aboutsummaryrefslogblamecommitdiffstats
path: root/fs/nfs/nfs4proc.c
blob: 0c5a308e49638171291d22aa3630fc431ac74528 (plain) (tree)
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
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167

















































                                                                           
                    













                                                                                                             







































                                                               
                                     


















                                                                                
                                         
















































































































































                                                                                                


                                                               






































































































































































































































                                                                                                                                             













                                                                                                                    











                                                                                   
                                                                      






























































































































































































































                                                                                                                                                  
                   


                         
                            
                                                     

                                                                       

                                                                         

                                                        







































































































































































































































































































































































                                                                                                         


                                              



                         









                                                                         
                                                                  
         
 

                                                          

                                                       
                          
                                                     
                          












































































































































































































































































































                                                                                                                  

                                                                      
                                 
                 














































































































































































































































































                                                                                     



                                                                   






                                                                                   
                                                          








































































































































































































































































































































































































































                                                                                                         

























                                                                          


                        
                     































































                                                                                          
 






                                             
                       




                                                                   
                                      

                













                                                                            
                                                        
















                                                                 


                   















                                                                               



















                                                                                 

                                                          


                   




























































































































































































































































































                                                                                                          




                                                     



















                                                        
    


























































                                                                                         
                   


                                

                                                     
                         
                                          
                                           






                                                                          
 
                        

                                                          
























                                                                                                    
                                                                  




















                                                                       




















                                                                                     



                                                              





                                                                          

                                                                              

                                                                            
         
                                        
                        
                                                                               
                                           
                                 




                                                                                 










                                                                 



                                                                                 










                                                                 








                                                                                         


                                                                
























































                                                                                                         


                      
 

                                              


                                                                          









                                                                 








                                                                             





                                                     



                                                                       
                                                      



                                

                                                       

 









                                                                      








                                                             



                                                                      
                                                      





























                                                 
                                             






                     
/*
 *  fs/nfs/nfs4proc.c
 *
 *  Client-side procedure declarations for NFSv4.
 *
 *  Copyright (c) 2002 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Kendrick Smith <kmsmith@umich.edu>
 *  Andy Adamson   <andros@umich.edu>
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *
 *  1. Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *  2. Redistributions in binary form must reproduce the above copyright
 *     notice, this list of conditions and the following disclaimer in the
 *     documentation and/or other materials provided with the distribution.
 *  3. Neither the name of the University nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
 *  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 *  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 *  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 *  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 *  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
 *  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
 *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
 *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <linux/mm.h>
#include <linux/utsname.h>
#include <linux/delay.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/sunrpc/clnt.h>
#include <linux/nfs.h>
#include <linux/nfs4.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_page.h>
#include <linux/smp_lock.h>
#include <linux/namei.h>

#include "nfs4_fs.h"
#include "delegation.h"

#define NFSDBG_FACILITY		NFSDBG_PROC

#define NFS4_POLL_RETRY_MIN	(1*HZ)
#define NFS4_POLL_RETRY_MAX	(15*HZ)

static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
static int nfs4_async_handle_error(struct rpc_task *, struct nfs_server *);
static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry);
static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus);
extern struct rpc_procinfo nfs4_procedures[];

/* Prevent leaks of NFSv4 errors into userland */
int nfs4_map_errors(int err)
{
	if (err < -1000) {
		dprintk("%s could not handle NFSv4 error %d\n",
				__FUNCTION__, -err);
		return -EIO;
	}
	return err;
}

/*
 * This is our standard bitmap for GETATTR requests.
 */
const u32 nfs4_fattr_bitmap[2] = {
	FATTR4_WORD0_TYPE
	| FATTR4_WORD0_CHANGE
	| FATTR4_WORD0_SIZE
	| FATTR4_WORD0_FSID
	| FATTR4_WORD0_FILEID,
	FATTR4_WORD1_MODE
	| FATTR4_WORD1_NUMLINKS
	| FATTR4_WORD1_OWNER
	| FATTR4_WORD1_OWNER_GROUP
	| FATTR4_WORD1_RAWDEV
	| FATTR4_WORD1_SPACE_USED
	| FATTR4_WORD1_TIME_ACCESS
	| FATTR4_WORD1_TIME_METADATA
	| FATTR4_WORD1_TIME_MODIFY
};

const u32 nfs4_statfs_bitmap[2] = {
	FATTR4_WORD0_FILES_AVAIL
	| FATTR4_WORD0_FILES_FREE
	| FATTR4_WORD0_FILES_TOTAL,
	FATTR4_WORD1_SPACE_AVAIL
	| FATTR4_WORD1_SPACE_FREE
	| FATTR4_WORD1_SPACE_TOTAL
};

const u32 nfs4_pathconf_bitmap[2] = {
	FATTR4_WORD0_MAXLINK
	| FATTR4_WORD0_MAXNAME,
	0
};

const u32 nfs4_fsinfo_bitmap[2] = { FATTR4_WORD0_MAXFILESIZE
			| FATTR4_WORD0_MAXREAD
			| FATTR4_WORD0_MAXWRITE
			| FATTR4_WORD0_LEASE_TIME,
			0
};

static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry,
		struct nfs4_readdir_arg *readdir)
{
	u32 *start, *p;

	BUG_ON(readdir->count < 80);
	if (cookie > 2) {
		readdir->cookie = cookie;
		memcpy(&readdir->verifier, verifier, sizeof(readdir->verifier));
		return;
	}

	readdir->cookie = 0;
	memset(&readdir->verifier, 0, sizeof(readdir->verifier));
	if (cookie == 2)
		return;
	
	/*
	 * NFSv4 servers do not return entries for '.' and '..'
	 * Therefore, we fake these entries here.  We let '.'
	 * have cookie 0 and '..' have cookie 1.  Note that
	 * when talking to the server, we always send cookie 0
	 * instead of 1 or 2.
	 */
	start = p = (u32 *)kmap_atomic(*readdir->pages, KM_USER0);
	
	if (cookie == 0) {
		*p++ = xdr_one;                                  /* next */
		*p++ = xdr_zero;                   /* cookie, first word */
		*p++ = xdr_one;                   /* cookie, second word */
		*p++ = xdr_one;                             /* entry len */
		memcpy(p, ".\0\0\0", 4);                        /* entry */
		p++;
		*p++ = xdr_one;                         /* bitmap length */
		*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
		*p++ = htonl(8);              /* attribute buffer length */
		p = xdr_encode_hyper(p, dentry->d_inode->i_ino);
	}
	
