~hc/RK356X_SDK_RELEASE.git

..	..	@@ -11,18 +11,15 @@
11	11	#include "xfs_trans_resv.h"
12	12	#include "xfs_sb.h"
13	13	#include "xfs_mount.h"
14		-#include "xfs_da_format.h"
15	14	#include "xfs_inode.h"
16	15	#include "xfs_btree.h"
17	16	#include "xfs_bmap.h"
18	17	#include "xfs_alloc.h"
19		-#include "xfs_error.h"
20	18	#include "xfs_fsops.h"
21	19	#include "xfs_trans.h"
22	20	#include "xfs_buf_item.h"
23	21	#include "xfs_log.h"
24	22	#include "xfs_log_priv.h"
25		-#include "xfs_da_btree.h"
26	23	#include "xfs_dir2.h"
27	24	#include "xfs_extfree_item.h"
28	25	#include "xfs_mru_cache.h"
..	..	@@ -39,417 +36,119 @@
39	36	#include "xfs_bmap_item.h"
40	37	#include "xfs_reflink.h"
41	38
42		-#include <linux/namei.h>
43		-#include <linux/dax.h>
44		-#include <linux/init.h>
45		-#include <linux/slab.h>
46		-#include <linux/mount.h>
47		-#include <linux/mempool.h>
48		-#include <linux/writeback.h>
49		-#include <linux/kthread.h>
50		-#include <linux/freezer.h>
51		-#include <linux/parser.h>
	39	+#include <linux/magic.h>
	40	+#include <linux/fs_context.h>
	41	+#include <linux/fs_parser.h>
52	42
53	43	static const struct super_operations xfs_super_operations;
54		-struct bio_set xfs_ioend_bioset;
55	44
56	45	static struct kset xfs_kset; / top-level xfs sysfs dir */
57	46	#ifdef DEBUG
58	47	static struct xfs_kobj xfs_dbg_kobj; /* global debug sysfs attrs */
59	48	#endif
60	49
	50	+enum xfs_dax_mode {
	51	+ XFS_DAX_INODE = 0,
	52	+ XFS_DAX_ALWAYS = 1,
	53	+ XFS_DAX_NEVER = 2,
	54	+};
	55	+
	56	+static void
	57	+xfs_mount_set_dax_mode(
	58	+ struct xfs_mount *mp,
	59	+ enum xfs_dax_mode mode)
	60	+{
	61	+ switch (mode) {
	62	+ case XFS_DAX_INODE:
	63	+ mp->m_flags &= ~(XFS_MOUNT_DAX_ALWAYS \| XFS_MOUNT_DAX_NEVER);
	64	+ break;
	65	+ case XFS_DAX_ALWAYS:
	66	+ mp->m_flags \|= XFS_MOUNT_DAX_ALWAYS;
	67	+ mp->m_flags &= ~XFS_MOUNT_DAX_NEVER;
	68	+ break;
	69	+ case XFS_DAX_NEVER:
	70	+ mp->m_flags \|= XFS_MOUNT_DAX_NEVER;
	71	+ mp->m_flags &= ~XFS_MOUNT_DAX_ALWAYS;
	72	+ break;
	73	+ }
	74	+}
	75	+
	76	+static const struct constant_table dax_param_enums[] = {
	77	+ {"inode", XFS_DAX_INODE },
	78	+ {"always", XFS_DAX_ALWAYS },
	79	+ {"never", XFS_DAX_NEVER },
	80	+ {}
	81	+};
	82	+
61	83	/*
62	84	* Table driven mount option parser.
63	85	*/
64	86	enum {
65		- Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev, Opt_biosize,
	87	+ Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev,
66	88	Opt_wsync, Opt_noalign, Opt_swalloc, Opt_sunit, Opt_swidth, Opt_nouuid,
67		- Opt_mtpt, Opt_grpid, Opt_nogrpid, Opt_bsdgroups, Opt_sysvgroups,
	89	+ Opt_grpid, Opt_nogrpid, Opt_bsdgroups, Opt_sysvgroups,
68	90	Opt_allocsize, Opt_norecovery, Opt_inode64, Opt_inode32, Opt_ikeep,
69	91	Opt_noikeep, Opt_largeio, Opt_nolargeio, Opt_attr2, Opt_noattr2,
70	92	Opt_filestreams, Opt_quota, Opt_noquota, Opt_usrquota, Opt_grpquota,
71	93	Opt_prjquota, Opt_uquota, Opt_gquota, Opt_pquota,
72	94	Opt_uqnoenforce, Opt_gqnoenforce, Opt_pqnoenforce, Opt_qnoenforce,
73		- Opt_discard, Opt_nodiscard, Opt_dax, Opt_err,
	95	+ Opt_discard, Opt_nodiscard, Opt_dax, Opt_dax_enum,
74	96	};
75	97
76		-static const match_table_t tokens = {
77		- {Opt_logbufs, "logbufs=%u"}, /* number of XFS log buffers */
78		- {Opt_logbsize, "logbsize=%s"}, /* size of XFS log buffers */
79		- {Opt_logdev, "logdev=%s"}, /* log device */
80		- {Opt_rtdev, "rtdev=%s"}, /* realtime I/O device */
81		- {Opt_biosize, "biosize=%u"}, /* log2 of preferred buffered io size */
82		- {Opt_wsync, "wsync"}, /* safe-mode nfs compatible mount */
83		- {Opt_noalign, "noalign"}, /* turn off stripe alignment */
84		- {Opt_swalloc, "swalloc"}, /* turn on stripe width allocation */
85		- {Opt_sunit, "sunit=%u"}, /* data volume stripe unit */
86		- {Opt_swidth, "swidth=%u"}, /* data volume stripe width */
87		- {Opt_nouuid, "nouuid"}, /* ignore filesystem UUID */
88		- {Opt_mtpt, "mtpt"}, /* filesystem mount point */
89		- {Opt_grpid, "grpid"}, /* group-ID from parent directory */
90		- {Opt_nogrpid, "nogrpid"}, /* group-ID from current process */
91		- {Opt_bsdgroups, "bsdgroups"}, /* group-ID from parent directory */
92		- {Opt_sysvgroups,"sysvgroups"}, /* group-ID from current process */
93		- {Opt_allocsize, "allocsize=%s"},/* preferred allocation size */
94		- {Opt_norecovery,"norecovery"}, /* don't run XFS recovery */
95		- {Opt_inode64, "inode64"}, /* inodes can be allocated anywhere */
96		- {Opt_inode32, "inode32"}, /* inode allocation limited to
97		- * XFS_MAXINUMBER_32 */
98		- {Opt_ikeep, "ikeep"}, /* do not free empty inode clusters */
99		- {Opt_noikeep, "noikeep"}, /* free empty inode clusters */
100		- {Opt_largeio, "largeio"}, /* report large I/O sizes in stat() */
101		- {Opt_nolargeio, "nolargeio"}, /* do not report large I/O sizes
102		- * in stat(). */
103		- {Opt_attr2, "attr2"}, /* do use attr2 attribute format */
104		- {Opt_noattr2, "noattr2"}, /* do not use attr2 attribute format */
105		- {Opt_filestreams,"filestreams"},/* use filestreams allocator */
106		- {Opt_quota, "quota"}, /* disk quotas (user) */
107		- {Opt_noquota, "noquota"}, /* no quotas */
108		- {Opt_usrquota, "usrquota"}, /* user quota enabled */
109		- {Opt_grpquota, "grpquota"}, /* group quota enabled */
110		- {Opt_prjquota, "prjquota"}, /* project quota enabled */
111		- {Opt_uquota, "uquota"}, /* user quota (IRIX variant) */
112		- {Opt_gquota, "gquota"}, /* group quota (IRIX variant) */
113		- {Opt_pquota, "pquota"}, /* project quota (IRIX variant) */
114		- {Opt_uqnoenforce,"uqnoenforce"},/* user quota limit enforcement */
115		- {Opt_gqnoenforce,"gqnoenforce"},/* group quota limit enforcement */
116		- {Opt_pqnoenforce,"pqnoenforce"},/* project quota limit enforcement */
117		- {Opt_qnoenforce, "qnoenforce"}, /* same as uqnoenforce */
118		- {Opt_discard, "discard"}, /* Discard unused blocks */
119		- {Opt_nodiscard, "nodiscard"}, /* Do not discard unused blocks */
120		- {Opt_dax, "dax"}, /* Enable direct access to bdev pages */
121		- {Opt_err, NULL},
	98	+static const struct fs_parameter_spec xfs_fs_parameters[] = {
	99	+ fsparam_u32("logbufs", Opt_logbufs),
	100	+ fsparam_string("logbsize", Opt_logbsize),
	101	+ fsparam_string("logdev", Opt_logdev),
	102	+ fsparam_string("rtdev", Opt_rtdev),
	103	+ fsparam_flag("wsync", Opt_wsync),
	104	+ fsparam_flag("noalign", Opt_noalign),
	105	+ fsparam_flag("swalloc", Opt_swalloc),
	106	+ fsparam_u32("sunit", Opt_sunit),
	107	+ fsparam_u32("swidth", Opt_swidth),
	108	+ fsparam_flag("nouuid", Opt_nouuid),
	109	+ fsparam_flag("grpid", Opt_grpid),
	110	+ fsparam_flag("nogrpid", Opt_nogrpid),
	111	+ fsparam_flag("bsdgroups", Opt_bsdgroups),
	112	+ fsparam_flag("sysvgroups", Opt_sysvgroups),
	113	+ fsparam_string("allocsize", Opt_allocsize),
	114	+ fsparam_flag("norecovery", Opt_norecovery),
	115	+ fsparam_flag("inode64", Opt_inode64),
	116	+ fsparam_flag("inode32", Opt_inode32),
	117	+ fsparam_flag("ikeep", Opt_ikeep),
	118	+ fsparam_flag("noikeep", Opt_noikeep),
	119	+ fsparam_flag("largeio", Opt_largeio),
	120	+ fsparam_flag("nolargeio", Opt_nolargeio),
	121	+ fsparam_flag("attr2", Opt_attr2),
	122	+ fsparam_flag("noattr2", Opt_noattr2),
	123	+ fsparam_flag("filestreams", Opt_filestreams),
	124	+ fsparam_flag("quota", Opt_quota),
	125	+ fsparam_flag("noquota", Opt_noquota),
	126	+ fsparam_flag("usrquota", Opt_usrquota),
	127	+ fsparam_flag("grpquota", Opt_grpquota),
	128	+ fsparam_flag("prjquota", Opt_prjquota),
	129	+ fsparam_flag("uquota", Opt_uquota),
	130	+ fsparam_flag("gquota", Opt_gquota),
	131	+ fsparam_flag("pquota", Opt_pquota),
	132	+ fsparam_flag("uqnoenforce", Opt_uqnoenforce),
	133	+ fsparam_flag("gqnoenforce", Opt_gqnoenforce),
	134	+ fsparam_flag("pqnoenforce", Opt_pqnoenforce),
	135	+ fsparam_flag("qnoenforce", Opt_qnoenforce),
	136	+ fsparam_flag("discard", Opt_discard),
	137	+ fsparam_flag("nodiscard", Opt_nodiscard),
	138	+ fsparam_flag("dax", Opt_dax),
	139	+ fsparam_enum("dax", Opt_dax_enum, dax_param_enums),
	140	+ {}
122	141	};
123		-
124		-
125		-STATIC int
126		-suffix_kstrtoint(const substring_t s, unsigned int base, int res)
127		-{
128		- int last, shift_left_factor = 0, _res;
129		- char *value;
130		- int ret = 0;
131		-
132		- value = match_strdup(s);
133		- if (!value)
134		- return -ENOMEM;
135		-
136		- last = strlen(value) - 1;
137		- if (value[last] == 'K' \|\| value[last] == 'k') {
138		- shift_left_factor = 10;
139		- value[last] = '\0';
140		- }
141		- if (value[last] == 'M' \|\| value[last] == 'm') {
142		- shift_left_factor = 20;
143		- value[last] = '\0';
144		- }
145		- if (value[last] == 'G' \|\| value[last] == 'g') {
146		- shift_left_factor = 30;
147		- value[last] = '\0';
148		- }
149		-
150		- if (kstrtoint(value, base, &_res))
151		- ret = -EINVAL;
152		- kfree(value);
153		- *res = _res << shift_left_factor;
154		- return ret;
155		-}
156		-
157		-/*
158		- * This function fills in xfs_mount_t fields based on mount args.
159		- * Note: the superblock has _not_ yet been read in.
160		- *
161		- * Note that this function leaks the various device name allocations on
162		- * failure. The caller takes care of them.
163		- *
164		- * *sb is const because this is also used to test options on the remount
165		- * path, and we don't want this to have any side effects at remount time.
166		- * Today this function does not change *sb, but just to future-proof...
167		- */
168		-STATIC int
169		-xfs_parseargs(
170		- struct xfs_mount *mp,
171		- char *options)
172		-{
173		- const struct super_block *sb = mp->m_super;
174		- char *p;
175		- substring_t args[MAX_OPT_ARGS];
176		- int dsunit = 0;
177		- int dswidth = 0;
178		- int iosize = 0;
179		- uint8_t iosizelog = 0;
180		-
181		- /*
182		- * set up the mount name first so all the errors will refer to the
183		- * correct device.
184		- */
185		- mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
186		- if (!mp->m_fsname)
187		- return -ENOMEM;
188		- mp->m_fsname_len = strlen(mp->m_fsname) + 1;
189		-
190		- /*
191		- * Copy binary VFS mount flags we are interested in.
192		- */
193		- if (sb_rdonly(sb))
194		- mp->m_flags \|= XFS_MOUNT_RDONLY;
195		- if (sb->s_flags & SB_DIRSYNC)
196		- mp->m_flags \|= XFS_MOUNT_DIRSYNC;
197		- if (sb->s_flags & SB_SYNCHRONOUS)
198		- mp->m_flags \|= XFS_MOUNT_WSYNC;
199		-
200		- /*
201		- * Set some default flags that could be cleared by the mount option
202		- * parsing.
203		- */
204		- mp->m_flags \|= XFS_MOUNT_COMPAT_IOSIZE;
205		-
206		- /*
207		- * These can be overridden by the mount option parsing.
