext3_next_entry(de); if (le32_to_cpu(de->inode) != inode->i_ino || !le32_to_cpu(de1->inode) || strcmp (".", de->name) || strcmp ("..", de1->name)) { ext3_warning (inode->i_sb, "empty_dir", "bad directory (dir #%lu) - no `.' or `..'", inode->i_ino); brelse (bh); return 1; } offset = ext3_rec_len_from_disk(de->rec_len) + ext3_rec_len_from_disk(de1->rec_len); de = ext3_next_entry(de1); while (offset < inode->i_size ) { if (!bh || (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { err = 0; brelse (bh); bh = ext3_bread (NULL, inode, offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) ext3_error(sb, __func__, "error %d reading directory" " #%lu offset %lu", err, inode->i_ino, offset); offset += sb->s_blocksize; continue; } de = (struct ext3_dir_entry_2 *) bh->b_data; } if (!ext3_check_dir_entry("empty_dir", inode, de, bh, offset)) { de = (struct ext3_dir_entry_2 *)(bh->b_data + sb->s_blocksize); offset = (offset | (sb->s_blocksize - 1)) + 1; continue; } if (le32_to_cpu(de->inode)) { brelse (bh); return 0; } offset += ext3_rec_len_from_disk(de->rec_len); de = ext3_next_entry(de); } brelse (bh); return 1; } /* ext3_orphan_add() links an unlinked or truncated inode into a list of * such inodes, starting at the superblock, in case we crash before the * file is closed/deleted, or in case the inode truncate spans multiple * transactions and the last transaction is not recovered after a crash. * * At filesystem recovery time, we walk this list deleting unlinked * inodes and truncating linked inodes in ext3_orphan_cleanup(). */ int ext3_orphan_add(handle_t *handle, struct inode *inode) { struct super_block *sb = inode->i_sb; struct ext3_iloc iloc; int err = 0, rc; lock_super(sb); if (!list_empty(&EXT3_I(inode)->i_orphan)) goto out_unlock; /* Orphan handling is only valid for files with data blocks * being truncated, or files being unlinked. */ /* @@@ FIXME: Observation from aviro: * I think I can trigger J_ASSERT in ext3_orphan_add(). We block * here (on lock_super()), so race with ext3_link() which might bump * ->i_nlink. For, say it, character device. Not a regular file, * not a directory, not a symlink and ->i_nlink > 0. */ J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); BUFFER_TRACE(EXT3_SB(sb)->s_sbh, "get_write_access"); err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh); if (err) goto out_unlock; err = ext3_reserve_inode_write(handle, inode, &iloc); if (err) goto out_unlock; /* Insert this inode at the head of the on-disk orphan list... */ NEXT_ORPHAN(inode) = le32_to_cpu(EXT3_SB(sb)->s_es->s_last_orphan); EXT3_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); err = ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); rc = ext3_mark_iloc_dirty(handle, inode, &iloc); if (!err) err = rc; /* Only add to the head of the in-memory list if all the * previous operations succeeded. If the orphan_add is going to * fail (possibly taking the journal offline), we can't risk * leaving the inode on the orphan list: stray orphan-list * entries can cause panics at unmount time. * * This is safe: on error we're going to ignore the orphan list * anyway on the next recovery. */ if (!err) list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan); jbd_debug(4, "superblock will point to %lu\n", inode->i_ino); jbd_debug(4, "orphan inode %lu will point to %d\n", inode->i_ino, NEXT_ORPHAN(inode)); out_unlock: unlock_super(sb); ext3_std_error(inode->i_sb, err); return err; } /* * ext3_orphan_del() removes an unlinked or truncated inode from the list * of such inodes stored on disk, because it is finally being