	*p++ = xdr_one;                                  /* next */
	*p++ = xdr_zero;                   /* cookie, first word */
	*p++ = xdr_two;                   /* cookie, second word */
	*p++ = xdr_two;                             /* entry len */
	memcpy(p, "..\0\0", 4);                         /* entry */
	p++;
	*p++ = xdr_one;                         /* bitmap length */
	*p++ = htonl(FATTR4_WORD0_FILEID);             /* bitmap */
	*p++ = htonl(8);              /* attribute buffer length */
	p = xdr_encode_hyper(p, dentry->d_parent->d_inode->i_ino);

	readdir->pgbase = (char *)p - (char *)start;
	readdir->count -= readdir->pgbase;
	kunmap_atomic(start, KM_USER0);
}

static void
renew_lease(struct nfs_server *server, unsigned long timestamp)
{
	struct nfs4_client *clp = server->nfs4_state;
	spin_lock(&clp->cl_lock);
	if (time_before(clp->cl_last_renewal,timestamp))
		clp->cl_last_renewal = timestamp;
	spin_unlock(&clp->cl_lock);
}

static void update_changeattr(struct inode *inode, struct nfs4_change_info *cinfo)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (cinfo->before == nfsi->change_attr && cinfo->atomic)
		nfsi->change_attr = cinfo->after;
}

static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, int open_flags)
{
	struct inode *inode = state->inode;

	open_flags &= (FMODE_READ|FMODE_WRITE);
	/* Protect against nfs4_find_state() */
	spin_lock(&inode->i_lock);
	state->state |= open_flags;
	/* NB! List reordering - see the reclaim code for why.  */
	if ((open_flags & FMODE_WRITE) && 0 == state->nwriters++)
		list_move(&state->open_states, &state->owner->so_states);
	if (open_flags & FMODE_READ)
		state->nreaders++;
	memcpy(&state->stateid, stateid, sizeof(state->stateid));
	spin_unlock(&inode->i_lock);
}

/*
 * OPEN_RECLAIM:
 * 	reclaim state on the server after a reboot.
 * 	Assumes caller is holding the sp->so_sem
 */
static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	struct inode *inode = state->inode;
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs_delegation *delegation = NFS_I(inode)->delegation;
	struct nfs_openargs o_arg = {
		.fh = NFS_FH(inode),
		.seqid = sp->so_seqid,
		.id = sp->so_id,
		.open_flags = state->state,
		.clientid = server->nfs4_state->cl_clientid,
		.claim = NFS4_OPEN_CLAIM_PREVIOUS,
		.bitmask = server->attr_bitmask,
	};
	struct nfs_openres o_res = {
		.server = server,	/* Grrr */
	};
	struct rpc_message msg = {
		.rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR],
		.rpc_argp       = &o_arg,
		.rpc_resp	= &o_res,
		.rpc_cred	= sp->so_cred,
	};
	int status;

	if (delegation != NULL) {
		if (!(delegation->flags & NFS_DELEGATION_NEED_RECLAIM)) {
			memcpy(&state->stateid, &delegation->stateid,
					sizeof(state->stateid));
			set_bit(NFS_DELEGATED_STATE, &state->flags);
			return 0;
		}
		o_arg.u.delegation_type = delegation->type;
	}
	status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
	nfs4_increment_seqid(status, sp);
	if (status == 0) {
		memcpy(&state->stateid, &o_res.stateid, sizeof(state->stateid));
		if (o_res.delegation_type != 0) {
			nfs_inode_reclaim_delegation(inode, sp->so_cred, &o_res);
			/* Did the server issue an immediate delegation recall? */
			if (o_res.do_recall)
				nfs_async_inode_return_delegation(inode, &o_res.stateid);
		}
	}
	clear_bit(NFS_DELEGATED_STATE, &state->flags);
	/* Ensure we update the inode attributes */
	NFS_CACHEINV(inode);
	return status;
}

static int nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_exception exception = { };
	int err;
	do {
		err = _nfs4_open_reclaim(sp, state);
		if (err != -NFS4ERR_DELAY)
			break;
		nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state)
{
	struct nfs4_state_owner  *sp  = state->owner;
	struct inode *inode = dentry->d_inode;
	struct nfs_server *server = NFS_SERVER(inode);
	struct dentry *parent = dget_parent(dentry);
	struct nfs_openargs arg = {
		.fh = NFS_FH(parent->d_inode),
		.clientid = server->nfs4_state->cl_clientid,
		.name = &dentry->d_name,
		.id = sp->so_id,
		.server = server,
		.bitmask = server->attr_bitmask,
		.claim = NFS4_OPEN_CLAIM_DELEGATE_CUR,
	};
	struct nfs_openres res = {
		.server = server,
	};
	struct 	rpc_message msg = {
		.rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR],
		.rpc_argp       = &arg,
		.rpc_resp       = &res,
		.rpc_cred	= sp->so_cred,
	};
	int status = 0;

	down(&sp->so_sema);
	if (!test_bit(NFS_DELEGATED_STATE, &state->flags))
		goto out;
	if (state->state == 0)
		goto out;
	arg.seqid = sp->so_seqid;
	arg.open_flags = state->state;
	memcpy(arg.u.delegation.data, state->stateid.data, sizeof(arg.u.delegation.data));
	status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
	nfs4_increment_seqid(status, sp);
	if (status >= 0) {
		memcpy(state->stateid.data, res.stateid.data,
				sizeof(state->stateid.data));
		clear_bit(NFS_DELEGATED_STATE, &state->flags);
	}
out:
	up(&sp->so_sema);
	dput(parent);
	return status;
}

int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state)
{
	struct nfs4_exception exception = { };
	struct nfs_server *server = NFS_SERVER(dentry->d_inode);
	int err;
	do {
		err = _nfs4_open_delegation_recall(dentry, state);
		switch (err) {
			case 0:
				return err;
			case -NFS4ERR_STALE_CLIENTID:
			case -NFS4ERR_STALE_STATEID:
			case -NFS4ERR_EXPIRED:
				/* Don't recall a delegation if it was lost */
				nfs4_schedule_state_recovery(server->nfs4_state);
				return err;
		}
		err = nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static inline int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid)
{
	struct nfs_open_confirmargs arg = {
		.fh             = fh,
		.seqid          = sp->so_seqid,
		.stateid	= *stateid,
	};
	struct nfs_open_confirmres res;
	struct 	rpc_message msg = {
		.rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_OPEN_CONFIRM],
		.rpc_argp       = &arg,
		.rpc_resp       = &res,
		.rpc_cred	= sp->so_cred,
	};
	int status;

	status = rpc_call_sync(clnt, &msg, RPC_TASK_NOINTR);
	nfs4_increment_seqid(status, sp);
	if (status >= 0)
		memcpy(stateid, &res.stateid, sizeof(*stateid));
	return status;
}

static int _nfs4_proc_open(struct inode *dir, struct nfs4_state_owner  *sp, struct nfs_openargs *o_arg, struct nfs_openres *o_res)
{
	struct nfs_server *server = NFS_SERVER(dir);
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN],
		.rpc_argp = o_arg,
		.rpc_resp = o_res,
		.rpc_cred = sp->so_cred,
	};
	int status;

	/* Update sequence id. The caller must serialize! */
	o_arg->seqid = sp->so_seqid;
	o_arg->id = sp->so_id;
	o_arg->clientid = sp->so_client->cl_clientid;

	status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
	nfs4_increment_seqid(status, sp);
	if (status != 0)
		goto out;
	update_changeattr(dir, &o_res->cinfo);
	if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
		status = _nfs4_proc_open_confirm(server->client, &o_res->fh,
				sp, &o_res->stateid);
		if (status != 0)
			goto out;
	}
	if (!(o_res->f_attr->valid & NFS_ATTR_FATTR))
		status = server->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr);
out:
	return status;
}

static int _nfs4_do_access(struct inode *inode, struct rpc_cred *cred, int openflags)
{
	struct nfs_access_entry cache;
	int mask = 0;
	int status;

	if (openflags & FMODE_READ)
		mask |= MAY_READ;
	if (openflags & FMODE_WRITE)
		mask |= MAY_WRITE;
	status = nfs_access_get_cached(inode, cred, &cache);
	if (status == 0)
		goto out;