208		- */
209		- mp->m_logbufs = -1;
210		- mp->m_logbsize = -1;
211		-
212		- if (!options)
213		- goto done;
214		-
215		- while ((p = strsep(&options, ",")) != NULL) {
216		- int token;
217		-
218		- if (!*p)
219		- continue;
220		-
221		- token = match_token(p, tokens, args);
222		- switch (token) {
223		- case Opt_logbufs:
224		- if (match_int(args, &mp->m_logbufs))
225		- return -EINVAL;
226		- break;
227		- case Opt_logbsize:
228		- if (suffix_kstrtoint(args, 10, &mp->m_logbsize))
229		- return -EINVAL;
230		- break;
231		- case Opt_logdev:
232		- kfree(mp->m_logname);
233		- mp->m_logname = match_strdup(args);
234		- if (!mp->m_logname)
235		- return -ENOMEM;
236		- break;
237		- case Opt_mtpt:
238		- xfs_warn(mp, "%s option not allowed on this system", p);
239		- return -EINVAL;
240		- case Opt_rtdev:
241		- kfree(mp->m_rtname);
242		- mp->m_rtname = match_strdup(args);
243		- if (!mp->m_rtname)
244		- return -ENOMEM;
245		- break;
246		- case Opt_allocsize:
247		- case Opt_biosize:
248		- if (suffix_kstrtoint(args, 10, &iosize))
249		- return -EINVAL;
250		- iosizelog = ffs(iosize) - 1;
251		- break;
252		- case Opt_grpid:
253		- case Opt_bsdgroups:
254		- mp->m_flags \|= XFS_MOUNT_GRPID;
255		- break;
256		- case Opt_nogrpid:
257		- case Opt_sysvgroups:
258		- mp->m_flags &= ~XFS_MOUNT_GRPID;
259		- break;
260		- case Opt_wsync:
261		- mp->m_flags \|= XFS_MOUNT_WSYNC;
262		- break;
263		- case Opt_norecovery:
264		- mp->m_flags \|= XFS_MOUNT_NORECOVERY;
265		- break;
266		- case Opt_noalign:
267		- mp->m_flags \|= XFS_MOUNT_NOALIGN;
268		- break;
269		- case Opt_swalloc:
270		- mp->m_flags \|= XFS_MOUNT_SWALLOC;
271		- break;
272		- case Opt_sunit:
273		- if (match_int(args, &dsunit))
274		- return -EINVAL;
275		- break;
276		- case Opt_swidth:
277		- if (match_int(args, &dswidth))
278		- return -EINVAL;
279		- break;
280		- case Opt_inode32:
281		- mp->m_flags \|= XFS_MOUNT_SMALL_INUMS;
282		- break;
283		- case Opt_inode64:
284		- mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
285		- break;
286		- case Opt_nouuid:
287		- mp->m_flags \|= XFS_MOUNT_NOUUID;
288		- break;
289		- case Opt_ikeep:
290		- mp->m_flags \|= XFS_MOUNT_IKEEP;
291		- break;
292		- case Opt_noikeep:
293		- mp->m_flags &= ~XFS_MOUNT_IKEEP;
294		- break;
295		- case Opt_largeio:
296		- mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
297		- break;
298		- case Opt_nolargeio:
299		- mp->m_flags \|= XFS_MOUNT_COMPAT_IOSIZE;
300		- break;
301		- case Opt_attr2:
302		- mp->m_flags \|= XFS_MOUNT_ATTR2;
303		- break;
304		- case Opt_noattr2:
305		- mp->m_flags &= ~XFS_MOUNT_ATTR2;
306		- mp->m_flags \|= XFS_MOUNT_NOATTR2;
307		- break;
308		- case Opt_filestreams:
309		- mp->m_flags \|= XFS_MOUNT_FILESTREAMS;
310		- break;
311		- case Opt_noquota:
312		- mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
313		- mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
314		- mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
315		- break;
316		- case Opt_quota:
317		- case Opt_uquota:
318		- case Opt_usrquota:
319		- mp->m_qflags \|= (XFS_UQUOTA_ACCT \| XFS_UQUOTA_ACTIVE \|
320		- XFS_UQUOTA_ENFD);
321		- break;
322		- case Opt_qnoenforce:
323		- case Opt_uqnoenforce:
324		- mp->m_qflags \|= (XFS_UQUOTA_ACCT \| XFS_UQUOTA_ACTIVE);
325		- mp->m_qflags &= ~XFS_UQUOTA_ENFD;
326		- break;
327		- case Opt_pquota:
328		- case Opt_prjquota:
329		- mp->m_qflags \|= (XFS_PQUOTA_ACCT \| XFS_PQUOTA_ACTIVE \|
330		- XFS_PQUOTA_ENFD);
331		- break;
332		- case Opt_pqnoenforce:
333		- mp->m_qflags \|= (XFS_PQUOTA_ACCT \| XFS_PQUOTA_ACTIVE);
334		- mp->m_qflags &= ~XFS_PQUOTA_ENFD;
335		- break;
336		- case Opt_gquota:
337		- case Opt_grpquota:
338		- mp->m_qflags \|= (XFS_GQUOTA_ACCT \| XFS_GQUOTA_ACTIVE \|
339		- XFS_GQUOTA_ENFD);
340		- break;
341		- case Opt_gqnoenforce:
342		- mp->m_qflags \|= (XFS_GQUOTA_ACCT \| XFS_GQUOTA_ACTIVE);
343		- mp->m_qflags &= ~XFS_GQUOTA_ENFD;
344		- break;
345		- case Opt_discard:
346		- mp->m_flags \|= XFS_MOUNT_DISCARD;
347		- break;
348		- case Opt_nodiscard:
349		- mp->m_flags &= ~XFS_MOUNT_DISCARD;
350		- break;
351		-#ifdef CONFIG_FS_DAX
352		- case Opt_dax:
353		- mp->m_flags \|= XFS_MOUNT_DAX;
354		- break;
355		-#endif
356		- default:
357		- xfs_warn(mp, "unknown mount option [%s].", p);
358		- return -EINVAL;
359		- }
360		- }
361		-
362		- /*
363		- * no recovery flag requires a read-only mount
364		- */
365		- if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
366		- !(mp->m_flags & XFS_MOUNT_RDONLY)) {
367		- xfs_warn(mp, "no-recovery mounts must be read-only.");
368		- return -EINVAL;
369		- }
370		-
371		- if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit \|\| dswidth)) {
372		- xfs_warn(mp,
373		- "sunit and swidth options incompatible with the noalign option");
374		- return -EINVAL;
375		- }
376		-
377		-#ifndef CONFIG_XFS_QUOTA
378		- if (XFS_IS_QUOTA_RUNNING(mp)) {
379		- xfs_warn(mp, "quota support not available in this kernel.");
380		- return -EINVAL;
381		- }
382		-#endif
383		-
384		- if ((dsunit && !dswidth) \|\| (!dsunit && dswidth)) {
385		- xfs_warn(mp, "sunit and swidth must be specified together");
386		- return -EINVAL;
387		- }
388		-
389		- if (dsunit && (dswidth % dsunit != 0)) {
390		- xfs_warn(mp,
391		- "stripe width (%d) must be a multiple of the stripe unit (%d)",
392		- dswidth, dsunit);
393		- return -EINVAL;
394		- }
395		-
396		-done:
397		- if (dsunit && !(mp->m_flags & XFS_MOUNT_NOALIGN)) {
398		- /*
399		- * At this point the superblock has not been read
400		- * in, therefore we do not know the block size.
401		- * Before the mount call ends we will convert
402		- * these to FSBs.
403		- */
404		- mp->m_dalign = dsunit;
405		- mp->m_swidth = dswidth;
406		- }
407		-
408		- if (mp->m_logbufs != -1 &&
409		- mp->m_logbufs != 0 &&
410		- (mp->m_logbufs < XLOG_MIN_ICLOGS \|\|
411		- mp->m_logbufs > XLOG_MAX_ICLOGS)) {
412		- xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
413		- mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
414		- return -EINVAL;
415		- }
416		- if (mp->m_logbsize != -1 &&
417		- mp->m_logbsize != 0 &&
418		- (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE \|\|
419		- mp->m_logbsize > XLOG_MAX_RECORD_BSIZE \|\|
420		- !is_power_of_2(mp->m_logbsize))) {
421		- xfs_warn(mp,
422		- "invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
423		- mp->m_logbsize);
424		- return -EINVAL;
425		- }
426		-
427		- if (iosizelog) {
428		- if (iosizelog > XFS_MAX_IO_LOG \|\|
429		- iosizelog < XFS_MIN_IO_LOG) {
430		- xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
431		- iosizelog, XFS_MIN_IO_LOG,
432		- XFS_MAX_IO_LOG);
433		- return -EINVAL;
434		- }
435		-
436		- mp->m_flags \|= XFS_MOUNT_DFLT_IOSIZE;
437		- mp->m_readio_log = iosizelog;
438		- mp->m_writeio_log = iosizelog;
439		- }
440		-
441		- return 0;
442		-}
443	142
444	143	struct proc_xfs_info {
445	144	uint64_t flag;
446	145	char *str;
447	146	};
448	147
449		-STATIC int
450		-xfs_showargs(
451		- struct xfs_mount *mp,
452		- struct seq_file *m)
	148	+static int
	149	+xfs_fs_show_options(
	150	+ struct seq_file *m,
	151	+ struct dentry *root)
453	152	{
454	153	static struct proc_xfs_info xfs_info_set[] = {
455	154	/* the few simple ones we can get from the mount struct */
..	..	@@ -463,30 +162,25 @@
463	162	{ XFS_MOUNT_FILESTREAMS, ",filestreams" },
464	163	{ XFS_MOUNT_GRPID, ",grpid" },
465	164	{ XFS_MOUNT_DISCARD, ",discard" },
466		- { XFS_MOUNT_SMALL_INUMS, ",inode32" },
467		- { XFS_MOUNT_DAX, ",dax" },
	165	+ { XFS_MOUNT_LARGEIO, ",largeio" },
	166	+ { XFS_MOUNT_DAX_ALWAYS, ",dax=always" },
	167	+ { XFS_MOUNT_DAX_NEVER, ",dax=never" },
468	168	{ 0, NULL }
469	169	};
470		- static struct proc_xfs_info xfs_info_unset[] = {
471		- /* the few simple ones we can get from the mount struct */
472		- { XFS_MOUNT_COMPAT_IOSIZE, ",largeio" },
473		- { XFS_MOUNT_SMALL_INUMS, ",inode64" },
474		- { 0, NULL }
475		- };
	170	+ struct xfs_mount *mp = XFS_M(root->d_sb);
476	171	struct proc_xfs_info *xfs_infop;
477	172
478	173	for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) {
479	174	if (mp->m_flags & xfs_infop->flag)
480	175	seq_puts(m, xfs_infop->str);
481	176	}
482		- for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) {
483		- if (!(mp->m_flags & xfs_infop->flag))
484		- seq_puts(m, xfs_infop->str);
485		- }
486	177
487		- if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)
	178	+ seq_printf(m, ",inode%d",
	179	+ (mp->m_flags & XFS_MOUNT_SMALL_INUMS) ? 32 : 64);
	180	+
	181	+ if (mp->m_flags & XFS_MOUNT_ALLOCSIZE)
488	182	seq_printf(m, ",allocsize=%dk",
489		- (int)(1 << mp->m_writeio_log) >> 10);
	183	+ (1 << mp->m_allocsize_log) >> 10);
490	184
491	185	if (mp->m_logbufs > 0)
492	186	seq_printf(m, ",logbufs=%d", mp->m_logbufs);
..	..	@@ -505,10 +199,12 @@
505	199	seq_printf(m, ",swidth=%d",
506	200	(int)XFS_FSB_TO_BB(mp, mp->m_swidth));
507	201
508		- if (mp->m_qflags & (XFS_UQUOTA_ACCT\|XFS_UQUOTA_ENFD))
509		- seq_puts(m, ",usrquota");
510		- else if (mp->m_qflags & XFS_UQUOTA_ACCT)
511		- seq_puts(m, ",uqnoenforce");
	202	+ if (mp->m_qflags & XFS_UQUOTA_ACCT) {
	203	+ if (mp->m_qflags & XFS_UQUOTA_ENFD)
	204	+ seq_puts(m, ",usrquota");
	205	+ else
	206	+ seq_puts(m, ",uqnoenforce");
	207	+ }
512	208
513	209	if (mp->m_qflags & XFS_PQUOTA_ACCT) {
514	210	if (mp->m_qflags & XFS_PQUOTA_ENFD)
..	..	@@ -527,39 +223,6 @@
527	223	seq_puts(m, ",noquota");
528	224
529	225	return 0;
530		-}
531		-static uint64_t
532		-xfs_max_file_offset(
533		- unsigned int blockshift)
534		-{
535		- unsigned int pagefactor = 1;
536		- unsigned int bitshift = BITS_PER_LONG - 1;
537		-
538		- /* Figure out maximum filesize, on Linux this can depend on
539		- * the filesystem blocksize (on 32 bit platforms).
540		- * __block_write_begin does this in an [unsigned] long...
541		- * page->index << (PAGE_SHIFT - bbits)
542		- * So, for page sized blocks (4K on 32 bit platforms),
543		- * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is
544		- * (((u64)PAGE_SIZE << (BITS_PER_LONG-1))-1)
545		- * but for smaller blocksizes it is less (bbits = log2 bsize).
546		- * Note1: get_block_t takes a long (implicit cast from above)
547		- * Note2: The Large Block Device (LBD and HAVE_SECTOR_T) patch
548		- * can optionally convert the [unsigned] long from above into
549		- * an [unsigned] long long.
550		- */
551		-
552		-#if BITS_PER_LONG == 32
553		-# if defined(CONFIG_LBDAF)
554		- ASSERT(sizeof(sector_t) == 8);
555		- pagefactor = PAGE_SIZE;
556		- bitshift = BITS_PER_LONG;
557		-# else
558		- pagefactor = PAGE_SIZE >> (PAGE_SHIFT - blockshift);
559		-# endif
560		-#endif
561		-
562		- return (((uint64_t)pagefactor) << bitshift) - 1;
563	226	}
564	227
565	228	/*
..	..	@@ -593,7 +256,7 @@
593	256	* Calculate how much should be reserved for inodes to meet
594	257	* the max inode percentage. Used only for inode32.