cleaned up. */ int ext3_orphan_del(handle_t *handle, struct inode *inode) { struct list_head *prev; struct ext3_inode_info *ei = EXT3_I(inode); struct ext3_sb_info *sbi; unsigned long ino_next; struct ext3_iloc iloc; int err = 0; lock_super(inode->i_sb); if (list_empty(&ei->i_orphan)) { unlock_super(inode->i_sb); return 0; } ino_next = NEXT_ORPHAN(inode); prev = ei->i_orphan.prev; sbi = EXT3_SB(inode->i_sb); jbd_debug(4, "remove inode %lu from orphan list\n", inode->i_ino); list_del_init(&ei->i_orphan); /* If we're on an error path, we may not have a valid * transaction handle with which to update the orphan list on * disk, but we still need to remove the inode from the linked * list in memory. */ if (!handle) goto out; err = ext3_reserve_inode_write(handle, inode, &iloc); if (err) goto out_err; if (prev == &sbi->s_orphan) { jbd_debug(4, "superblock will point to %lu\n", ino_next); BUFFER_TRACE(sbi->s_sbh, "get_write_access"); err = ext3_journal_get_write_access(handle, sbi->s_sbh); if (err) goto out_brelse; sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); err = ext3_journal_dirty_metadata(handle, sbi->s_sbh); } else { struct ext3_iloc iloc2; struct inode *i_prev = &list_entry(prev, struct ext3_inode_info, i_orphan)->vfs_inode; jbd_debug(4, "orphan inode %lu will point to %lu\n", i_prev->i_ino, ino_next); err = ext3_reserve_inode_write(handle, i_prev, &iloc2); if (err) goto out_brelse; NEXT_ORPHAN(i_prev) = ino_next; err = ext3_mark_iloc_dirty(handle, i_prev, &iloc2); } if (err) goto out_brelse; NEXT_ORPHAN(inode) = 0; err = ext3_mark_iloc_dirty(handle, inode, &iloc); out_err: ext3_std_error(inode->i_sb, err); out: unlock_super(inode->i_sb); return err; out_brelse: brelse(iloc.bh); goto out_err; } static int ext3_rmdir (struct inode * dir, struct dentry *dentry) { int retval; struct inode * inode; struct buffer_head * bh; struct ext3_dir_entry_2 * de; handle_t *handle; /* Initialize quotas before so that eventual writes go in * separate transaction */ DQUOT_INIT(dentry->d_inode); handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); retval = -ENOENT; bh = ext3_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_rmdir; if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode = dentry->d_inode; retval = -EIO; if (le32_to_cpu(de->inode) != inode->i_ino) goto end_rmdir; retval = -ENOTEMPTY; if (!empty_dir (inode)) goto end_rmdir; retval = ext3_delete_entry(handle, dir, de, bh); if (retval) goto end_rmdir; if (inode->i_nlink != 2) ext3_warning (inode->i_sb, "ext3_rmdir", "empty directory has nlink!=2 (%d)", inode->i_nlink); inode->i_version++; clear_nlink(inode); /* There's no need to set i_disksize: the fact that i_nlink is * zero will ensure that the right thing happens during any * recovery. */ inode->i_size = 0; ext3_orphan_add(handle, inode); inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; ext3_mark_inode_dirty(handle, inode); drop_nlink(dir); ext3_update_dx_flag(dir); ext3_mark_inode_dirty(handle, dir); end_rmdir: ext3_journal_stop(handle); brelse (bh); return retval; } static int ext3_unlink(struct inode * dir, struct dentry *dentry) { int retval; struct inode * inode; struct buffer_head * bh; struct ext3_dir_entry_2 * de; handle_t *handle; /* Initialize quotas before so that eventual writes go * in separate transaction */ DQUOT_INIT(dentry->d_inode); handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; retval = -ENOENT; bh = ext3_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_unlink; inode = dentry->d_inode; retval = -EIO; if (le32_to_cpu(de->inode) != inode->i_ino) goto