	/* Be clever: ask server to check for all possible rights */
	cache.mask = MAY_EXEC | MAY_WRITE | MAY_READ;
	cache.cred = cred;
	cache.jiffies = jiffies;
	status = _nfs4_proc_access(inode, &cache);
	if (status != 0)
		return status;
	nfs_access_add_cache(inode, &cache);
out:
	if ((cache.mask & mask) == mask)
		return 0;
	return -EACCES;
}

/*
 * OPEN_EXPIRED:
 * 	reclaim state on the server after a network partition.
 * 	Assumes caller holds the appropriate lock
 */
static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state, struct dentry *dentry)
{
	struct dentry *parent = dget_parent(dentry);
	struct inode *dir = parent->d_inode;
	struct inode *inode = state->inode;
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs_delegation *delegation = NFS_I(inode)->delegation;
	struct nfs_fattr        f_attr = {
		.valid = 0,
	};
	struct nfs_openargs o_arg = {
		.fh = NFS_FH(dir),
		.open_flags = state->state,
		.name = &dentry->d_name,
		.bitmask = server->attr_bitmask,
		.claim = NFS4_OPEN_CLAIM_NULL,
	};
	struct nfs_openres o_res = {
		.f_attr = &f_attr,
		.server = server,
	};
	int status = 0;

	if (delegation != NULL && !(delegation->flags & NFS_DELEGATION_NEED_RECLAIM)) {
		status = _nfs4_do_access(inode, sp->so_cred, state->state);
		if (status < 0)
			goto out;
		memcpy(&state->stateid, &delegation->stateid, sizeof(state->stateid));
		set_bit(NFS_DELEGATED_STATE, &state->flags);
		goto out;
	}
	status = _nfs4_proc_open(dir, sp, &o_arg, &o_res);
	if (status != 0)
		goto out_nodeleg;
	/* Check if files differ */
	if ((f_attr.mode & S_IFMT) != (inode->i_mode & S_IFMT))
		goto out_stale;
	/* Has the file handle changed? */
	if (nfs_compare_fh(&o_res.fh, NFS_FH(inode)) != 0) {
		/* Verify if the change attributes are the same */
		if (f_attr.change_attr != NFS_I(inode)->change_attr)
			goto out_stale;
		if (nfs_size_to_loff_t(f_attr.size) != inode->i_size)
			goto out_stale;
		/* Lets just pretend that this is the same file */
		nfs_copy_fh(NFS_FH(inode), &o_res.fh);
		NFS_I(inode)->fileid = f_attr.fileid;
	}
	memcpy(&state->stateid, &o_res.stateid, sizeof(state->stateid));
	if (o_res.delegation_type != 0) {
		if (!(delegation->flags & NFS_DELEGATION_NEED_RECLAIM))
			nfs_inode_set_delegation(inode, sp->so_cred, &o_res);
		else
			nfs_inode_reclaim_delegation(inode, sp->so_cred, &o_res);
	}
out_nodeleg:
	clear_bit(NFS_DELEGATED_STATE, &state->flags);
out:
	dput(parent);
	return status;
out_stale:
	status = -ESTALE;
	/* Invalidate the state owner so we don't ever use it again */
	nfs4_drop_state_owner(sp);
	d_drop(dentry);
	/* Should we be trying to close that stateid? */
	goto out_nodeleg;
}

static inline int nfs4_do_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state, struct dentry *dentry)
{
	struct nfs_server *server = NFS_SERVER(dentry->d_inode);
	struct nfs4_exception exception = { };
	int err;

	do {
		err = _nfs4_open_expired(sp, state, dentry);
		if (err == -NFS4ERR_DELAY)
			nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static int nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *state)
{
	struct nfs_inode *nfsi = NFS_I(state->inode);
	struct nfs_open_context *ctx;
	int status;

	spin_lock(&state->inode->i_lock);
	list_for_each_entry(ctx, &nfsi->open_files, list) {
		if (ctx->state != state)
			continue;
		get_nfs_open_context(ctx);
		spin_unlock(&state->inode->i_lock);
		status = nfs4_do_open_expired(sp, state, ctx->dentry);
		put_nfs_open_context(ctx);
		return status;
	}
	spin_unlock(&state->inode->i_lock);
	return -ENOENT;
}

/*
 * Returns an nfs4_state + an extra reference to the inode
 */
static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred *cred, struct nfs4_state **res)
{
	struct nfs_delegation *delegation;
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_client *clp = server->nfs4_state;
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs4_state_owner *sp = NULL;
	struct nfs4_state *state = NULL;
	int open_flags = flags & (FMODE_READ|FMODE_WRITE);
	int err;

	/* Protect against reboot recovery - NOTE ORDER! */
	down_read(&clp->cl_sem);
	/* Protect against delegation recall */
	down_read(&nfsi->rwsem);
	delegation = NFS_I(inode)->delegation;
	err = -ENOENT;
	if (delegation == NULL || (delegation->type & open_flags) != open_flags)
		goto out_err;
	err = -ENOMEM;
	if (!(sp = nfs4_get_state_owner(server, cred))) {
		dprintk("%s: nfs4_get_state_owner failed!\n", __FUNCTION__);
		goto out_err;
	}
	down(&sp->so_sema);
	state = nfs4_get_open_state(inode, sp);
	if (state == NULL)
		goto out_err;

	err = -ENOENT;
	if ((state->state & open_flags) == open_flags) {
		spin_lock(&inode->i_lock);
		if (open_flags & FMODE_READ)
			state->nreaders++;
		if (open_flags & FMODE_WRITE)
			state->nwriters++;
		spin_unlock(&inode->i_lock);
		goto out_ok;
	} else if (state->state != 0)
		goto out_err;

	lock_kernel();
	err = _nfs4_do_access(inode, cred, open_flags);
	unlock_kernel();
	if (err != 0)
		goto out_err;
	set_bit(NFS_DELEGATED_STATE, &state->flags);
	update_open_stateid(state, &delegation->stateid, open_flags);
out_ok:
	up(&sp->so_sema);
	nfs4_put_state_owner(sp);
	up_read(&nfsi->rwsem);
	up_read(&clp->cl_sem);
	igrab(inode);
	*res = state;
	return 0; 
out_err:
	if (sp != NULL) {
		if (state != NULL)
			nfs4_put_open_state(state);
		up(&sp->so_sema);
		nfs4_put_state_owner(sp);
	}
	up_read(&nfsi->rwsem);
	up_read(&clp->cl_sem);
	return err;
}

static struct nfs4_state *nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred *cred)
{
	struct nfs4_exception exception = { };
	struct nfs4_state *res;
	int err;

	do {
		err = _nfs4_open_delegated(inode, flags, cred, &res);
		if (err == 0)
			break;
		res = ERR_PTR(nfs4_handle_exception(NFS_SERVER(inode),
					err, &exception));
	} while (exception.retry);
	return res;
}