595	258	*/
596		- if (mp->m_maxicount) {
	259	+ if (M_IGEO(mp)->maxicount) {
597	260	uint64_t icount;
598	261
599	262	icount = sbp->sb_dblocks * sbp->sb_imax_pct;
..	..	@@ -606,7 +269,7 @@
606	269	}
607	270
608	271	/* Get the last possible inode in the filesystem */
609		- agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
	272	+ agino = XFS_AGB_TO_AGINO(mp, sbp->sb_agblocks - 1);
610	273	ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);
611	274
612	275	/*
..	..	@@ -679,7 +342,7 @@
679	342	xfs_blkdev_issue_flush(
680	343	xfs_buftarg_t *buftarg)
681	344	{
682		- blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS, NULL);
	345	+ blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS);
683	346	}
684	347
685	348	STATIC void
..	..	@@ -832,43 +495,33 @@
832	495	struct xfs_mount *mp)
833	496	{
834	497	mp->m_buf_workqueue = alloc_workqueue("xfs-buf/%s",
835		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 1, mp->m_fsname);
	498	+ WQ_MEM_RECLAIM\|WQ_FREEZABLE, 1, mp->m_super->s_id);
836	499	if (!mp->m_buf_workqueue)
837	500	goto out;
838	501
839		- mp->m_data_workqueue = alloc_workqueue("xfs-data/%s",
840		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_fsname);
841		- if (!mp->m_data_workqueue)
	502	+ mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s",
	503	+ WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_super->s_id);
	504	+ if (!mp->m_unwritten_workqueue)
842	505	goto out_destroy_buf;
843	506
844		- mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s",
845		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_fsname);
846		- if (!mp->m_unwritten_workqueue)
847		- goto out_destroy_data_iodone_queue;
848		-
849	507	mp->m_cil_workqueue = alloc_workqueue("xfs-cil/%s",
850		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_fsname);
	508	+ WQ_MEM_RECLAIM \| WQ_FREEZABLE \| WQ_UNBOUND,
	509	+ 0, mp->m_super->s_id);
851	510	if (!mp->m_cil_workqueue)
852	511	goto out_destroy_unwritten;
853	512
854	513	mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s",
855		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_fsname);
	514	+ WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_super->s_id);
856	515	if (!mp->m_reclaim_workqueue)
857	516	goto out_destroy_cil;
858	517
859		- mp->m_log_workqueue = alloc_workqueue("xfs-log/%s",
860		- WQ_MEM_RECLAIM\|WQ_FREEZABLE\|WQ_HIGHPRI, 0,
861		- mp->m_fsname);
862		- if (!mp->m_log_workqueue)
	518	+ mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
	519	+ WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_super->s_id);
	520	+ if (!mp->m_eofblocks_workqueue)
863	521	goto out_destroy_reclaim;
864	522
865		- mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
866		- WQ_MEM_RECLAIM\|WQ_FREEZABLE, 0, mp->m_fsname);
867		- if (!mp->m_eofblocks_workqueue)
868		- goto out_destroy_log;
869		-
870	523	mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", WQ_FREEZABLE, 0,
871		- mp->m_fsname);
	524	+ mp->m_super->s_id);
872	525	if (!mp->m_sync_workqueue)
873	526	goto out_destroy_eofb;
874	527
..	..	@@ -876,16 +529,12 @@
876	529
877	530	out_destroy_eofb:
878	531	destroy_workqueue(mp->m_eofblocks_workqueue);
879		-out_destroy_log:
880		- destroy_workqueue(mp->m_log_workqueue);
881	532	out_destroy_reclaim:
882	533	destroy_workqueue(mp->m_reclaim_workqueue);
883	534	out_destroy_cil:
884	535	destroy_workqueue(mp->m_cil_workqueue);
885	536	out_destroy_unwritten:
886	537	destroy_workqueue(mp->m_unwritten_workqueue);
887		-out_destroy_data_iodone_queue:
888		- destroy_workqueue(mp->m_data_workqueue);
889	538	out_destroy_buf:
890	539	destroy_workqueue(mp->m_buf_workqueue);
891	540	out:
..	..	@@ -898,12 +547,24 @@
898	547	{
899	548	destroy_workqueue(mp->m_sync_workqueue);
900	549	destroy_workqueue(mp->m_eofblocks_workqueue);
901		- destroy_workqueue(mp->m_log_workqueue);
902	550	destroy_workqueue(mp->m_reclaim_workqueue);
903	551	destroy_workqueue(mp->m_cil_workqueue);
904		- destroy_workqueue(mp->m_data_workqueue);
905	552	destroy_workqueue(mp->m_unwritten_workqueue);
906	553	destroy_workqueue(mp->m_buf_workqueue);
	554	+}
	555	+
	556	+static void
	557	+xfs_flush_inodes_worker(
	558	+ struct work_struct *work)
	559	+{
	560	+ struct xfs_mount *mp = container_of(work, struct xfs_mount,
	561	+ m_flush_inodes_work);
	562	+ struct super_block *sb = mp->m_super;
	563	+
	564	+ if (down_read_trylock(&sb->s_umount)) {
	565	+ sync_inodes_sb(sb);
	566	+ up_read(&sb->s_umount);
	567	+ }
907	568	}
908	569
909	570	/*
..	..	@@ -916,12 +577,15 @@
916	577	xfs_flush_inodes(
917	578	struct xfs_mount *mp)
918	579	{
919		- struct super_block *sb = mp->m_super;
	580	+ /*
	581	+ * If flush_work() returns true then that means we waited for a flush
	582	+ * which was already in progress. Don't bother running another scan.
	583	+ */
	584	+ if (flush_work(&mp->m_flush_inodes_work))
	585	+ return;
920	586
921		- if (down_read_trylock(&sb->s_umount)) {
922		- sync_inodes_sb(sb);
923		- up_read(&sb->s_umount);
924		- }
	587	+ queue_work(mp->m_sync_workqueue, &mp->m_flush_inodes_work);
	588	+ flush_work(&mp->m_flush_inodes_work);
925	589	}
926	590
927	591	/* Catch misguided souls that try to use this interface on XFS */
..	..	@@ -932,6 +596,32 @@
932	596	BUG();
933	597	return NULL;
934	598	}
	599	+
	600	+#ifdef DEBUG
	601	+static void
	602	+xfs_check_delalloc(
	603	+ struct xfs_inode *ip,
	604	+ int whichfork)
	605	+{
	606	+ struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
	607	+ struct xfs_bmbt_irec got;
	608	+ struct xfs_iext_cursor icur;
	609	+
	610	+ if (!ifp \|\| !xfs_iext_lookup_extent(ip, ifp, 0, &icur, &got))
	611	+ return;
	612	+ do {
	613	+ if (isnullstartblock(got.br_startblock)) {
	614	+ xfs_warn(ip->i_mount,
	615	+ "ino %llx %s fork has delalloc extent at [0x%llx:0x%llx]",
	616	+ ip->i_ino,
	617	+ whichfork == XFS_DATA_FORK ? "data" : "cow",
	618	+ got.br_startoff, got.br_blockcount);
	619	+ }
	620	+ } while (xfs_iext_next_extent(ifp, &icur, &got));
	621	+}
	622	+#else
	623	+#define xfs_check_delalloc(ip, whichfork) do { } while (0)
	624	+#endif
935	625
936	626	/*
937	627	* Now that the generic code is guaranteed not to be accessing
..	..	@@ -951,7 +641,12 @@
951	641
952	642	xfs_inactive(ip);
953	643
954		- ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) \|\| ip->i_delayed_blks == 0);
	644	+ if (!XFS_FORCED_SHUTDOWN(ip->i_mount) && ip->i_delayed_blks) {
	645	+ xfs_check_delalloc(ip, XFS_DATA_FORK);
	646	+ xfs_check_delalloc(ip, XFS_COW_FORK);
	647	+ ASSERT(0);
	648	+ }
	649	+
955	650	XFS_STATS_INC(ip->i_mount, vn_reclaim);
956	651
957	652	/*
..	..	@@ -961,11 +656,11 @@
961	656	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM));
962	657
963	658	/*
964		- * We always use background reclaim here because even if the
965		- * inode is clean, it still may be under IO and hence we have
966		- * to take the flush lock. The background reclaim path handles
967		- * this more efficiently than we can here, so simply let background
968		- * reclaim tear down all inodes.
	659	+ * We always use background reclaim here because even if the inode is
	660	+ * clean, it still may be under IO and hence we have wait for IO
	661	+ * completion to occur before we can reclaim the inode. The background
	662	+ * reclaim path handles this more efficiently than we can here, so
	663	+ * simply let background reclaim tear down all inodes.
969	664	*/
970	665	xfs_inode_set_reclaim_tag(ip);
971	666	}
..	..	@@ -1044,16 +739,16 @@
1044	739	return 0;
1045	740	}
1046	741
1047		- return generic_drop_inode(inode) \|\| (ip->i_flags & XFS_IDONTCACHE);
	742	+ return generic_drop_inode(inode);
1048	743	}
1049	744
1050		-STATIC void
1051		-xfs_free_fsname(
	745	+static void
	746	+xfs_mount_free(
1052	747	struct xfs_mount *mp)
1053	748	{
1054		- kfree(mp->m_fsname);
1055	749	kfree(mp->m_rtname);
1056	750	kfree(mp->m_logname);
	751	+ kmem_free(mp);
1057	752	}
1058	753
1059	754	STATIC int
..	..	@@ -1062,6 +757,7 @@
1062	757	int wait)
1063	758	{
1064	759	struct xfs_mount *mp = XFS_M(sb);
	760	+ int error;
1065	761
1066	762	/*
1067	763	* Doing anything during the async pass would be counterproductive.
..	..	@@ -1069,7 +765,10 @@
1069	765	if (!wait)
1070	766	return 0;
1071	767
1072		- xfs_log_force(mp, XFS_LOG_SYNC);
	768	+ error = xfs_log_force(mp, XFS_LOG_SYNC);
	769	+ if (error)
	770	+ return error;
	771	+
1073	772	if (laptop_mode) {
1074	773	/*
1075	774	* The disk must be active because we're syncing.
..	..	@@ -1097,12 +796,11 @@
1097	796	xfs_extlen_t lsize;
1098	797	int64_t ffree;
1099	798
1100		- statp->f_type = XFS_SB_MAGIC;
	799	+ statp->f_type = XFS_SUPER_MAGIC;
1101	800	statp->f_namelen = MAXNAMELEN - 1;
1102	801
1103	802	id = huge_encode_dev(mp->m_ddev_targp->bt_dev);
1104		- statp->f_fsid.val[0] = (u32)id;
1105		- statp->f_fsid.val[1] = (u32)(id >> 32);
	803	+ statp->f_fsid = u64_to_fsid(id);
1106	804
1107	805	icount = percpu_counter_sum(&mp->m_icount);
1108	806	ifree = percpu_counter_sum(&mp->m_ifree);
..	..	@@ -1114,15 +812,16 @@
1114	812	statp->f_blocks = sbp->sb_dblocks - lsize;
1115	813	spin_unlock(&mp->m_sb_lock);
1116	814
1117		- statp->f_bfree = fdblocks - mp->m_alloc_set_aside;
	815	+ /* make sure statp->f_bfree does not underflow */
	816	+ statp->f_bfree = max_t(int64_t, fdblocks - mp->m_alloc_set_aside, 0);
1118	817	statp->f_bavail = statp->f_bfree;
1119	818
1120		- fakeinos = statp->f_bfree << sbp->sb_inopblog;
	819	+ fakeinos = XFS_FSB_TO_INO(mp, statp->f_bfree);
1121	820	statp->f_files = min(icount + fakeinos, (uint64_t)XFS_MAXINUMBER);
1122		- if (mp->m_maxicount)
	821	+ if (M_IGEO(mp)->maxicount)
1123	822	statp->f_files = min_t(typeof(statp->f_files),
1124	823	statp->f_files,
1125		- mp->m_maxicount);
	824	+ M_IGEO(mp)->maxicount);
1126	825
1127	826	/* If sb_icount overshot maxicount, report actual allocation */
1128	827	statp->f_files = max_t(typeof(statp->f_files),
..	..	@@ -1180,8 +879,10 @@
1180	879	* there is no log replay required to write the inodes to disk - this is the
1181	880	* primary difference between a sync and a quiesce.
1182	881	*
1183		- * Note: xfs_log_quiesce() stops background log work - the callers must ensure
1184		- * it is started again when appropriate.
	882	+ * We cancel log work early here to ensure all transactions the log worker may
	883	+ * run have finished before we clean up and log the superblock and write an
	884	+ * unmount record. The unfreeze process is responsible for restarting the log
	885	+ * worker correctly.
1185	886	*/
1186	887	void
1187	888	xfs_quiesce_attr(
..	..	@@ -1189,204 +890,18 @@
1189	890	{
1190	891	int error = 0;
1191	892
1192		- /* wait for all modifications to complete */
1193		- while (atomic_read(&mp->m_active_trans) > 0)
1194		- delay(100);
	893	+ cancel_delayed_work_sync(&mp->m_log->l_work);
1195	894
1196	895	/* force the log to unpin objects from the now complete transactions */
1197	896	xfs_log_force(mp, XFS_LOG_SYNC);
1198	897
1199		- /* reclaim inodes to do any IO before the freeze completes */
1200		- xfs_reclaim_inodes(mp, 0);
1201		- xfs_reclaim_inodes(mp, SYNC_WAIT);
1202	898
1203	899	/* Push the superblock and write an unmount record */
1204	900	error = xfs_log_sbcount(mp);
1205	901	if (error)
1206	902	xfs_warn(mp, "xfs_attr_quiesce: failed to log sb changes. "
1207	903	"Frozen image may not be consistent.");
1208		- /*
1209		- * Just warn here till VFS can correctly support
1210		- * read-only remount without racing.