end_unlink; if (!inode->i_nlink) { ext3_warning (inode->i_sb, "ext3_unlink", "Deleting nonexistent file (%lu), %d", inode->i_ino, inode->i_nlink); inode->i_nlink = 1; } retval = ext3_delete_entry(handle, dir, de, bh); if (retval) goto end_unlink; dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; ext3_update_dx_flag(dir); ext3_mark_inode_dirty(handle, dir); drop_nlink(inode); if (!inode->i_nlink) ext3_orphan_add(handle, inode); inode->i_ctime = dir->i_ctime; ext3_mark_inode_dirty(handle, inode); retval = 0; end_unlink: ext3_journal_stop(handle); brelse (bh); return retval; } static int ext3_symlink (struct inode * dir, struct dentry *dentry, const char * symname) { handle_t *handle; struct inode * inode; int l, err, retries = 0; l = strlen(symname)+1; if (l > dir->i_sb->s_blocksize) return -ENAMETOOLONG; retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode = ext3_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; if (l > sizeof (EXT3_I(inode)->i_data)) { inode->i_op = &ext3_symlink_inode_operations; ext3_set_aops(inode); /* * page_symlink() calls into ext3_prepare/commit_write. * We have a transaction open. All is sweetness. It also sets * i_size in generic_commit_write(). */ err = __page_symlink(inode, symname, l, mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS); if (err) { drop_nlink(inode); ext3_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } } else { inode->i_op = &ext3_fast_symlink_inode_operations; memcpy((char*)&EXT3_I(inode)->i_data,symname,l); inode->i_size = l-1; } EXT3_I(inode)->i_disksize = inode->i_size; err = ext3_add_nondir(handle, dentry, inode); out_stop: ext3_journal_stop(handle); if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } static int ext3_link (struct dentry * old_dentry, struct inode * dir, struct dentry *dentry) { handle_t *handle; struct inode *inode = old_dentry->d_inode; int err, retries = 0; if (inode->i_nlink >= EXT3_LINK_MAX) return -EMLINK; /* * Return -ENOENT if we've raced with unlink and i_nlink is 0. Doing * otherwise has the potential to corrupt the orphan inode list. */ if (inode->i_nlink == 0) return -ENOENT; retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; inode->i_ctime = CURRENT_TIME_SEC; inc_nlink(inode); atomic_inc(&inode->i_count); err = ext3_add_nondir(handle, dentry, inode); ext3_journal_stop(handle); if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } #define PARENT_INO(buffer) \ (ext3_next_entry((struct ext3_dir_entry_2 *)(buffer))->inode) /* * Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, struct inode * new_dir,struct dentry *new_dentry) { handle_t *handle; struct inode * old_inode, * new_inode; struct buffer_head * old_bh, * new_bh, * dir_bh; struct ext3_dir_entry_2 * old_de, * new_de; int retval; old_bh = new_bh = dir_bh = NULL; /* Initialize quotas before so that eventual writes go * in separate transaction */ if (new_dentry->d_inode) DQUOT_INIT(new_dentry->d_inode); handle = ext3_journal_start(old_dir, 2 * EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) handle->h_sync = 1; old_bh = ext3_find_entry(old_dir, &old_dentry->d_name, &old_de); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process * and merrily kill the link to whatever was created under the * same name. Goodbye sticky bit ;-< */ old_inode = old_dentry->d_inode; retval = -ENOENT; if (!old_bh || le32_to_cpu(old_de->inode) != old_inode->i_ino) goto end_rename; new_inode = new_dentry->d_inode; new_bh = ext3_find_entry(new_dir, &new_dentry->d_name, &new_de); if (new_bh) { if (!new_inode) { brelse (new_bh); new_bh = NULL; } } if (S_ISDIR(old_inode->i_mode)) { if (new_inode) { retval = -ENOTEMPTY; if (!empty_dir (new_inode)) goto end_rename; } retval = -EIO; dir_bh = ext3_bread (handle, old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) goto end_rename; retval = -EMLINK; if (!new_inode && new_dir!