/*
 * Returns an nfs4_state + an referenced inode
 */
static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, struct iattr *sattr, struct rpc_cred *cred, struct nfs4_state **res)
{
	struct nfs4_state_owner  *sp;
	struct nfs4_state     *state = NULL;
	struct nfs_server       *server = NFS_SERVER(dir);
	struct nfs4_client *clp = server->nfs4_state;
	struct inode *inode = NULL;
	int                     status;
	struct nfs_fattr        f_attr = {
		.valid          = 0,
	};
	struct nfs_openargs o_arg = {
		.fh             = NFS_FH(dir),
		.open_flags	= flags,
		.name           = &dentry->d_name,
		.server         = server,
		.bitmask = server->attr_bitmask,
		.claim = NFS4_OPEN_CLAIM_NULL,
	};
	struct nfs_openres o_res = {
		.f_attr         = &f_attr,
		.server         = server,
	};

	/* Protect against reboot recovery conflicts */
	down_read(&clp->cl_sem);
	status = -ENOMEM;
	if (!(sp = nfs4_get_state_owner(server, cred))) {
		dprintk("nfs4_do_open: nfs4_get_state_owner failed!\n");
		goto out_err;
	}
	if (flags & O_EXCL) {
		u32 *p = (u32 *) o_arg.u.verifier.data;
		p[0] = jiffies;
		p[1] = current->pid;
	} else
		o_arg.u.attrs = sattr;
	/* Serialization for the sequence id */
	down(&sp->so_sema);

	status = _nfs4_proc_open(dir, sp, &o_arg, &o_res);
	if (status != 0)
		goto out_err;

	status = -ENOMEM;
	inode = nfs_fhget(dir->i_sb, &o_res.fh, &f_attr);
	if (!inode)
		goto out_err;
	state = nfs4_get_open_state(inode, sp);
	if (!state)
		goto out_err;
	update_open_stateid(state, &o_res.stateid, flags);
	if (o_res.delegation_type != 0)
		nfs_inode_set_delegation(inode, cred, &o_res);
	up(&sp->so_sema);
	nfs4_put_state_owner(sp);
	up_read(&clp->cl_sem);
	*res = state;
	return 0;
out_err:
	if (sp != NULL) {
		if (state != NULL)
			nfs4_put_open_state(state);
		up(&sp->so_sema);
		nfs4_put_state_owner(sp);
	}
	/* Note: clp->cl_sem must be released before nfs4_put_open_state()! */
	up_read(&clp->cl_sem);
	if (inode != NULL)
		iput(inode);
	*res = NULL;
	return status;
}


static struct nfs4_state *nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, struct iattr *sattr, struct rpc_cred *cred)
{
	struct nfs4_exception exception = { };
	struct nfs4_state *res;
	int status;

	do {
		status = _nfs4_do_open(dir, dentry, flags, sattr, cred, &res);
		if (status == 0)
			break;
		/* NOTE: BAD_SEQID means the server and client disagree about the
		 * book-keeping w.r.t. state-changing operations
		 * (OPEN/CLOSE/LOCK/LOCKU...)
		 * It is actually a sign of a bug on the client or on the server.
		 *
		 * If we receive a BAD_SEQID error in the particular case of
		 * doing an OPEN, we assume that nfs4_increment_seqid() will
		 * have unhashed the old state_owner for us, and that we can
		 * therefore safely retry using a new one. We should still warn
		 * the user though...
		 */
		if (status == -NFS4ERR_BAD_SEQID) {
			printk(KERN_WARNING "NFS: v4 server returned a bad sequence-id error!\n");
			exception.retry = 1;
			continue;
		}
		res = ERR_PTR(nfs4_handle_exception(NFS_SERVER(dir),
					status, &exception));
	} while (exception.retry);
	return res;
}

static int _nfs4_do_setattr(struct nfs_server *server, struct nfs_fattr *fattr,
                struct nfs_fh *fhandle, struct iattr *sattr,
                struct nfs4_state *state)
{
        struct nfs_setattrargs  arg = {
                .fh             = fhandle,
                .iap            = sattr,
		.server		= server,
		.bitmask = server->attr_bitmask,
        };
        struct nfs_setattrres  res = {
		.fattr		= fattr,
		.server		= server,
        };
        struct rpc_message msg = {
                .rpc_proc       = &nfs4_procedures[NFSPROC4_CLNT_SETATTR],
                .rpc_argp       = &arg,
                .rpc_resp       = &res,
        };
	int status;

        fattr->valid = 0;

	if (state != NULL) {
		msg.rpc_cred = state->owner->so_cred;
		nfs4_copy_stateid(&arg.stateid, state, current->files);
	} else
		memcpy(&arg.stateid, &zero_stateid, sizeof(arg.stateid));

	status = rpc_call_sync(server->client, &msg, 0);
	return status;
}

static int nfs4_do_setattr(struct nfs_server *server, struct nfs_fattr *fattr,
                struct nfs_fh *fhandle, struct iattr *sattr,
                struct nfs4_state *state)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_do_setattr(server, fattr, fhandle, sattr,
					state),
				&exception);
	} while (exception.retry);
	return err;
}

struct nfs4_closedata {
	struct inode *inode;
	struct nfs4_state *state;
	struct nfs_closeargs arg;
	struct nfs_closeres res;
};

static void nfs4_close_done(struct rpc_task *task)
{
	struct nfs4_closedata *calldata = (struct nfs4_closedata *)task->tk_calldata;
	struct nfs4_state *state = calldata->state;
	struct nfs4_state_owner *sp = state->owner;
	struct nfs_server *server = NFS_SERVER(calldata->inode);

        /* hmm. we are done with the inode, and in the process of freeing
	 * the state_owner. we keep this around to process errors
	 */
	nfs4_increment_seqid(task->tk_status, sp);
	switch (task->tk_status) {
		case 0:
			memcpy(&state->stateid, &calldata->res.stateid,
					sizeof(state->stateid));
			break;
		case -NFS4ERR_STALE_STATEID:
		case -NFS4ERR_EXPIRED:
			state->state = calldata->arg.open_flags;
			nfs4_schedule_state_recovery(server->nfs4_state);
			break;
		default:
			if (nfs4_async_handle_error(task, server) == -EAGAIN) {
				rpc_restart_call(task);
				return;
			}
	}
	state->state = calldata->arg.open_flags;
	nfs4_put_open_state(state);
	up(&sp->so_sema);
	nfs4_put_state_owner(sp);
	up_read(&server->nfs4_state->cl_sem);
	kfree(calldata);
}

static inline int nfs4_close_call(struct rpc_clnt *clnt, struct nfs4_closedata *calldata)
{
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE],
		.rpc_argp = &calldata->arg,
		.rpc_resp = &calldata->res,
		.rpc_cred = calldata->state->owner->so_cred,
	};
	if (calldata->arg.open_flags != 0)
		msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
	return rpc_call_async(clnt, &msg, 0, nfs4_close_done, calldata);
}