1211		- */
1212		- WARN_ON(atomic_read(&mp->m_active_trans) != 0);
1213		-
1214	904	xfs_log_quiesce(mp);
1215		-}
1216		-
1217		-STATIC int
1218		-xfs_test_remount_options(
1219		- struct super_block *sb,
1220		- char *options)
1221		-{
1222		- int error = 0;
1223		- struct xfs_mount *tmp_mp;
1224		-
1225		- tmp_mp = kmem_zalloc(sizeof(*tmp_mp), KM_MAYFAIL);
1226		- if (!tmp_mp)
1227		- return -ENOMEM;
1228		-
1229		- tmp_mp->m_super = sb;
1230		- error = xfs_parseargs(tmp_mp, options);
1231		- xfs_free_fsname(tmp_mp);
1232		- kmem_free(tmp_mp);
1233		-
1234		- return error;
1235		-}
1236		-
1237		-STATIC int
1238		-xfs_fs_remount(
1239		- struct super_block *sb,
1240		- int *flags,
1241		- char *options)
1242		-{
1243		- struct xfs_mount *mp = XFS_M(sb);
1244		- xfs_sb_t *sbp = &mp->m_sb;
1245		- substring_t args[MAX_OPT_ARGS];
1246		- char *p;
1247		- int error;
1248		-
1249		- /* First, check for complete junk; i.e. invalid options */
1250		- error = xfs_test_remount_options(sb, options);
1251		- if (error)
1252		- return error;
1253		-
1254		- sync_filesystem(sb);
1255		- while ((p = strsep(&options, ",")) != NULL) {
1256		- int token;
1257		-
1258		- if (!*p)
1259		- continue;
1260		-
1261		- token = match_token(p, tokens, args);
1262		- switch (token) {
1263		- case Opt_inode64:
1264		- mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
1265		- mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
1266		- break;
1267		- case Opt_inode32:
1268		- mp->m_flags \|= XFS_MOUNT_SMALL_INUMS;
1269		- mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
1270		- break;
1271		- default:
1272		- /*
1273		- * Logically we would return an error here to prevent
1274		- * users from believing they might have changed
1275		- * mount options using remount which can't be changed.
1276		- *
1277		- * But unfortunately mount(8) adds all options from
1278		- * mtab and fstab to the mount arguments in some cases
1279		- * so we can't blindly reject options, but have to
1280		- * check for each specified option if it actually
1281		- * differs from the currently set option and only
1282		- * reject it if that's the case.
1283		- *
1284		- * Until that is implemented we return success for
1285		- * every remount request, and silently ignore all
1286		- * options that we can't actually change.
1287		- */
1288		-#if 0
1289		- xfs_info(mp,
1290		- "mount option \"%s\" not supported for remount", p);
1291		- return -EINVAL;
1292		-#else
1293		- break;
1294		-#endif
1295		- }
1296		- }
1297		-
1298		- /* ro -> rw */
1299		- if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & SB_RDONLY)) {
1300		- if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
1301		- xfs_warn(mp,
1302		- "ro->rw transition prohibited on norecovery mount");
1303		- return -EINVAL;
1304		- }
1305		-
1306		- if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
1307		- xfs_sb_has_ro_compat_feature(sbp,
1308		- XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
1309		- xfs_warn(mp,
1310		-"ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
1311		- (sbp->sb_features_ro_compat &
1312		- XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
1313		- return -EINVAL;
1314		- }
1315		-
1316		- mp->m_flags &= ~XFS_MOUNT_RDONLY;
1317		-
1318		- /*
1319		- * If this is the first remount to writeable state we
1320		- * might have some superblock changes to update.
1321		- */
1322		- if (mp->m_update_sb) {
1323		- error = xfs_sync_sb(mp, false);
1324		- if (error) {
1325		- xfs_warn(mp, "failed to write sb changes");
1326		- return error;
1327		- }
1328		- mp->m_update_sb = false;
1329		- }
1330		-
1331		- /*
1332		- * Fill out the reserve pool if it is empty. Use the stashed
1333		- * value if it is non-zero, otherwise go with the default.
1334		- */
1335		- xfs_restore_resvblks(mp);
1336		- xfs_log_work_queue(mp);
1337		-
1338		- /* Recover any CoW blocks that never got remapped. */
1339		- error = xfs_reflink_recover_cow(mp);
1340		- if (error) {
1341		- xfs_err(mp,
1342		- "Error %d recovering leftover CoW allocations.", error);
1343		- xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
1344		- return error;
1345		- }
1346		- xfs_icache_enable_reclaim(mp);
1347		-
1348		- /* Create the per-AG metadata reservation pool .*/
1349		- error = xfs_fs_reserve_ag_blocks(mp);
1350		- if (error && error != -ENOSPC)
1351		- return error;
1352		- }
1353		-
1354		- /* rw -> ro */
1355		- if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & SB_RDONLY)) {
1356		- /*
1357		- * Cancel background eofb scanning so it cannot race with the
1358		- * final log force+buftarg wait and deadlock the remount.
1359		- */
1360		- xfs_icache_disable_reclaim(mp);
1361		-
1362		- /* Get rid of any leftover CoW reservations... */
1363		- error = xfs_icache_free_cowblocks(mp, NULL);
1364		- if (error) {
1365		- xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
1366		- return error;
1367		- }
1368		-
1369		- /* Free the per-AG metadata reservation pool. */
1370		- error = xfs_fs_unreserve_ag_blocks(mp);
1371		- if (error) {
1372		- xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
1373		- return error;
1374		- }
1375		-
1376		- /*
1377		- * Before we sync the metadata, we need to free up the reserve
1378		- * block pool so that the used block count in the superblock on
1379		- * disk is correct at the end of the remount. Stash the current
1380		- * reserve pool size so that if we get remounted rw, we can
1381		- * return it to the same size.
1382		- */
1383		- xfs_save_resvblks(mp);
1384		-
1385		- xfs_quiesce_attr(mp);
1386		- mp->m_flags \|= XFS_MOUNT_RDONLY;
1387		- }
1388		-
1389		- return 0;
1390	905	}
1391	906
1392	907	/*
..	..	@@ -1400,11 +915,21 @@
1400	915	struct super_block *sb)
1401	916	{
1402	917	struct xfs_mount *mp = XFS_M(sb);
	918	+ unsigned int flags;
	919	+ int ret;
1403	920
1404		- xfs_icache_disable_reclaim(mp);
	921	+ /*
	922	+ * The filesystem is now frozen far enough that memory reclaim
	923	+ * cannot safely operate on the filesystem. Hence we need to
	924	+ * set a GFP_NOFS context here to avoid recursion deadlocks.
	925	+ */
	926	+ flags = memalloc_nofs_save();
	927	+ xfs_stop_block_reaping(mp);
1405	928	xfs_save_resvblks(mp);
1406	929	xfs_quiesce_attr(mp);
1407		- return xfs_sync_sb(mp, true);
	930	+ ret = xfs_sync_sb(mp, true);
	931	+ memalloc_nofs_restore(flags);
	932	+ return ret;
1408	933	}
1409	934
1410	935	STATIC int
..	..	@@ -1415,16 +940,8 @@
1415	940
1416	941	xfs_restore_resvblks(mp);
1417	942	xfs_log_work_queue(mp);
1418		- xfs_icache_enable_reclaim(mp);
	943	+ xfs_start_block_reaping(mp);
1419	944	return 0;
1420		-}
1421		-
1422		-STATIC int
1423		-xfs_fs_show_options(
1424		- struct seq_file *m,
1425		- struct dentry *root)
1426		-{
1427		- return xfs_showargs(XFS_M(root->d_sb), m);
1428	945	}
1429	946
1430	947	/*
..	..	@@ -1513,8 +1030,14 @@
1513	1030	if (error)
1514	1031	goto free_ifree;
1515	1032
	1033	+ error = percpu_counter_init(&mp->m_delalloc_blks, 0, GFP_KERNEL);
	1034	+ if (error)
	1035	+ goto free_fdblocks;
	1036	+
1516	1037	return 0;
1517	1038
	1039	+free_fdblocks:
	1040	+ percpu_counter_destroy(&mp->m_fdblocks);
1518	1041	free_ifree:
1519	1042	percpu_counter_destroy(&mp->m_ifree);
1520	1043	free_icount:
..	..	@@ -1538,228 +1061,12 @@
1538	1061	percpu_counter_destroy(&mp->m_icount);
1539	1062	percpu_counter_destroy(&mp->m_ifree);
1540	1063	percpu_counter_destroy(&mp->m_fdblocks);
	1064	+ ASSERT(XFS_FORCED_SHUTDOWN(mp) \|\|
	1065	+ percpu_counter_sum(&mp->m_delalloc_blks) == 0);
	1066	+ percpu_counter_destroy(&mp->m_delalloc_blks);
1541	1067	}
1542	1068
1543		-static struct xfs_mount *
1544		-xfs_mount_alloc(
1545		- struct super_block *sb)
1546		-{
1547		- struct xfs_mount *mp;
1548		-
1549		- mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
1550		- if (!mp)
1551		- return NULL;
1552		-
1553		- mp->m_super = sb;
1554		- spin_lock_init(&mp->m_sb_lock);
1555		- spin_lock_init(&mp->m_agirotor_lock);
1556		- INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
1557		- spin_lock_init(&mp->m_perag_lock);
1558		- mutex_init(&mp->m_growlock);
1559		- atomic_set(&mp->m_active_trans, 0);
1560		- INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
1561		- INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
1562		- INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
1563		- mp->m_kobj.kobject.kset = xfs_kset;
1564		- /*
1565		- * We don't create the finobt per-ag space reservation until after log
1566		- * recovery, so we must set this to true so that an ifree transaction
1567		- * started during log recovery will not depend on space reservations
1568		- * for finobt expansion.
1569		- */
1570		- mp->m_finobt_nores = true;
1571		- return mp;
1572		-}
1573		-
1574		-
1575		-STATIC int
1576		-xfs_fs_fill_super(
1577		- struct super_block *sb,
1578		- void *data,
1579		- int silent)
1580		-{
1581		- struct inode *root;
1582		- struct xfs_mount *mp = NULL;
1583		- int flags = 0, error = -ENOMEM;
1584		-
1585		- /*
1586		- * allocate mp and do all low-level struct initializations before we
1587		- * attach it to the super
1588		- */
1589		- mp = xfs_mount_alloc(sb);
1590		- if (!mp)
1591		- goto out;
1592		- sb->s_fs_info = mp;
1593		-
1594		- error = xfs_parseargs(mp, (char *)data);
1595		- if (error)
1596		- goto out_free_fsname;
1597		-
1598		- sb_min_blocksize(sb, BBSIZE);
1599		- sb->s_xattr = xfs_xattr_handlers;
1600		- sb->s_export_op = &xfs_export_operations;
1601		-#ifdef CONFIG_XFS_QUOTA
1602		- sb->s_qcop = &xfs_quotactl_operations;
1603		- sb->s_quota_types = QTYPE_MASK_USR \| QTYPE_MASK_GRP \| QTYPE_MASK_PRJ;
1604		-#endif
1605		- sb->s_op = &xfs_super_operations;
1606		-
1607		- /*
1608		- * Delay mount work if the debug hook is set. This is debug
1609		- * instrumention to coordinate simulation of xfs mount failures with
1610		- * VFS superblock operations
1611		- */
1612		- if (xfs_globals.mount_delay) {
1613		- xfs_notice(mp, "Delaying mount for %d seconds.",
1614		- xfs_globals.mount_delay);
1615		- msleep(xfs_globals.mount_delay * 1000);
1616		- }
1617		-
1618		- if (silent)
1619		- flags \|= XFS_MFSI_QUIET;
1620		-
1621		- error = xfs_open_devices(mp);
1622		- if (error)
1623		- goto out_free_fsname;
1624		-
1625		- error = xfs_init_mount_workqueues(mp);
1626		- if (error)
1627		- goto out_close_devices;
1628		-
1629		- error = xfs_init_percpu_counters(mp);
1630		- if (error)
1631		- goto out_destroy_workqueues;
1632		-
1633		- /* Allocate stats memory before we do operations that might use it */
1634		- mp->m_stats.xs_stats = alloc_percpu(struct xfsstats);
1635		- if (!mp->m_stats.xs_stats) {
1636		- error = -ENOMEM;
1637		- goto out_destroy_counters;
1638		- }
1639		-
1640		- error = xfs_readsb(mp, flags);
1641		- if (error)
1642		- goto out_free_stats;
1643		-
1644		- error = xfs_finish_flags(mp);
1645		- if (error)
1646		- goto out_free_sb;
1647		-
1648		- error = xfs_setup_devices(mp);
1649		- if (error)
1650		- goto out_free_sb;
1651		-
1652		- error = xfs_filestream_mount(mp);
1653		- if (error)
1654		- goto out_free_sb;
1655		-
1656		- /*
1657		- * we must configure the block size in the superblock before we run the
1658		- * full mount process as the mount process can lookup and cache inodes.