=old_dir && new_dir->i_nlink >= EXT3_LINK_MAX) goto end_rename; } if (!new_bh) { retval = ext3_add_entry (handle, new_dentry, old_inode); if (retval) goto end_rename; } else { BUFFER_TRACE(new_bh, "get write access"); ext3_journal_get_write_access(handle, new_bh); new_de->inode = cpu_to_le32(old_inode->i_ino); if (EXT3_HAS_INCOMPAT_FEATURE(new_dir->i_sb, EXT3_FEATURE_INCOMPAT_FILETYPE)) new_de->file_type = old_de->file_type; new_dir->i_version++; new_dir->i_ctime = new_dir->i_mtime = CURRENT_TIME_SEC; ext3_mark_inode_dirty(handle, new_dir); BUFFER_TRACE(new_bh, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, new_bh); brelse(new_bh); new_bh = NULL; } /* * Like most other Unix systems, set the ctime for inodes on a * rename. */ old_inode->i_ctime = CURRENT_TIME_SEC; ext3_mark_inode_dirty(handle, old_inode); /* * ok, that's it */ if (le32_to_cpu(old_de->inode) != old_inode->i_ino || old_de->name_len != old_dentry->d_name.len || strncmp(old_de->name, old_dentry->d_name.name, old_de->name_len) || (retval = ext3_delete_entry(handle, old_dir, old_de, old_bh)) == -ENOENT) { /* old_de could have moved from under us during htree split, so * make sure that we are deleting the right entry. We might * also be pointing to a stale entry in the unused part of * old_bh so just checking inum and the name isn't enough. */ struct buffer_head *old_bh2; struct ext3_dir_entry_2 *old_de2; old_bh2 = ext3_find_entry(old_dir, &old_dentry->d_name, &old_de2); if (old_bh2) { retval = ext3_delete_entry(handle, old_dir, old_de2, old_bh2); brelse(old_bh2); } } if (retval) { ext3_warning(old_dir->i_sb, "ext3_rename", "Deleting old file (%lu), %d, error=%d", old_dir->i_ino, old_dir->i_nlink, retval); } if (new_inode) { drop_nlink(new_inode); new_inode->i_ctime = CURRENT_TIME_SEC; } old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC; ext3_update_dx_flag(old_dir); if (dir_bh) { BUFFER_TRACE(dir_bh, "get_write_access"); ext3_journal_get_write_access(handle, dir_bh); PARENT_INO(dir_bh->b_data) = cpu_to_le32(new_dir->i_ino); BUFFER_TRACE(dir_bh, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, dir_bh); drop_nlink(old_dir); if (new_inode) { drop_nlink(new_inode); } else { inc_nlink(new_dir); ext3_update_dx_flag(new_dir); ext3_mark_inode_dirty(handle, new_dir); } } ext3_mark_inode_dirty(handle, old_dir); if (new_inode) { ext3_mark_inode_dirty(handle, new_inode); if (!new_inode->i_nlink) ext3_orphan_add(handle, new_inode); } retval = 0; end_rename: brelse (dir_bh); brelse (old_bh); brelse (new_bh); ext3_journal_stop(handle); return retval; } /* * directories can handle most operations... */ const struct inode_operations ext3_dir_inode_operations = { .create = ext3_create, .lookup = ext3_lookup, .link = ext3_link, .unlink = ext3_unlink, .symlink = ext3_symlink, .mkdir = ext3_mkdir, .rmdir = ext3_rmdir, .mknod = ext3_mknod, .rename = ext3_rename, .setattr = ext3_setattr, #ifdef CONFIG_EXT3_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext3_listxattr, .removexattr = generic_removexattr, #endif .permission = ext3_permission, }; const struct inode_operations ext3_special_inode_operations = { .setattr = ext3_setattr, #ifdef CONFIG_EXT3_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext3_listxattr, .removexattr = generic_removexattr, #endif .permission = ext3_permission, };