/* 
 * It is possible for data to be read/written from a mem-mapped file 
 * after the sys_close call (which hits the vfs layer as a flush).
 * This means that we can't safely call nfsv4 close on a file until 
 * the inode is cleared. This in turn means that we are not good
 * NFSv4 citizens - we do not indicate to the server to update the file's 
 * share state even when we are done with one of the three share 
 * stateid's in the inode.
 *
 * NOTE: Caller must be holding the sp->so_owner semaphore!
 */
int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode) 
{
	struct nfs4_closedata *calldata;
	int status;

	/* Tell caller we're done */
	if (test_bit(NFS_DELEGATED_STATE, &state->flags)) {
		state->state = mode;
		return 0;
	}
	calldata = (struct nfs4_closedata *)kmalloc(sizeof(*calldata), GFP_KERNEL);
	if (calldata == NULL)
		return -ENOMEM;
	calldata->inode = inode;
	calldata->state = state;
	calldata->arg.fh = NFS_FH(inode);
	/* Serialization for the sequence id */
	calldata->arg.seqid = state->owner->so_seqid;
	calldata->arg.open_flags = mode;
	memcpy(&calldata->arg.stateid, &state->stateid,
			sizeof(calldata->arg.stateid));
	status = nfs4_close_call(NFS_SERVER(inode)->client, calldata);
	/*
	 * Return -EINPROGRESS on success in order to indicate to the
	 * caller that an asynchronous RPC call has been launched, and
	 * that it will release the semaphores on completion.
	 */
	return (status == 0) ? -EINPROGRESS : status;
}

struct inode *
nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
{
	struct iattr attr;
	struct rpc_cred *cred;
	struct nfs4_state *state;

	if (nd->flags & LOOKUP_CREATE) {
		attr.ia_mode = nd->intent.open.create_mode;
		attr.ia_valid = ATTR_MODE;
		if (!IS_POSIXACL(dir))
			attr.ia_mode &= ~current->fs->umask;
	} else {
		attr.ia_valid = 0;
		BUG_ON(nd->intent.open.flags & O_CREAT);
	}

	cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0);
	if (IS_ERR(cred))
		return (struct inode *)cred;
	state = nfs4_do_open(dir, dentry, nd->intent.open.flags, &attr, cred);
	put_rpccred(cred);
	if (IS_ERR(state))
		return (struct inode *)state;
	return state->inode;
}

int
nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags)
{
	struct rpc_cred *cred;
	struct nfs4_state *state;
	struct inode *inode;

	cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0);
	if (IS_ERR(cred))
		return PTR_ERR(cred);
	state = nfs4_open_delegated(dentry->d_inode, openflags, cred);
	if (IS_ERR(state))
		state = nfs4_do_open(dir, dentry, openflags, NULL, cred);
	put_rpccred(cred);
	if (state == ERR_PTR(-ENOENT) && dentry->d_inode == 0)
		return 1;
	if (IS_ERR(state))
		return 0;
	inode = state->inode;
	if (inode == dentry->d_inode) {
		iput(inode);
		return 1;
	}
	d_drop(dentry);
	nfs4_close_state(state, openflags);
	iput(inode);
	return 0;
}


static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
{
	struct nfs4_server_caps_res res = {};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS],
		.rpc_argp = fhandle,
		.rpc_resp = &res,
	};
	int status;

	status = rpc_call_sync(server->client, &msg, 0);
	if (status == 0) {
		memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
		if (res.attr_bitmask[0] & FATTR4_WORD0_ACL)
			server->caps |= NFS_CAP_ACLS;
		if (res.has_links != 0)
			server->caps |= NFS_CAP_HARDLINKS;
		if (res.has_symlinks != 0)
			server->caps |= NFS_CAP_SYMLINKS;
		server->acl_bitmask = res.acl_bitmask;
	}
	return status;
}

static int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_server_capabilities(server, fhandle),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_fsinfo *info)
{
	struct nfs_fattr *	fattr = info->fattr;
	struct nfs4_lookup_root_arg args = {
		.bitmask = nfs4_fattr_bitmap,
	};
	struct nfs4_lookup_res res = {
		.server = server,
		.fattr = fattr,
		.fh = fhandle,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP_ROOT],
		.rpc_argp = &args,
		.rpc_resp = &res,
	};
	fattr->valid = 0;
	return rpc_call_sync(server->client, &msg, 0);
}

static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_fsinfo *info)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_lookup_root(server, fhandle, info),
				&exception);
	} while (exception.retry);
	return err;
}

static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_fsinfo *info)
{
	struct nfs_fattr *	fattr = info->fattr;
	unsigned char *		p;
	struct qstr		q;
	struct nfs4_lookup_arg args = {
		.dir_fh = fhandle,
		.name = &q,
		.bitmask = nfs4_fattr_bitmap,
	};
	struct nfs4_lookup_res res = {
		.server = server,
		.fattr = fattr,
		.fh = fhandle,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
		.rpc_argp = &args,
		.rpc_resp = &res,
	};
	int status;

	/*
	 * Now we do a separate LOOKUP for each component of the mount path.
	 * The LOOKUPs are done separately so that we can conveniently
	 * catch an ERR_WRONGSEC if it occurs along the way...
	 */
	status = nfs4_lookup_root(server, fhandle, info);
	if (status)
		goto out;

	p = server->mnt_path;
	for (;;) {
		struct nfs4_exception exception = { };

		while (*p == '/')
			p++;
		if (!*p)
			break;
		q.name = p;
		while (*p && (*p != '/'))
			p++;
		q.len = p - q.name;

		do {
			fattr->valid = 0;
			status = nfs4_handle_exception(server,
					rpc_call_sync(server->client, &msg, 0),
					&exception);
		} while (exception.retry);
		if (status == 0)
			continue;
		if (status == -ENOENT) {
			printk(KERN_NOTICE "NFS: mount path %s does not exist!\n", server->mnt_path);
			printk(KERN_NOTICE "NFS: suggestion: try mounting '/' instead.\n");
		}
		break;
	}
	if (status == 0)
		status = nfs4_server_capabilities(server, fhandle);
	if (status == 0)
		status = nfs4_do_fsinfo(server, fhandle, info);
out:
	return status;
}

static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
{
	struct nfs4_getattr_arg args = {
		.fh = fhandle,
		.bitmask = server->attr_bitmask,
	};
	struct nfs4_getattr_res res = {
		.fattr = fattr,
		.server = server,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETATTR],
		.rpc_argp = &args,
		.rpc_resp = &res,
	};
	
	fattr->valid = 0;
	return rpc_call_sync(server->client, &msg, 0);
}

static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_proc_getattr(server, fhandle, fattr),
				&exception);
	} while (exception.retry);
	return err;
}