1659		- */
1660		- sb->s_magic = XFS_SB_MAGIC;
1661		- sb->s_blocksize = mp->m_sb.sb_blocksize;
1662		- sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1;
1663		- sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits);
1664		- sb->s_max_links = XFS_MAXLINK;
1665		- sb->s_time_gran = 1;
1666		- set_posix_acl_flag(sb);
1667		-
1668		- /* version 5 superblocks support inode version counters. */
1669		- if (XFS_SB_VERSION_NUM(&mp->m_sb) == XFS_SB_VERSION_5)
1670		- sb->s_flags \|= SB_I_VERSION;
1671		-
1672		- if (mp->m_flags & XFS_MOUNT_DAX) {
1673		- bool rtdev_is_dax = false, datadev_is_dax;
1674		-
1675		- xfs_warn(mp,
1676		- "DAX enabled. Warning: EXPERIMENTAL, use at your own risk");
1677		-
1678		- datadev_is_dax = bdev_dax_supported(mp->m_ddev_targp->bt_bdev,
1679		- sb->s_blocksize);
1680		- if (mp->m_rtdev_targp)
1681		- rtdev_is_dax = bdev_dax_supported(
1682		- mp->m_rtdev_targp->bt_bdev, sb->s_blocksize);
1683		- if (!rtdev_is_dax && !datadev_is_dax) {
1684		- xfs_alert(mp,
1685		- "DAX unsupported by block device. Turning off DAX.");
1686		- mp->m_flags &= ~XFS_MOUNT_DAX;
1687		- }
1688		- if (xfs_sb_version_hasreflink(&mp->m_sb)) {
1689		- xfs_alert(mp,
1690		- "DAX and reflink cannot be used together!");
1691		- error = -EINVAL;
1692		- goto out_filestream_unmount;
1693		- }
1694		- }
1695		-
1696		- if (mp->m_flags & XFS_MOUNT_DISCARD) {
1697		- struct request_queue *q = bdev_get_queue(sb->s_bdev);
1698		-
1699		- if (!blk_queue_discard(q)) {
1700		- xfs_warn(mp, "mounting with \"discard\" option, but "
1701		- "the device does not support discard");
1702		- mp->m_flags &= ~XFS_MOUNT_DISCARD;
1703		- }
1704		- }
1705		-
1706		- if (xfs_sb_version_hasreflink(&mp->m_sb) && mp->m_sb.sb_rblocks) {
1707		- xfs_alert(mp,
1708		- "reflink not compatible with realtime device!");
1709		- error = -EINVAL;
1710		- goto out_filestream_unmount;
1711		- }
1712		-
1713		- if (xfs_sb_version_hasrmapbt(&mp->m_sb) && mp->m_sb.sb_rblocks) {
1714		- xfs_alert(mp,
1715		- "reverse mapping btree not compatible with realtime device!");
1716		- error = -EINVAL;
1717		- goto out_filestream_unmount;
1718		- }
1719		-
1720		- error = xfs_mountfs(mp);
1721		- if (error)
1722		- goto out_filestream_unmount;
1723		-
1724		- root = igrab(VFS_I(mp->m_rootip));
1725		- if (!root) {
1726		- error = -ENOENT;
1727		- goto out_unmount;
1728		- }
1729		- sb->s_root = d_make_root(root);
1730		- if (!sb->s_root) {
1731		- error = -ENOMEM;
1732		- goto out_unmount;
1733		- }
1734		-
1735		- return 0;
1736		-
1737		- out_filestream_unmount:
1738		- xfs_filestream_unmount(mp);
1739		- out_free_sb:
1740		- xfs_freesb(mp);
1741		- out_free_stats:
1742		- free_percpu(mp->m_stats.xs_stats);
1743		- out_destroy_counters:
1744		- xfs_destroy_percpu_counters(mp);
1745		- out_destroy_workqueues:
1746		- xfs_destroy_mount_workqueues(mp);
1747		- out_close_devices:
1748		- xfs_close_devices(mp);
1749		- out_free_fsname:
1750		- sb->s_fs_info = NULL;
1751		- xfs_free_fsname(mp);
1752		- kfree(mp);
1753		- out:
1754		- return error;
1755		-
1756		- out_unmount:
1757		- xfs_filestream_unmount(mp);
1758		- xfs_unmountfs(mp);
1759		- goto out_free_sb;
1760		-}
1761		-
1762		-STATIC void
	1069	+static void
1763	1070	xfs_fs_put_super(
1764	1071	struct super_block *sb)
1765	1072	{
..	..	@@ -1780,18 +1087,7 @@
1780	1087	xfs_close_devices(mp);
1781	1088
1782	1089	sb->s_fs_info = NULL;
1783		- xfs_free_fsname(mp);
1784		- kfree(mp);
1785		-}
1786		-
1787		-STATIC struct dentry *
1788		-xfs_fs_mount(
1789		- struct file_system_type *fs_type,
1790		- int flags,
1791		- const char *dev_name,
1792		- void *data)
1793		-{
1794		- return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super);
	1090	+ xfs_mount_free(mp);
1795	1091	}
1796	1092
1797	1093	static long
..	..	@@ -1823,16 +1119,790 @@
1823	1119	.freeze_fs = xfs_fs_freeze,
1824	1120	.unfreeze_fs = xfs_fs_unfreeze,
1825	1121	.statfs = xfs_fs_statfs,
1826		- .remount_fs = xfs_fs_remount,
1827	1122	.show_options = xfs_fs_show_options,
1828	1123	.nr_cached_objects = xfs_fs_nr_cached_objects,
1829	1124	.free_cached_objects = xfs_fs_free_cached_objects,
1830	1125	};
1831	1126
	1127	+static int
	1128	+suffix_kstrtoint(
	1129	+ const char *s,
	1130	+ unsigned int base,
	1131	+ int *res)
	1132	+{
	1133	+ int last, shift_left_factor = 0, _res;
	1134	+ char *value;
	1135	+ int ret = 0;
	1136	+
	1137	+ value = kstrdup(s, GFP_KERNEL);
	1138	+ if (!value)
	1139	+ return -ENOMEM;
	1140	+
	1141	+ last = strlen(value) - 1;
	1142	+ if (value[last] == 'K' \|\| value[last] == 'k') {
	1143	+ shift_left_factor = 10;
	1144	+ value[last] = '\0';
	1145	+ }
	1146	+ if (value[last] == 'M' \|\| value[last] == 'm') {
	1147	+ shift_left_factor = 20;
	1148	+ value[last] = '\0';
	1149	+ }
	1150	+ if (value[last] == 'G' \|\| value[last] == 'g') {
	1151	+ shift_left_factor = 30;
	1152	+ value[last] = '\0';
	1153	+ }
	1154	+
	1155	+ if (kstrtoint(value, base, &_res))
	1156	+ ret = -EINVAL;
	1157	+ kfree(value);
	1158	+ *res = _res << shift_left_factor;
	1159	+ return ret;
	1160	+}
	1161	+
	1162	+static inline void
	1163	+xfs_fs_warn_deprecated(
	1164	+ struct fs_context *fc,
	1165	+ struct fs_parameter *param,
	1166	+ uint64_t flag,
	1167	+ bool value)
	1168	+{
	1169	+ /* Don't print the warning if reconfiguring and current mount point
	1170	+ * already had the flag set
	1171	+ */
	1172	+ if ((fc->purpose & FS_CONTEXT_FOR_RECONFIGURE) &&
	1173	+ !!(XFS_M(fc->root->d_sb)->m_flags & flag) == value)
	1174	+ return;
	1175	+ xfs_warn(fc->s_fs_info, "%s mount option is deprecated.", param->key);
	1176	+}
	1177	+
	1178	+/*
	1179	+ * Set mount state from a mount option.
	1180	+ *
	1181	+ * NOTE: mp->m_super is NULL here!
	1182	+ */
	1183	+static int
	1184	+xfs_fc_parse_param(
	1185	+ struct fs_context *fc,
	1186	+ struct fs_parameter *param)
	1187	+{
	1188	+ struct xfs_mount *parsing_mp = fc->s_fs_info;
	1189	+ struct fs_parse_result result;
	1190	+ int size = 0;
	1191	+ int opt;
	1192	+
	1193	+ opt = fs_parse(fc, xfs_fs_parameters, param, &result);
	1194	+ if (opt < 0)
	1195	+ return opt;
	1196	+
	1197	+ switch (opt) {
	1198	+ case Opt_logbufs:
	1199	+ parsing_mp->m_logbufs = result.uint_32;
	1200	+ return 0;
	1201	+ case Opt_logbsize:
	1202	+ if (suffix_kstrtoint(param->string, 10, &parsing_mp->m_logbsize))
	1203	+ return -EINVAL;
	1204	+ return 0;
	1205	+ case Opt_logdev:
	1206	+ kfree(parsing_mp->m_logname);
	1207	+ parsing_mp->m_logname = kstrdup(param->string, GFP_KERNEL);
	1208	+ if (!parsing_mp->m_logname)
	1209	+ return -ENOMEM;
	1210	+ return 0;
	1211	+ case Opt_rtdev:
	1212	+ kfree(parsing_mp->m_rtname);
	1213	+ parsing_mp->m_rtname = kstrdup(param->string, GFP_KERNEL);
	1214	+ if (!parsing_mp->m_rtname)
	1215	+ return -ENOMEM;
	1216	+ return 0;
	1217	+ case Opt_allocsize:
	1218	+ if (suffix_kstrtoint(param->string, 10, &size))
	1219	+ return -EINVAL;
	1220	+ parsing_mp->m_allocsize_log = ffs(size) - 1;
	1221	+ parsing_mp->m_flags \|= XFS_MOUNT_ALLOCSIZE;
	1222	+ return 0;
	1223	+ case Opt_grpid:
	1224	+ case Opt_bsdgroups:
	1225	+ parsing_mp->m_flags \|= XFS_MOUNT_GRPID;
	1226	+ return 0;
	1227	+ case Opt_nogrpid:
	1228	+ case Opt_sysvgroups:
	1229	+ parsing_mp->m_flags &= ~XFS_MOUNT_GRPID;
	1230	+ return 0;
	1231	+ case Opt_wsync:
	1232	+ parsing_mp->m_flags \|= XFS_MOUNT_WSYNC;
	1233	+ return 0;
	1234	+ case Opt_norecovery:
	1235	+ parsing_mp->m_flags \|= XFS_MOUNT_NORECOVERY;
	1236	+ return 0;
	1237	+ case Opt_noalign:
	1238	+ parsing_mp->m_flags \|= XFS_MOUNT_NOALIGN;
	1239	+ return 0;
	1240	+ case Opt_swalloc:
	1241	+ parsing_mp->m_flags \|= XFS_MOUNT_SWALLOC;
	1242	+ return 0;
	1243	+ case Opt_sunit:
	1244	+ parsing_mp->m_dalign = result.uint_32;
	1245	+ return 0;
	1246	+ case Opt_swidth:
	1247	+ parsing_mp->m_swidth = result.uint_32;
	1248	+ return 0;
	1249	+ case Opt_inode32:
	1250	+ parsing_mp->m_flags \|= XFS_MOUNT_SMALL_INUMS;
	1251	+ return 0;
	1252	+ case Opt_inode64:
	1253	+ parsing_mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
	1254	+ return 0;
	1255	+ case Opt_nouuid:
	1256	+ parsing_mp->m_flags \|= XFS_MOUNT_NOUUID;
	1257	+ return 0;
	1258	+ case Opt_largeio:
	1259	+ parsing_mp->m_flags \|= XFS_MOUNT_LARGEIO;
	1260	+ return 0;
	1261	+ case Opt_nolargeio:
	1262	+ parsing_mp->m_flags &= ~XFS_MOUNT_LARGEIO;
	1263	+ return 0;
	1264	+ case Opt_filestreams:
	1265	+ parsing_mp->m_flags \|= XFS_MOUNT_FILESTREAMS;
	1266	+ return 0;
	1267	+ case Opt_noquota:
	1268	+ parsing_mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
	1269	+ parsing_mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
	1270	+ parsing_mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
	1271	+ return 0;
	1272	+ case Opt_quota:
	1273	+ case Opt_uquota:
	1274	+ case Opt_usrquota:
	1275	+ parsing_mp->m_qflags \|= (XFS_UQUOTA_ACCT \| XFS_UQUOTA_ACTIVE \|
	1276	+ XFS_UQUOTA_ENFD);
	1277	+ return 0;
	1278	+ case Opt_qnoenforce:
	1279	+ case Opt_uqnoenforce:
	1280	+ parsing_mp->m_qflags \|= (XFS_UQUOTA_ACCT \| XFS_UQUOTA_ACTIVE);
	1281	+ parsing_mp->m_qflags &= ~XFS_UQUOTA_ENFD;
	1282	+ return 0;
	1283	+ case Opt_pquota:
	1284	+ case Opt_prjquota:
	1285	+ parsing_mp->m_qflags \|= (XFS_PQUOTA_ACCT \| XFS_PQUOTA_ACTIVE \|
	1286	+ XFS_PQUOTA_ENFD);
	1287	+ return 0;
	1288	+ case Opt_pqnoenforce:
	1289	+ parsing_mp->m_qflags \|= (XFS_PQUOTA_ACCT \| XFS_PQUOTA_ACTIVE);
	1290	+ parsing_mp->m_qflags &= ~XFS_PQUOTA_ENFD;
	1291	+ return 0;
	1292	+ case Opt_gquota:
	1293	+ case Opt_grpquota:
	1294	+ parsing_mp->m_qflags \|= (XFS_GQUOTA_ACCT \| XFS_GQUOTA_ACTIVE \|
	1295	+ XFS_GQUOTA_ENFD);
	1296	+ return 0;
	1297	+ case Opt_gqnoenforce:
	1298	+ parsing_mp->m_qflags \|= (XFS_GQUOTA_ACCT \| XFS_GQUOTA_ACTIVE);
	1299	+ parsing_mp->m_qflags &= ~XFS_GQUOTA_ENFD;
	1300	+ return 0;
	1301	+ case Opt_discard:
	1302	+ parsing_mp->m_flags \|= XFS_MOUNT_DISCARD;
	1303	+ return 0;
	1304	+ case Opt_nodiscard:
	1305	+ parsing_mp->m_flags &= ~XFS_MOUNT_DISCARD;
	1306	+ return 0;
	1307	+#ifdef CONFIG_FS_DAX
	1308	+ case Opt_dax:
	1309	+ xfs_mount_set_dax_mode(parsing_mp, XFS_DAX_ALWAYS);
	1310	+ return 0;
	1311	+ case Opt_dax_enum:
	1312	+ xfs_mount_set_dax_mode(parsing_mp, result.uint_32);
	1313	+ return 0;
	1314	+#endif
	1315	+ /* Following mount options will be removed in September 2025 */
	1316	+ case Opt_ikeep:
	1317	+ xfs_fs_warn_deprecated(fc, param, XFS_MOUNT_IKEEP, true);
	1318	+ parsing_mp->m_flags \|= XFS_MOUNT_IKEEP;
	1319	+ return 0;
	1320	+ case Opt_noikeep:
	1321	+ xfs_fs_warn_deprecated(fc, param, XFS_MOUNT_IKEEP, false);
	1322	+ parsing_mp->m_flags &= ~XFS_MOUNT_IKEEP;
	1323	+ return 0;
	1324	+ case Opt_attr2:
	1325	+ xfs_fs_warn_deprecated(fc, param, XFS_MOUNT_ATTR2, true);
	1326	+ parsing_mp->m_flags \|= XFS_MOUNT_ATTR2;
	1327	+ return 0;
	1328	+ case Opt_noattr2:
	1329	+ xfs_fs_warn_deprecated(fc, param, XFS_MOUNT_NOATTR2, true);
	1330	+ parsing_mp->m_flags &= ~XFS_MOUNT_ATTR2;
	1331	+ parsing_mp->m_flags \|= XFS_MOUNT_NOATTR2;
	1332	+ return 0;
	1333	+ default:
	1334	+ xfs_warn(parsing_mp, "unknown mount option [%s].", param->key);
	1335	+ return -EINVAL;
	1336	+ }
	1337	+
	1338	+ return 0;
	1339	+}
	1340	+
	1341	+static int
	1342	+xfs_fc_validate_params(
	1343	+ struct xfs_mount *mp)
	1344	+{
	1345	+ /*
	1346	+ * no recovery flag requires a read-only mount
	1347	+ */
	1348	+ if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
	1349	+ !(mp->m_flags & XFS_MOUNT_RDONLY)) {
	1350	+ xfs_warn(mp, "no-recovery mounts must be read-only.");
	1351	+ return -EINVAL;
	1352	+ }
	1353	+
	1354	+ if ((mp->m_flags & XFS_MOUNT_NOALIGN) &&
	1355	+ (mp->m_dalign \|\| mp->m_swidth)) {
	1356	+ xfs_warn(mp,
	1357	+ "sunit and swidth options incompatible with the noalign option");
	1358	+ return -EINVAL;
	1359	+ }
	1360	+
	1361	+ if (!IS_ENABLED(CONFIG_XFS_QUOTA) && mp->m_qflags != 0) {
	1362	+ xfs_warn(mp, "quota support not available in this kernel.");
	1363	+ return -EINVAL;
	1364	+ }
	1365	+
	1366	+ if ((mp->m_dalign && !mp->m_swidth) \|\|
	1367	+ (!mp->m_dalign && mp->m_swidth)) {
	1368	+ xfs_warn(mp, "sunit and swidth must be specified together");
	1369	+ return -EINVAL;
	1370	+ }
	1371	+
	1372	+ if (mp->m_dalign && (mp->m_swidth % mp->m_dalign != 0)) {
	1373	+ xfs_warn(mp,
	1374	+ "stripe width (%d) must be a multiple of the stripe unit (%d)",
	1375	+ mp->m_swidth, mp->m_dalign);
	1376	+ return -EINVAL;
	1377	+ }
	1378	+
	1379	+ if (mp->m_logbufs != -1 &&
	1380	+ mp->m_logbufs != 0 &&
	1381	+ (mp->m_logbufs < XLOG_MIN_ICLOGS \|\|
	1382	+ mp->m_logbufs > XLOG_MAX_ICLOGS)) {
	1383	+ xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
	1384	+ mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
	1385	+ return -EINVAL;
	1386	+ }
	1387	+
	1388	+ if (mp->m_logbsize != -1 &&
	1389	+ mp->m_logbsize != 0 &&
	1390	+ (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE \|\|
	1391	+ mp->m_logbsize > XLOG_MAX_RECORD_BSIZE \|\|
	1392	+ !is_power_of_2(mp->m_logbsize))) {
	1393	+ xfs_warn(mp,
	1394	+ "invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
	1395	+ mp->m_logbsize);
	1396	+ return -EINVAL;
	1397	+ }
	1398	+
	1399	+ if ((mp->m_flags & XFS_MOUNT_ALLOCSIZE) &&
	1400	+ (mp->m_allocsize_log > XFS_MAX_IO_LOG \|\|
	1401	+ mp->m_allocsize_log < XFS_MIN_IO_LOG)) {
	1402	+ xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
	1403	+ mp->m_allocsize_log, XFS_MIN_IO_LOG, XFS_MAX_IO_LOG);
	1404	+ return -EINVAL;
	1405	+ }
	1406	+
	1407	+ return 0;
	1408	+}
	1409	+
	1410	+static int
	1411	+xfs_fc_fill_super(
	1412	+ struct super_block *sb,
	1413	+ struct fs_context *fc)
	1414	+{
	1415	+ struct xfs_mount *mp = sb->s_fs_info;
	1416	+ struct inode *root;
	1417	+ int flags = 0, error;
	1418	+
	1419	+ mp->m_super = sb;
	1420	+
	1421	+ error = xfs_fc_validate_params(mp);
	1422	+ if (error)
	1423	+ goto out_free_names;
	1424	+
	1425	+ sb_min_blocksize(sb, BBSIZE);
	1426	+ sb->s_xattr = xfs_xattr_handlers;
	1427	+ sb->s_export_op = &xfs_export_operations;
	1428	+#ifdef CONFIG_XFS_QUOTA
	1429	+ sb->s_qcop = &xfs_quotactl_operations;
	1430	+ sb->s_quota_types = QTYPE_MASK_USR \| QTYPE_MASK_GRP \| QTYPE_MASK_PRJ;
	1431	+#endif
	1432	+ sb->s_op = &xfs_super_operations;
	1433	+
	1434	+ /*
	1435	+ * Delay mount work if the debug hook is set. This is debug
	1436	+ * instrumention to coordinate simulation of xfs mount failures with
	1437	+ * VFS superblock operations
	1438	+ */
	1439	+ if (xfs_globals.mount_delay) {
	1440	+ xfs_notice(mp, "Delaying mount for %d seconds.",
	1441	+ xfs_globals.mount_delay);
	1442	+ msleep(xfs_globals.mount_delay * 1000);
	1443	+ }
	1444	+
	1445	+ if (fc->sb_flags & SB_SILENT)
	1446	+ flags \|= XFS_MFSI_QUIET;
	1447	+
	1448	+ error = xfs_open_devices(mp);
	1449	+ if (error)
	1450	+ goto out_free_names;
	1451	+
	1452	+ error = xfs_init_mount_workqueues(mp);
	1453	+ if (error)
	1454	+ goto out_close_devices;
	1455	+
	1456	+ error = xfs_init_percpu_counters(mp);
	1457	+ if (error)
	1458	+ goto out_destroy_workqueues;
	1459	+
	1460	+ /* Allocate stats memory before we do operations that might use it */
	1461	+ mp->m_stats.xs_stats = alloc_percpu(struct xfsstats);
	1462	+ if (!mp->m_stats.xs_stats) {
	1463	+ error = -ENOMEM;
	1464	+ goto out_destroy_counters;
	1465	+ }
	1466	+
	1467	+ error = xfs_readsb(mp, flags);
	1468	+ if (error)
	1469	+ goto out_free_stats;
	1470	+
	1471	+ error = xfs_finish_flags(mp);
	1472	+ if (error)
	1473	+ goto out_free_sb;
	1474	+
	1475	+ error = xfs_setup_devices(mp);
	1476	+ if (error)
	1477	+ goto out_free_sb;
	1478	+
	1479	+ /* V4 support is undergoing deprecation. */
	1480	+ if (!xfs_sb_version_hascrc(&mp->m_sb)) {
	1481	+#ifdef CONFIG_XFS_SUPPORT_V4
	1482	+ xfs_warn_once(mp,
	1483	+ "Deprecated V4 format (crc=0) will not be supported after September 2030.");
	1484	+#else
	1485	+ xfs_warn(mp,
	1486	+ "Deprecated V4 format (crc=0) not supported by kernel.");
	1487	+ error = -EINVAL;
	1488	+ goto out_free_sb;
	1489	+#endif
	1490	+ }
	1491	+
	1492	+ /*
	1493	+ * XFS block mappings use 54 bits to store the logical block offset.
	1494	+ * This should suffice to handle the maximum file size that the VFS
	1495	+ * supports (currently 2^63 bytes on 64-bit and ULONG_MAX << PAGE_SHIFT
	1496	+ * bytes on 32-bit), but as XFS and VFS have gotten the s_maxbytes
	1497	+ * calculation wrong on 32-bit kernels in the past, we'll add a WARN_ON
	1498	+ * to check this assertion.
	1499	+ *
	1500	+ * Avoid integer overflow by comparing the maximum bmbt offset to the
	1501	+ * maximum pagecache offset in units of fs blocks.
	1502	+ */
	1503	+ if (XFS_B_TO_FSBT(mp, MAX_LFS_FILESIZE) > XFS_MAX_FILEOFF) {
	1504	+ xfs_warn(mp,
	1505	+"MAX_LFS_FILESIZE block offset (%llu) exceeds extent map maximum (%llu)!",
	1506	+ XFS_B_TO_FSBT(mp, MAX_LFS_FILESIZE),
	1507	+ XFS_MAX_FILEOFF);
	1508	+ error = -EINVAL;
	1509	+ goto out_free_sb;
	1510	+ }
	1511	+
	1512	+ error = xfs_filestream_mount(mp);
	1513	+ if (error)
	1514	+ goto out_free_sb;
	1515	+
	1516	+ /*
	1517	+ * we must configure the block size in the superblock before we run the
	1518	+ * full mount process as the mount process can lookup and cache inodes.
	1519	+ */
	1520	+ sb->s_magic = XFS_SUPER_MAGIC;
	1521	+ sb->s_blocksize = mp->m_sb.sb_blocksize;
	1522	+ sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1;
	1523	+ sb->s_maxbytes = MAX_LFS_FILESIZE;
	1524	+ sb->s_max_links = XFS_MAXLINK;
	1525	+ sb->s_time_gran = 1;
	1526	+ if (xfs_sb_version_hasbigtime(&mp->m_sb)) {
	1527	+ sb->s_time_min = xfs_bigtime_to_unix(XFS_BIGTIME_TIME_MIN);
	1528	+ sb->s_time_max = xfs_bigtime_to_unix(XFS_BIGTIME_TIME_MAX);
	1529	+ } else {
	1530	+ sb->s_time_min = XFS_LEGACY_TIME_MIN;
	1531	+ sb->s_time_max = XFS_LEGACY_TIME_MAX;
	1532	+ }
	1533	+ trace_xfs_inode_timestamp_range(mp, sb->s_time_min, sb->s_time_max);
	1534	+ sb->s_iflags \|= SB_I_CGROUPWB;
	1535	+
	1536	+ set_posix_acl_flag(sb);
	1537	+
	1538	+ /* version 5 superblocks support inode version counters. */
	1539	+ if (XFS_SB_VERSION_NUM(&mp->m_sb) == XFS_SB_VERSION_5)
	1540	+ sb->s_flags \|= SB_I_VERSION;
	1541	+
	1542	+ if (xfs_sb_version_hasbigtime(&mp->m_sb))
	1543	+ xfs_warn(mp,
	1544	+ "EXPERIMENTAL big timestamp feature in use. Use at your own risk!");
	1545	+
	1546	+ if (mp->m_flags & XFS_MOUNT_DAX_ALWAYS) {
	1547	+ bool rtdev_is_dax = false, datadev_is_dax;
	1548	+
	1549	+ xfs_warn(mp,
	1550	+ "DAX enabled. Warning: EXPERIMENTAL, use at your own risk");
	1551	+
	1552	+ datadev_is_dax = bdev_dax_supported(mp->m_ddev_targp->bt_bdev,
	1553	+ sb->s_blocksize);
	1554	+ if (mp->m_rtdev_targp)
	1555	+ rtdev_is_dax = bdev_dax_supported(
	1556	+ mp->m_rtdev_targp->bt_bdev, sb->s_blocksize);
	1557	+ if (!rtdev_is_dax && !datadev_is_dax) {
	1558	+ xfs_alert(mp,
	1559	+ "DAX unsupported by block device. Turning off DAX.");
	1560	+ xfs_mount_set_dax_mode(mp, XFS_DAX_NEVER);
	1561	+ }
	1562	+ if (xfs_sb_version_hasreflink(&mp->m_sb)) {
	1563	+ xfs_alert(mp,
	1564	+ "DAX and reflink cannot be used together!");
	1565	+ error = -EINVAL;
	1566	+ goto out_filestream_unmount;
	1567	+ }
	1568	+ }
	1569	+
	1570	+ if (mp->m_flags & XFS_MOUNT_DISCARD) {
	1571	+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
	1572	+
	1573	+ if (!blk_queue_discard(q)) {
	1574	+ xfs_warn(mp, "mounting with \"discard\" option, but "
	1575	+ "the device does not support discard");
	1576	+ mp->m_flags &= ~XFS_MOUNT_DISCARD;
	1577	+ }
	1578	+ }
	1579	+
	1580	+ if (xfs_sb_version_hasreflink(&mp->m_sb)) {
	1581	+ if (mp->m_sb.sb_rblocks) {
	1582	+ xfs_alert(mp,
	1583	+ "reflink not compatible with realtime device!");
	1584	+ error = -EINVAL;
	1585	+ goto out_filestream_unmount;
	1586	+ }
	1587	+
	1588	+ if (xfs_globals.always_cow) {
	1589	+ xfs_info(mp, "using DEBUG-only always_cow mode.");
	1590	+ mp->m_always_cow = true;
	1591	+ }
	1592	+ }
	1593	+
	1594	+ if (xfs_sb_version_hasrmapbt(&mp->m_sb) && mp->m_sb.sb_rblocks) {
	1595	+ xfs_alert(mp,
	1596	+ "reverse mapping btree not compatible with realtime device!");
	1597	+ error = -EINVAL;
	1598	+ goto out_filestream_unmount;
	1599	+ }
	1600	+
	1601	+ if (xfs_sb_version_hasinobtcounts(&mp->m_sb))
	1602	+ xfs_warn(mp,
	1603	+ "EXPERIMENTAL inode btree counters feature in use. Use at your own risk!");
	1604	+
	1605	+ error = xfs_mountfs(mp);
	1606	+ if (error)
	1607	+ goto out_filestream_unmount;
	1608	+
	1609	+ root = igrab(VFS_I(mp->m_rootip));
	1610	+ if (!root) {
	1611	+ error = -ENOENT;
	1612	+ goto out_unmount;
	1613	+ }
	1614	+ sb->s_root = d_make_root(root);
	1615	+ if (!sb->s_root) {
	1616	+ error = -ENOMEM;
	1617	+ goto out_unmount;
	1618	+ }
	1619	+
	1620	+ return 0;
	1621	+
	1622	+ out_filestream_unmount:
	1623	+ xfs_filestream_unmount(mp);
	1624	+ out_free_sb:
	1625	+ xfs_freesb(mp);
	1626	+ out_free_stats:
	1627	+ free_percpu(mp->m_stats.xs_stats);
	1628	+ out_destroy_counters:
	1629	+ xfs_destroy_percpu_counters(mp);
	1630	+ out_destroy_workqueues:
	1631	+ xfs_destroy_mount_workqueues(mp);
	1632	+ out_close_devices:
	1633	+ xfs_close_devices(mp);
	1634	+ out_free_names:
	1635	+ sb->s_fs_info = NULL;
	1636	+ xfs_mount_free(mp);
	1637	+ return error;
	1638	+
	1639	+ out_unmount:
	1640	+ xfs_filestream_unmount(mp);
	1641	+ xfs_unmountfs(mp);
	1642	+ goto out_free_sb;
	1643	+}
	1644	+
	1645	+static int
	1646	+xfs_fc_get_tree(
	1647	+ struct fs_context *fc)
	1648	+{
	1649	+ return get_tree_bdev(fc, xfs_fc_fill_super);
	1650	+}
	1651	+
	1652	+static int
	1653	+xfs_remount_rw(
	1654	+ struct xfs_mount *mp)
	1655	+{
	1656	+ struct xfs_sb *sbp = &mp->m_sb;
	1657	+ int error;
	1658	+
	1659	+ if (mp->m_flags & XFS_MOUNT_NORECOVERY) {
	1660	+ xfs_warn(mp,
	1661	+ "ro->rw transition prohibited on norecovery mount");
	1662	+ return -EINVAL;
	1663	+ }
	1664	+
	1665	+ if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
	1666	+ xfs_sb_has_ro_compat_feature(sbp, XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
	1667	+ xfs_warn(mp,
	1668	+ "ro->rw transition prohibited on unknown (0x%x) ro-compat filesystem",
	1669	+ (sbp->sb_features_ro_compat &
	1670	+ XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
	1671	+ return -EINVAL;
	1672	+ }
	1673	+
	1674	+ mp->m_flags &= ~XFS_MOUNT_RDONLY;
	1675	+
	1676	+ /*
	1677	+ * If this is the first remount to writeable state we might have some
	1678	+ * superblock changes to update.