/* 
 * The file is not closed if it is opened due to the a request to change
 * the size of the file. The open call will not be needed once the
 * VFS layer lookup-intents are implemented.
 *
 * Close is called when the inode is destroyed.
 * If we haven't opened the file for O_WRONLY, we
 * need to in the size_change case to obtain a stateid.
 *
 * Got race?
 * Because OPEN is always done by name in nfsv4, it is
 * possible that we opened a different file by the same
 * name.  We can recognize this race condition, but we
 * can't do anything about it besides returning an error.
 *
 * This will be fixed with VFS changes (lookup-intent).
 */
static int
nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
		  struct iattr *sattr)
{
	struct rpc_cred *cred;
	struct inode *inode = dentry->d_inode;
	struct nfs4_state *state;
	int status;

	fattr->valid = 0;
	
	cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0);
	if (IS_ERR(cred))
		return PTR_ERR(cred);
	/* Search for an existing WRITE delegation first */
	state = nfs4_open_delegated(inode, FMODE_WRITE, cred);
	if (!IS_ERR(state)) {
		/* NB: nfs4_open_delegated() bumps the inode->i_count */
		iput(inode);
	} else {
		/* Search for an existing open(O_WRITE) stateid */
		state = nfs4_find_state(inode, cred, FMODE_WRITE);
	}

	status = nfs4_do_setattr(NFS_SERVER(inode), fattr,
			NFS_FH(inode), sattr, state);
	if (status == 0)
		nfs_setattr_update_inode(inode, sattr);
	if (state != NULL)
		nfs4_close_state(state, FMODE_WRITE);
	put_rpccred(cred);
	return status;
}

static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name,
		struct nfs_fh *fhandle, struct nfs_fattr *fattr)
{
	int		       status;
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs4_lookup_arg args = {
		.bitmask = server->attr_bitmask,
		.dir_fh = NFS_FH(dir),
		.name = name,
	};
	struct nfs4_lookup_res res = {
		.server = server,
		.fattr = fattr,
		.fh = fhandle,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
		.rpc_argp = &args,
		.rpc_resp = &res,
	};
	
	fattr->valid = 0;
	
	dprintk("NFS call  lookup %s\n", name->name);
	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	dprintk("NFS reply lookup: %d\n", status);
	return status;
}

static int nfs4_proc_lookup(struct inode *dir, struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dir),
				_nfs4_proc_lookup(dir, name, fhandle, fattr),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry)
{
	struct nfs4_accessargs args = {
		.fh = NFS_FH(inode),
	};
	struct nfs4_accessres res = { 0 };
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_ACCESS],
		.rpc_argp = &args,
		.rpc_resp = &res,
		.rpc_cred = entry->cred,
	};
	int mode = entry->mask;
	int status;

	/*
	 * Determine which access bits we want to ask for...
	 */
	if (mode & MAY_READ)
		args.access |= NFS4_ACCESS_READ;
	if (S_ISDIR(inode->i_mode)) {
		if (mode & MAY_WRITE)
			args.access |= NFS4_ACCESS_MODIFY | NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE;
		if (mode & MAY_EXEC)
			args.access |= NFS4_ACCESS_LOOKUP;
	} else {
		if (mode & MAY_WRITE)
			args.access |= NFS4_ACCESS_MODIFY | NFS4_ACCESS_EXTEND;
		if (mode & MAY_EXEC)
			args.access |= NFS4_ACCESS_EXECUTE;
	}
	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
	if (!status) {
		entry->mask = 0;
		if (res.access & NFS4_ACCESS_READ)
			entry->mask |= MAY_READ;
		if (res.access & (NFS4_ACCESS_MODIFY | NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE))
			entry->mask |= MAY_WRITE;
		if (res.access & (NFS4_ACCESS_LOOKUP|NFS4_ACCESS_EXECUTE))
			entry->mask |= MAY_EXEC;
	}
	return status;
}

static int nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(inode),
				_nfs4_proc_access(inode, entry),
				&exception);
	} while (exception.retry);
	return err;
}

/*
 * TODO: For the time being, we don't try to get any attributes
 * along with any of the zero-copy operations READ, READDIR,
 * READLINK, WRITE.
 *
 * In the case of the first three, we want to put the GETATTR
 * after the read-type operation -- this is because it is hard
 * to predict the length of a GETATTR response in v4, and thus
 * align the READ data correctly.  This means that the GETATTR
 * may end up partially falling into the page cache, and we should
 * shift it into the 'tail' of the xdr_buf before processing.
 * To do this efficiently, we need to know the total length
 * of data received, which doesn't seem to be available outside
 * of the RPC layer.
 *
 * In the case of WRITE, we also want to put the GETATTR after
 * the operation -- in this case because we want to make sure
 * we get the post-operation mtime and size.  This means that
 * we can't use xdr_encode_pages() as written: we need a variant
 * of it which would leave room in the 'tail' iovec.
 *
 * Both of these changes to the XDR layer would in fact be quite
 * minor, but I decided to leave them for a subsequent patch.
 */
static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
		unsigned int pgbase, unsigned int pglen)
{
	struct nfs4_readlink args = {
		.fh       = NFS_FH(inode),
		.pgbase	  = pgbase,
		.pglen    = pglen,
		.pages    = &page,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READLINK],
		.rpc_argp = &args,
		.rpc_resp = NULL,
	};

	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
}

static int nfs4_proc_readlink(struct inode *inode, struct page *page,
		unsigned int pgbase, unsigned int pglen)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(inode),
				_nfs4_proc_readlink(inode, page, pgbase, pglen),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_read(struct nfs_read_data *rdata)
{
	int flags = rdata->flags;
	struct inode *inode = rdata->inode;
	struct nfs_fattr *fattr = rdata->res.fattr;
	struct nfs_server *server = NFS_SERVER(inode);
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_READ],
		.rpc_argp	= &rdata->args,
		.rpc_resp	= &rdata->res,
		.rpc_cred	= rdata->cred,
	};
	unsigned long timestamp = jiffies;
	int status;

	dprintk("NFS call  read %d @ %Ld\n", rdata->args.count,
			(long long) rdata->args.offset);

	fattr->valid = 0;
	status = rpc_call_sync(server->client, &msg, flags);
	if (!status)
		renew_lease(server, timestamp);
	dprintk("NFS reply read: %d\n", status);
	return status;
}

static int nfs4_proc_read(struct nfs_read_data *rdata)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(rdata->inode),
				_nfs4_proc_read(rdata),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_write(struct nfs_write_data *wdata)
{
	int rpcflags = wdata->flags;
	struct inode *inode = wdata->inode;
	struct nfs_fattr *fattr = wdata->res.fattr;
	struct nfs_server *server = NFS_SERVER(inode);
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_WRITE],
		.rpc_argp	= &wdata->args,
		.rpc_resp	= &wdata->res,
		.rpc_cred	= wdata->cred,
	};
	int status;

	dprintk("NFS call  write %d @ %Ld\n", wdata->args.count,
			(long long) wdata->args.offset);

	fattr->valid = 0;
	status = rpc_call_sync(server->client, &msg, rpcflags);
	dprintk("NFS reply write: %d\n", status);
	return status;
}

static int nfs4_proc_write(struct nfs_write_data *wdata)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(wdata->inode),
				_nfs4_proc_write(wdata),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_commit(struct nfs_write_data *cdata)
{
	struct inode *inode = cdata->inode;
	struct nfs_fattr *fattr = cdata->res.fattr;
	struct nfs_server *server = NFS_SERVER(inode);
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_COMMIT],
		.rpc_argp	= &cdata->args,
		.rpc_resp	= &cdata->res,
		.rpc_cred	= cdata->cred,
	};
	int status;

	dprintk("NFS call  commit %d @ %Ld\n", cdata->args.count,
			(long long) cdata->args.offset);

	fattr->valid = 0;
	status = rpc_call_sync(server->client, &msg, 0);
	dprintk("NFS reply commit: %d\n", status);
	return status;
}

static int nfs4_proc_commit(struct nfs_write_data *cdata)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(cdata->inode),
				_nfs4_proc_commit(cdata),
				&exception);
	} while (exception.retry);
	return err;
}