	1679	+ */
	1680	+ if (mp->m_update_sb) {
	1681	+ error = xfs_sync_sb(mp, false);
	1682	+ if (error) {
	1683	+ xfs_warn(mp, "failed to write sb changes");
	1684	+ return error;
	1685	+ }
	1686	+ mp->m_update_sb = false;
	1687	+ }
	1688	+
	1689	+ /*
	1690	+ * Fill out the reserve pool if it is empty. Use the stashed value if
	1691	+ * it is non-zero, otherwise go with the default.
	1692	+ */
	1693	+ xfs_restore_resvblks(mp);
	1694	+ xfs_log_work_queue(mp);
	1695	+
	1696	+ /* Recover any CoW blocks that never got remapped. */
	1697	+ error = xfs_reflink_recover_cow(mp);
	1698	+ if (error) {
	1699	+ xfs_err(mp,
	1700	+ "Error %d recovering leftover CoW allocations.", error);
	1701	+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
	1702	+ return error;
	1703	+ }
	1704	+ xfs_start_block_reaping(mp);
	1705	+
	1706	+ /* Create the per-AG metadata reservation pool .*/
	1707	+ error = xfs_fs_reserve_ag_blocks(mp);
	1708	+ if (error && error != -ENOSPC)
	1709	+ return error;
	1710	+
	1711	+ return 0;
	1712	+}
	1713	+
	1714	+static int
	1715	+xfs_remount_ro(
	1716	+ struct xfs_mount *mp)
	1717	+{
	1718	+ struct xfs_eofblocks eofb = {
	1719	+ .eof_flags = XFS_EOF_FLAGS_SYNC,
	1720	+ };
	1721	+ int error;
	1722	+
	1723	+ /* Flush all the dirty data to disk. */
	1724	+ error = sync_filesystem(mp->m_super);
	1725	+ if (error)
	1726	+ return error;
	1727	+
	1728	+ /*
	1729	+ * Cancel background eofb scanning so it cannot race with the final
	1730	+ * log force+buftarg wait and deadlock the remount.
	1731	+ */
	1732	+ xfs_stop_block_reaping(mp);
	1733	+
	1734	+ /*
	1735	+ * Clear out all remaining COW staging extents and speculative post-EOF
	1736	+ * preallocations so that we don't leave inodes requiring inactivation
	1737	+ * cleanups during reclaim on a read-only mount. We must process every
	1738	+ * cached inode, so this requires a synchronous cache scan.
	1739	+ */
	1740	+ error = xfs_icache_free_cowblocks(mp, &eofb);
	1741	+ if (error) {
	1742	+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
	1743	+ return error;
	1744	+ }
	1745	+
	1746	+ /* Free the per-AG metadata reservation pool. */
	1747	+ error = xfs_fs_unreserve_ag_blocks(mp);
	1748	+ if (error) {
	1749	+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
	1750	+ return error;
	1751	+ }
	1752	+
	1753	+ /*
	1754	+ * Before we sync the metadata, we need to free up the reserve block
	1755	+ * pool so that the used block count in the superblock on disk is
	1756	+ * correct at the end of the remount. Stash the current* reserve pool
	1757	+ * size so that if we get remounted rw, we can return it to the same
	1758	+ * size.
	1759	+ */
	1760	+ xfs_save_resvblks(mp);
	1761	+
	1762	+ xfs_quiesce_attr(mp);
	1763	+ mp->m_flags \|= XFS_MOUNT_RDONLY;
	1764	+
	1765	+ return 0;
	1766	+}
	1767	+
	1768	+/*
	1769	+ * Logically we would return an error here to prevent users from believing
	1770	+ * they might have changed mount options using remount which can't be changed.
	1771	+ *
	1772	+ * But unfortunately mount(8) adds all options from mtab and fstab to the mount
	1773	+ * arguments in some cases so we can't blindly reject options, but have to
	1774	+ * check for each specified option if it actually differs from the currently
	1775	+ * set option and only reject it if that's the case.
	1776	+ *
	1777	+ * Until that is implemented we return success for every remount request, and
	1778	+ * silently ignore all options that we can't actually change.
	1779	+ */
	1780	+static int
	1781	+xfs_fc_reconfigure(
	1782	+ struct fs_context *fc)
	1783	+{
	1784	+ struct xfs_mount *mp = XFS_M(fc->root->d_sb);
	1785	+ struct xfs_mount *new_mp = fc->s_fs_info;
	1786	+ xfs_sb_t *sbp = &mp->m_sb;
	1787	+ int flags = fc->sb_flags;
	1788	+ int error;
	1789	+
	1790	+ /* version 5 superblocks always support version counters. */
	1791	+ if (XFS_SB_VERSION_NUM(&mp->m_sb) == XFS_SB_VERSION_5)
	1792	+ fc->sb_flags \|= SB_I_VERSION;
	1793	+
	1794	+ error = xfs_fc_validate_params(new_mp);
	1795	+ if (error)
	1796	+ return error;
	1797	+
	1798	+ /* inode32 -> inode64 */
	1799	+ if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
	1800	+ !(new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
	1801	+ mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
	1802	+ mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
	1803	+ }
	1804	+
	1805	+ /* inode64 -> inode32 */
	1806	+ if (!(mp->m_flags & XFS_MOUNT_SMALL_INUMS) &&
	1807	+ (new_mp->m_flags & XFS_MOUNT_SMALL_INUMS)) {
	1808	+ mp->m_flags \|= XFS_MOUNT_SMALL_INUMS;
	1809	+ mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
	1810	+ }
	1811	+
	1812	+ /* ro -> rw */
	1813	+ if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(flags & SB_RDONLY)) {
	1814	+ error = xfs_remount_rw(mp);
	1815	+ if (error)
	1816	+ return error;
	1817	+ }
	1818	+
	1819	+ /* rw -> ro */
	1820	+ if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (flags & SB_RDONLY)) {
	1821	+ error = xfs_remount_ro(mp);
	1822	+ if (error)
	1823	+ return error;
	1824	+ }
	1825	+
	1826	+ return 0;
	1827	+}
	1828	+
	1829	+static void xfs_fc_free(
	1830	+ struct fs_context *fc)
	1831	+{
	1832	+ struct xfs_mount *mp = fc->s_fs_info;
	1833	+
	1834	+ /*
	1835	+ * mp is stored in the fs_context when it is initialized.
	1836	+ * mp is transferred to the superblock on a successful mount,
	1837	+ * but if an error occurs before the transfer we have to free
	1838	+ * it here.
	1839	+ */
	1840	+ if (mp)
	1841	+ xfs_mount_free(mp);
	1842	+}
	1843	+
	1844	+static const struct fs_context_operations xfs_context_ops = {
	1845	+ .parse_param = xfs_fc_parse_param,
	1846	+ .get_tree = xfs_fc_get_tree,
	1847	+ .reconfigure = xfs_fc_reconfigure,
	1848	+ .free = xfs_fc_free,
	1849	+};
	1850	+
	1851	+static int xfs_init_fs_context(
	1852	+ struct fs_context *fc)
	1853	+{
	1854	+ struct xfs_mount *mp;
	1855	+
	1856	+ mp = kmem_alloc(sizeof(struct xfs_mount), KM_ZERO);
	1857	+ if (!mp)
	1858	+ return -ENOMEM;
	1859	+
	1860	+ spin_lock_init(&mp->m_sb_lock);
	1861	+ spin_lock_init(&mp->m_agirotor_lock);
	1862	+ INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
	1863	+ spin_lock_init(&mp->m_perag_lock);
	1864	+ mutex_init(&mp->m_growlock);
	1865	+ INIT_WORK(&mp->m_flush_inodes_work, xfs_flush_inodes_worker);
	1866	+ INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
	1867	+ INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
	1868	+ INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker);
	1869	+ mp->m_kobj.kobject.kset = xfs_kset;
	1870	+ /*
	1871	+ * We don't create the finobt per-ag space reservation until after log
	1872	+ * recovery, so we must set this to true so that an ifree transaction
	1873	+ * started during log recovery will not depend on space reservations
	1874	+ * for finobt expansion.
	1875	+ */
	1876	+ mp->m_finobt_nores = true;
	1877	+
	1878	+ /*
	1879	+ * These can be overridden by the mount option parsing.
	1880	+ */
	1881	+ mp->m_logbufs = -1;
	1882	+ mp->m_logbsize = -1;
	1883	+ mp->m_allocsize_log = 16; /* 64k */
	1884	+
	1885	+ /*
	1886	+ * Copy binary VFS mount flags we are interested in.
	1887	+ */
	1888	+ if (fc->sb_flags & SB_RDONLY)
	1889	+ mp->m_flags \|= XFS_MOUNT_RDONLY;
	1890	+ if (fc->sb_flags & SB_DIRSYNC)
	1891	+ mp->m_flags \|= XFS_MOUNT_DIRSYNC;
	1892	+ if (fc->sb_flags & SB_SYNCHRONOUS)
	1893	+ mp->m_flags \|= XFS_MOUNT_WSYNC;
	1894	+
	1895	+ fc->s_fs_info = mp;
	1896	+ fc->ops = &xfs_context_ops;
	1897	+
	1898	+ return 0;
	1899	+}
	1900	+
1832	1901	static struct file_system_type xfs_fs_type = {
1833	1902	.owner = THIS_MODULE,
1834	1903	.name = "xfs",
1835		- .mount = xfs_fs_mount,
	1904	+ .init_fs_context = xfs_init_fs_context,
	1905	+ .parameters = xfs_fs_parameters,
1836	1906	.kill_sb = kill_block_super,
1837	1907	.fs_flags = FS_REQUIRES_DEV,
1838	1908	};
..	..	@@ -1841,37 +1911,39 @@
1841	1911	STATIC int __init
1842	1912	xfs_init_zones(void)
1843	1913	{
1844		- if (bioset_init(&xfs_ioend_bioset, 4 * (PAGE_SIZE / SECTOR_SIZE),
1845		- offsetof(struct xfs_ioend, io_inline_bio),
1846		- BIOSET_NEED_BVECS))
	1914	+ xfs_log_ticket_zone = kmem_cache_create("xfs_log_ticket",
	1915	+ sizeof(struct xlog_ticket),
	1916	+ 0, 0, NULL);
	1917	+ if (!xfs_log_ticket_zone)
1847	1918	goto out;
1848	1919
1849		- xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t),
1850		- "xfs_log_ticket");
1851		- if (!xfs_log_ticket_zone)
1852		- goto out_free_ioend_bioset;
1853		-
1854		- xfs_bmap_free_item_zone = kmem_zone_init(
1855		- sizeof(struct xfs_extent_free_item),
1856		- "xfs_bmap_free_item");
	1920	+ xfs_bmap_free_item_zone = kmem_cache_create("xfs_bmap_free_item",
	1921	+ sizeof(struct xfs_extent_free_item),
	1922	+ 0, 0, NULL);
1857	1923	if (!xfs_bmap_free_item_zone)
1858	1924	goto out_destroy_log_ticket_zone;
1859	1925
1860		- xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
1861		- "xfs_btree_cur");
	1926	+ xfs_btree_cur_zone = kmem_cache_create("xfs_btree_cur",
	1927	+ sizeof(struct xfs_btree_cur),
	1928	+ 0, 0, NULL);
1862	1929	if (!xfs_btree_cur_zone)
1863	1930	goto out_destroy_bmap_free_item_zone;
1864	1931
1865		- xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t),
1866		- "xfs_da_state");
	1932	+ xfs_da_state_zone = kmem_cache_create("xfs_da_state",
	1933	+ sizeof(struct xfs_da_state),
	1934	+ 0, 0, NULL);
1867	1935	if (!xfs_da_state_zone)
1868	1936	goto out_destroy_btree_cur_zone;
1869	1937
1870		- xfs_ifork_zone = kmem_zone_init(sizeof(struct xfs_ifork), "xfs_ifork");
	1938	+ xfs_ifork_zone = kmem_cache_create("xfs_ifork",
	1939	+ sizeof(struct xfs_ifork),
	1940	+ 0, 0, NULL);
1871	1941	if (!xfs_ifork_zone)
1872	1942	goto out_destroy_da_state_zone;
1873	1943
1874		- xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
	1944	+ xfs_trans_zone = kmem_cache_create("xfs_trans",
	1945	+ sizeof(struct xfs_trans),
	1946	+ 0, 0, NULL);
1875	1947	if (!xfs_trans_zone)
1876	1948	goto out_destroy_ifork_zone;
1877	1949
..	..	@@ -1881,111 +1953,122 @@
1881	1953	* size possible under XFS. This wastes a little bit of memory,
1882	1954	* but it is much faster.
1883	1955	*/
1884		- xfs_buf_item_zone = kmem_zone_init(sizeof(struct xfs_buf_log_item),
1885		- "xfs_buf_item");
	1956	+ xfs_buf_item_zone = kmem_cache_create("xfs_buf_item",
	1957	+ sizeof(struct xfs_buf_log_item),
	1958	+ 0, 0, NULL);
1886	1959	if (!xfs_buf_item_zone)
1887	1960	goto out_destroy_trans_zone;
1888	1961
1889		- xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
1890		- ((XFS_EFD_MAX_FAST_EXTENTS - 1) *
1891		- sizeof(xfs_extent_t))), "xfs_efd_item");
	1962	+ xfs_efd_zone = kmem_cache_create("xfs_efd_item",
	1963	+ (sizeof(struct xfs_efd_log_item) +
	1964	+ (XFS_EFD_MAX_FAST_EXTENTS - 1) *
	1965	+ sizeof(struct xfs_extent)),
	1966	+ 0, 0, NULL);
1892	1967	if (!xfs_efd_zone)
1893	1968	goto out_destroy_buf_item_zone;
1894	1969
1895		- xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) +
1896		- ((XFS_EFI_MAX_FAST_EXTENTS - 1) *
1897		- sizeof(xfs_extent_t))), "xfs_efi_item");
	1970	+ xfs_efi_zone = kmem_cache_create("xfs_efi_item",
	1971	+ (sizeof(struct xfs_efi_log_item) +
	1972	+ (XFS_EFI_MAX_FAST_EXTENTS - 1) *
	1973	+ sizeof(struct xfs_extent)),
	1974	+ 0, 0, NULL);
1898	1975	if (!xfs_efi_zone)
1899	1976	goto out_destroy_efd_zone;
1900	1977
1901		- xfs_inode_zone =
1902		- kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
1903		- KM_ZONE_HWALIGN \| KM_ZONE_RECLAIM \| KM_ZONE_SPREAD \|
1904		- KM_ZONE_ACCOUNT, xfs_fs_inode_init_once);
	1978	+ xfs_inode_zone = kmem_cache_create("xfs_inode",
	1979	+ sizeof(struct xfs_inode), 0,
	1980	+ (SLAB_HWCACHE_ALIGN \|
	1981	+ SLAB_RECLAIM_ACCOUNT \|
	1982	+ SLAB_MEM_SPREAD \| SLAB_ACCOUNT),
	1983	+ xfs_fs_inode_init_once);
1905	1984	if (!xfs_inode_zone)
1906	1985	goto out_destroy_efi_zone;
1907	1986
1908		- xfs_ili_zone =
1909		- kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
1910		- KM_ZONE_SPREAD, NULL);
	1987	+ xfs_ili_zone = kmem_cache_create("xfs_ili",
	1988	+ sizeof(struct xfs_inode_log_item), 0,
	1989	+ SLAB_RECLAIM_ACCOUNT \| SLAB_MEM_SPREAD,
	1990	+ NULL);
1911	1991	if (!xfs_ili_zone)
1912	1992	goto out_destroy_inode_zone;
1913		- xfs_icreate_zone = kmem_zone_init(sizeof(struct xfs_icreate_item),
1914		- "xfs_icr");
	1993	+
	1994	+ xfs_icreate_zone = kmem_cache_create("xfs_icr",
	1995	+ sizeof(struct xfs_icreate_item),
	1996	+ 0, 0, NULL);
1915	1997	if (!xfs_icreate_zone)
1916	1998	goto out_destroy_ili_zone;
1917	1999
1918		- xfs_rud_zone = kmem_zone_init(sizeof(struct xfs_rud_log_item),
1919		- "xfs_rud_item");
	2000	+ xfs_rud_zone = kmem_cache_create("xfs_rud_item",
	2001	+ sizeof(struct xfs_rud_log_item),
	2002	+ 0, 0, NULL);
1920	2003	if (!xfs_rud_zone)
1921	2004	goto out_destroy_icreate_zone;
1922	2005
1923		- xfs_rui_zone = kmem_zone_init(
	2006	+ xfs_rui_zone = kmem_cache_create("xfs_rui_item",
1924	2007	xfs_rui_log_item_sizeof(XFS_RUI_MAX_FAST_EXTENTS),
1925		- "xfs_rui_item");
	2008	+ 0, 0, NULL);
1926	2009	if (!xfs_rui_zone)
1927	2010	goto out_destroy_rud_zone;
1928	2011
1929		- xfs_cud_zone = kmem_zone_init(sizeof(struct xfs_cud_log_item),
1930		- "xfs_cud_item");
	2012	+ xfs_cud_zone = kmem_cache_create("xfs_cud_item",
	2013	+ sizeof(struct xfs_cud_log_item),
	2014	+ 0, 0, NULL);
1931	2015	if (!xfs_cud_zone)
1932	2016	goto out_destroy_rui_zone;
1933	2017
1934		- xfs_cui_zone = kmem_zone_init(
	2018	+ xfs_cui_zone = kmem_cache_create("xfs_cui_item",
1935	2019	xfs_cui_log_item_sizeof(XFS_CUI_MAX_FAST_EXTENTS),
1936		- "xfs_cui_item");
	2020	+ 0, 0, NULL);
1937	2021	if (!xfs_cui_zone)
1938	2022	goto out_destroy_cud_zone;
1939	2023
1940		- xfs_bud_zone = kmem_zone_init(sizeof(struct xfs_bud_log_item),
1941		- "xfs_bud_item");
	2024	+ xfs_bud_zone = kmem_cache_create("xfs_bud_item",
	2025	+ sizeof(struct xfs_bud_log_item),
	2026	+ 0, 0, NULL);
1942	2027	if (!xfs_bud_zone)
1943	2028	goto out_destroy_cui_zone;
1944	2029
1945		- xfs_bui_zone = kmem_zone_init(
	2030	+ xfs_bui_zone = kmem_cache_create("xfs_bui_item",
1946	2031	xfs_bui_log_item_sizeof(XFS_BUI_MAX_FAST_EXTENTS),
1947		- "xfs_bui_item");
	2032	+ 0, 0, NULL);
1948	2033	if (!xfs_bui_zone)
1949	2034	goto out_destroy_bud_zone;
1950	2035
1951	2036	return 0;
1952	2037
1953	2038	out_destroy_bud_zone:
1954		- kmem_zone_destroy(xfs_bud_zone);
	2039	+ kmem_cache_destroy(xfs_bud_zone);
1955	2040	out_destroy_cui_zone:
1956		- kmem_zone_destroy(xfs_cui_zone);
	2041	+ kmem_cache_destroy(xfs_cui_zone);
1957	2042	out_destroy_cud_zone:
1958		- kmem_zone_destroy(xfs_cud_zone);
	2043	+ kmem_cache_destroy(xfs_cud_zone);
1959	2044	out_destroy_rui_zone:
1960		- kmem_zone_destroy(xfs_rui_zone);
	2045	+ kmem_cache_destroy(xfs_rui_zone);
1961	2046	out_destroy_rud_zone:
1962		- kmem_zone_destroy(xfs_rud_zone);
	2047	+ kmem_cache_destroy(xfs_rud_zone);
1963	2048	out_destroy_icreate_zone:
1964		- kmem_zone_destroy(xfs_icreate_zone);
	2049	+ kmem_cache_destroy(xfs_icreate_zone);
1965	2050	out_destroy_ili_zone:
1966		- kmem_zone_destroy(xfs_ili_zone);
	2051	+ kmem_cache_destroy(xfs_ili_zone);
1967	2052	out_destroy_inode_zone:
1968		- kmem_zone_destroy(xfs_inode_zone);
	2053	+ kmem_cache_destroy(xfs_inode_zone);
1969	2054	out_destroy_efi_zone:
1970		- kmem_zone_destroy(xfs_efi_zone);
	2055	+ kmem_cache_destroy(xfs_efi_zone);
1971	2056	out_destroy_efd_zone:
1972		- kmem_zone_destroy(xfs_efd_zone);
	2057	+ kmem_cache_destroy(xfs_efd_zone);
1973	2058	out_destroy_buf_item_zone:
1974		- kmem_zone_destroy(xfs_buf_item_zone);
	2059	+ kmem_cache_destroy(xfs_buf_item_zone);
1975	2060	out_destroy_trans_zone:
1976		- kmem_zone_destroy(xfs_trans_zone);
	2061	+ kmem_cache_destroy(xfs_trans_zone);
1977	2062	out_destroy_ifork_zone:
1978		- kmem_zone_destroy(xfs_ifork_zone);
	2063	+ kmem_cache_destroy(xfs_ifork_zone);
1979	2064	out_destroy_da_state_zone:
1980		- kmem_zone_destroy(xfs_da_state_zone);
	2065	+ kmem_cache_destroy(xfs_da_state_zone);
1981	2066	out_destroy_btree_cur_zone:
1982		- kmem_zone_destroy(xfs_btree_cur_zone);
	2067	+ kmem_cache_destroy(xfs_btree_cur_zone);
1983	2068	out_destroy_bmap_free_item_zone:
1984		- kmem_zone_destroy(xfs_bmap_free_item_zone);
	2069	+ kmem_cache_destroy(xfs_bmap_free_item_zone);
1985	2070	out_destroy_log_ticket_zone:
1986		- kmem_zone_destroy(xfs_log_ticket_zone);
1987		- out_free_ioend_bioset:
1988		- bioset_exit(&xfs_ioend_bioset);
	2071	+ kmem_cache_destroy(xfs_log_ticket_zone);
1989	2072	out:
1990	2073	return -ENOMEM;
1991	2074	}
..	..	@@ -1998,25 +2081,24 @@
1998	2081	* destroy caches.
1999	2082	*/
2000	2083	rcu_barrier();
2001		- kmem_zone_destroy(xfs_bui_zone);
2002		- kmem_zone_destroy(xfs_bud_zone);
2003		- kmem_zone_destroy(xfs_cui_zone);
2004		- kmem_zone_destroy(xfs_cud_zone);
2005		- kmem_zone_destroy(xfs_rui_zone);
2006		- kmem_zone_destroy(xfs_rud_zone);
2007		- kmem_zone_destroy(xfs_icreate_zone);
2008		- kmem_zone_destroy(xfs_ili_zone);
2009		- kmem_zone_destroy(xfs_inode_zone);
2010		- kmem_zone_destroy(xfs_efi_zone);
2011		- kmem_zone_destroy(xfs_efd_zone);
2012		- kmem_zone_destroy(xfs_buf_item_zone);
2013		- kmem_zone_destroy(xfs_trans_zone);
2014		- kmem_zone_destroy(xfs_ifork_zone);
2015		- kmem_zone_destroy(xfs_da_state_zone);
2016		- kmem_zone_destroy(xfs_btree_cur_zone);
2017		- kmem_zone_destroy(xfs_bmap_free_item_zone);
2018		- kmem_zone_destroy(xfs_log_ticket_zone);
2019		- bioset_exit(&xfs_ioend_bioset);
	2084	+ kmem_cache_destroy(xfs_bui_zone);
	2085	+ kmem_cache_destroy(xfs_bud_zone);
	2086	+ kmem_cache_destroy(xfs_cui_zone);
	2087	+ kmem_cache_destroy(xfs_cud_zone);
	2088	+ kmem_cache_destroy(xfs_rui_zone);
	2089	+ kmem_cache_destroy(xfs_rud_zone);
	2090	+ kmem_cache_destroy(xfs_icreate_zone);
	2091	+ kmem_cache_destroy(xfs_ili_zone);
	2092	+ kmem_cache_destroy(xfs_inode_zone);
	2093	+ kmem_cache_destroy(xfs_efi_zone);
	2094	+ kmem_cache_destroy(xfs_efd_zone);
	2095	+ kmem_cache_destroy(xfs_buf_item_zone);
	2096	+ kmem_cache_destroy(xfs_trans_zone);
	2097	+ kmem_cache_destroy(xfs_ifork_zone);
	2098	+ kmem_cache_destroy(xfs_da_state_zone);
	2099	+ kmem_cache_destroy(xfs_btree_cur_zone);
	2100	+ kmem_cache_destroy(xfs_bmap_free_item_zone);
	2101	+ kmem_cache_destroy(xfs_log_ticket_zone);
2020	2102	}
2021	2103
2022	2104	STATIC int __init
..	..	@@ -2059,11 +2141,6 @@
2059	2141
2060	2142	printk(KERN_INFO XFS_VERSION_STRING " with "
2061	2143	XFS_BUILD_OPTIONS " enabled\n");
2062		-
2063		- xfs_extent_free_init_defer_op();
2064		- xfs_rmap_update_init_defer_op();
2065		- xfs_refcount_update_init_defer_op();
2066		- xfs_bmap_update_init_defer_op();
2067	2144
2068	2145	xfs_dir_startup();
2069	2146
..	..	@@ -2180,3 +2257,4 @@
2180	2257	MODULE_AUTHOR("Silicon Graphics, Inc.");
2181	2258	MODULE_DESCRIPTION(XFS_VERSION_STRING " with " XFS_BUILD_OPTIONS " enabled");
2182	2259	MODULE_LICENSE("GPL");
	2260	+MODULE_IMPORT_NS(ANDROID_GKI_VFS_EXPORT_ONLY);