/*
 * Got race?
 * We will need to arrange for the VFS layer to provide an atomic open.
 * Until then, this create/open method is prone to inefficiency and race
 * conditions due to the lookup, create, and open VFS calls from sys_open()
 * placed on the wire.
 *
 * Given the above sorry state of affairs, I'm simply sending an OPEN.
 * The file will be opened again in the subsequent VFS open call
 * (nfs4_proc_file_open).
 *
 * The open for read will just hang around to be used by any process that
 * opens the file O_RDONLY. This will all be resolved with the VFS changes.
 */

static int
nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
                 int flags)
{
	struct nfs4_state *state;
	struct rpc_cred *cred;
	int status = 0;

	cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0);
	if (IS_ERR(cred)) {
		status = PTR_ERR(cred);
		goto out;
	}
	state = nfs4_do_open(dir, dentry, flags, sattr, cred);
	put_rpccred(cred);
	if (IS_ERR(state)) {
		status = PTR_ERR(state);
		goto out;
	}
	d_instantiate(dentry, state->inode);
	if (flags & O_EXCL) {
		struct nfs_fattr fattr;
		status = nfs4_do_setattr(NFS_SERVER(dir), &fattr,
		                     NFS_FH(state->inode), sattr, state);
		if (status == 0) {
			nfs_setattr_update_inode(state->inode, sattr);
			goto out;
		}
	} else if (flags != 0)
		goto out;
	nfs4_close_state(state, flags);
out:
	return status;
}

static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
{
	struct nfs4_remove_arg args = {
		.fh = NFS_FH(dir),
		.name = name,
	};
	struct nfs4_change_info	res;
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_REMOVE],
		.rpc_argp	= &args,
		.rpc_resp	= &res,
	};
	int			status;

	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	if (status == 0)
		update_changeattr(dir, &res);
	return status;
}

static int nfs4_proc_remove(struct inode *dir, struct qstr *name)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dir),
				_nfs4_proc_remove(dir, name),
				&exception);
	} while (exception.retry);
	return err;
}

struct unlink_desc {
	struct nfs4_remove_arg	args;
	struct nfs4_change_info	res;
};

static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir,
		struct qstr *name)
{
	struct unlink_desc *up;

	up = (struct unlink_desc *) kmalloc(sizeof(*up), GFP_KERNEL);
	if (!up)
		return -ENOMEM;
	
	up->args.fh = NFS_FH(dir->d_inode);
	up->args.name = name;
	
	msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE];
	msg->rpc_argp = &up->args;
	msg->rpc_resp = &up->res;
	return 0;
}

static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task)
{
	struct rpc_message *msg = &task->tk_msg;
	struct unlink_desc *up;
	
	if (msg->rpc_resp != NULL) {
		up = container_of(msg->rpc_resp, struct unlink_desc, res);
		update_changeattr(dir->d_inode, &up->res);
		kfree(up);
		msg->rpc_resp = NULL;
		msg->rpc_argp = NULL;
	}
	return 0;
}

static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
		struct inode *new_dir, struct qstr *new_name)
{
	struct nfs4_rename_arg arg = {
		.old_dir = NFS_FH(old_dir),
		.new_dir = NFS_FH(new_dir),
		.old_name = old_name,
		.new_name = new_name,
	};
	struct nfs4_rename_res res = { };
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME],
		.rpc_argp = &arg,
		.rpc_resp = &res,
	};
	int			status;
	
	status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0);

	if (!status) {
		update_changeattr(old_dir, &res.old_cinfo);
		update_changeattr(new_dir, &res.new_cinfo);
	}
	return status;
}

static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
		struct inode *new_dir, struct qstr *new_name)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(old_dir),
				_nfs4_proc_rename(old_dir, old_name,
					new_dir, new_name),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
{
	struct nfs4_link_arg arg = {
		.fh     = NFS_FH(inode),
		.dir_fh = NFS_FH(dir),
		.name   = name,
	};
	struct nfs4_change_info	cinfo = { };
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK],
		.rpc_argp = &arg,
		.rpc_resp = &cinfo,
	};
	int			status;

	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
	if (!status)
		update_changeattr(dir, &cinfo);

	return status;
}

static int nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(inode),
				_nfs4_proc_link(inode, dir, name),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name,
		struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle,
		struct nfs_fattr *fattr)
{
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs4_create_arg arg = {
		.dir_fh = NFS_FH(dir),
		.server = server,
		.name = name,
		.attrs = sattr,
		.ftype = NF4LNK,
		.bitmask = server->attr_bitmask,
	};
	struct nfs4_create_res res = {
		.server = server,
		.fh = fhandle,
		.fattr = fattr,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SYMLINK],
		.rpc_argp = &arg,
		.rpc_resp = &res,
	};
	int			status;

	if (path->len > NFS4_MAXPATHLEN)
		return -ENAMETOOLONG;
	arg.u.symlink = path;
	fattr->valid = 0;
	
	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	if (!status)
		update_changeattr(dir, &res.dir_cinfo);
	return status;
}

static int nfs4_proc_symlink(struct inode *dir, struct qstr *name,
		struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle,
		struct nfs_fattr *fattr)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dir),
				_nfs4_proc_symlink(dir, name, path, sattr,
					fhandle, fattr),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry,
		struct iattr *sattr)
{
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs_fh fhandle;
	struct nfs_fattr fattr;
	struct nfs4_create_arg arg = {
		.dir_fh = NFS_FH(dir),
		.server = server,
		.name = &dentry->d_name,
		.attrs = sattr,
		.ftype = NF4DIR,
		.bitmask = server->attr_bitmask,
	};
	struct nfs4_create_res res = {
		.server = server,
		.fh = &fhandle,
		.fattr = &fattr,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE],
		.rpc_argp = &arg,
		.rpc_resp = &res,
	};
	int			status;

	fattr.valid = 0;
	
	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	if (!status) {
		update_changeattr(dir, &res.dir_cinfo);
		status = nfs_instantiate(dentry, &fhandle, &fattr);
	}
	return status;
}

static int nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry,
		struct iattr *sattr)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dir),
				_nfs4_proc_mkdir(dir, dentry, sattr),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
                  u64 cookie, struct page *page, unsigned int count, int plus)
{
	struct inode		*dir = dentry->d_inode;
	struct nfs4_readdir_arg args = {
		.fh = NFS_FH(dir),
		.pages = &page,
		.pgbase = 0,
		.count = count,
		.bitmask = NFS_SERVER(dentry->d_inode)->attr_bitmask,
	};
	struct nfs4_readdir_res res;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READDIR],
		.rpc_argp = &args,
		.rpc_resp = &res,
		.rpc_cred = cred,
	};
	int			status;

	dprintk("%s: dentry = %s/%s, cookie = %Lu\n", __FUNCTION__,
			dentry->d_parent->d_name.name,
			dentry->d_name.name,
			(unsigned long long)cookie);
	lock_kernel();
	nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args);
	res.pgbase = args.pgbase;
	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	if (status == 0)
		memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE);
	unlock_kernel();
	dprintk("%s: returns %d\n", __FUNCTION__, status);
	return status;
}

static int nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
                  u64 cookie, struct page *page, unsigned int count, int plus)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dentry->d_inode),
				_nfs4_proc_readdir(dentry, cred, cookie,
					page, count, plus),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
		struct iattr *sattr, dev_t rdev)
{
	struct nfs_server *server = NFS_SERVER(dir);
	struct nfs_fh fh;
	struct nfs_fattr fattr;
	struct nfs4_create_arg arg = {
		.dir_fh = NFS_FH(dir),
		.server = server,
		.name = &dentry->d_name,
		.attrs = sattr,
		.bitmask = server->attr_bitmask,
	};
	struct nfs4_create_res res = {
		.server = server,
		.fh = &fh,
		.fattr = &fattr,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE],
		.rpc_argp = &arg,
		.rpc_resp = &res,
	};
	int			status;
	int                     mode = sattr->ia_mode;

	fattr.valid = 0;

	BUG_ON(!(sattr->ia_valid & ATTR_MODE));
	BUG_ON(!S_ISFIFO(mode) && !S_ISBLK(mode) && !S_ISCHR(mode) && !S_ISSOCK(mode));
	if (S_ISFIFO(mode))
		arg.ftype = NF4FIFO;
	else if (S_ISBLK(mode)) {
		arg.ftype = NF4BLK;
		arg.u.device.specdata1 = MAJOR(rdev);
		arg.u.device.specdata2 = MINOR(rdev);
	}
	else if (S_ISCHR(mode)) {
		arg.ftype = NF4CHR;
		arg.u.device.specdata1 = MAJOR(rdev);
		arg.u.device.specdata2 = MINOR(rdev);
	}
	else
		arg.ftype = NF4SOCK;
	
	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
	if (status == 0) {
		update_changeattr(dir, &res.dir_cinfo);
		status = nfs_instantiate(dentry, &fh, &fattr);
	}
	return status;
}

static int nfs4_proc_mknod(struct inode *dir, struct dentry *dentry,
		struct iattr *sattr, dev_t rdev)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(NFS_SERVER(dir),
				_nfs4_proc_mknod(dir, dentry, sattr, rdev),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
		 struct nfs_fsstat *fsstat)
{
	struct nfs4_statfs_arg args = {
		.fh = fhandle,
		.bitmask = server->attr_bitmask,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_STATFS],
		.rpc_argp = &args,
		.rpc_resp = fsstat,
	};

	fsstat->fattr->valid = 0;
	return rpc_call_sync(server->client, &msg, 0);
}

static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat)
{
	struct nfs4_exception exception = { };
	int err;
	do {
		err = nfs4_handle_exception(server,
				_nfs4_proc_statfs(server, fhandle, fsstat),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_fsinfo *fsinfo)
{
	struct nfs4_fsinfo_arg args = {
		.fh = fhandle,
		.bitmask = server->attr_bitmask,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSINFO],
		.rpc_argp = &args,
		.rpc_resp = fsinfo,
	};

	return rpc_call_sync(server->client, &msg, 0);
}

static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
{
	struct nfs4_exception exception = { };
	int err;

	do {
		err = nfs4_handle_exception(server,
				_nfs4_do_fsinfo(server, fhandle, fsinfo),
				&exception);
	} while (exception.retry);
	return err;
}

static int nfs4_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
{
	fsinfo->fattr->valid = 0;
	return nfs4_do_fsinfo(server, fhandle, fsinfo);
}

static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_pathconf *pathconf)
{
	struct nfs4_pathconf_arg args = {
		.fh = fhandle,
		.bitmask = server->attr_bitmask,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_PATHCONF],
		.rpc_argp = &args,
		.rpc_resp = pathconf,
	};

	/* None of the pathconf attributes are mandatory to implement */
	if ((args.bitmask[0] & nfs4_pathconf_bitmap[0]) == 0) {
		memset(pathconf, 0, sizeof(*pathconf));
		return 0;
	}

	pathconf->fattr->valid = 0;
	return rpc_call_sync(server->client, &msg, 0);
}

static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
		struct nfs_pathconf *pathconf)
{
	struct nfs4_exception exception = { };
	int err;

	do {
		err = nfs4_handle_exception(server,
				_nfs4_proc_pathconf(server, fhandle, pathconf),
				&exception);
	} while (exception.retry);
	return err;
}

static void
nfs4_read_done(struct rpc_task *task)
{
	struct nfs_read_data *data = (struct nfs_read_data *) task->tk_calldata;
	struct inode *inode = data->inode;

	if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
		rpc_restart_call(task);
		return;
	}
	if (task->tk_status > 0)
		renew_lease(NFS_SERVER(inode), data->timestamp);
	/* Call back common NFS readpage processing */
	nfs_readpage_result(task);
}

static void
nfs4_proc_read_setup(struct nfs_read_data *data)
{
	struct rpc_task	*task = &data->task;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ],
		.rpc_argp = &data->args,
		.rpc_resp = &data->res,
		.rpc_cred = data->cred,
	};
	struct inode *inode = data->inode;
	int flags;

	data->timestamp   = jiffies;

	/* N.B. Do we need to test? Never called for swapfile inode */
	flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0);

	/* Finalize the task. */
	rpc_init_task(task, NFS_CLIENT(inode), nfs4_read_done, flags);
	rpc_call_setup(task, &msg, 0);
}

static void
nfs4_write_done(struct rpc_task *task)
{
	struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
	struct inode *inode = data->inode;
	
	if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
		rpc_restart_call(task);
		return;
	}
	if (task->tk_status >= 0)
		renew_lease(NFS_SERVER(inode), data->timestamp);
	/* Call back common NFS writeback processing */
	nfs_writeback_done(task);
}

static void
nfs4_proc_write_setup(struct nfs_write_data *data, int how)
{
	struct rpc_task	*task = &data->task;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE],
		.rpc_argp = &data->args,
		.rpc_resp = &data->res,
		.rpc_cred = data->cred,
	};
	struct inode *inode = data->inode;
	int stable;
	int flags;
	
	if (how & FLUSH_STABLE) {
		if (!NFS_I(inode)->ncommit)
			stable = NFS_FILE_SYNC;
		else
			stable = NFS_DATA_SYNC;
	} else
		stable = NFS_UNSTABLE;
	data->args.stable = stable;

	data->timestamp   = jiffies;

	/* Set the initial flags for the task.  */
	flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;

	/* Finalize the task. */
	rpc_init_task(task, NFS_CLIENT(inode), nfs4_write_done, flags);
	rpc_call_setup(task, &msg, 0);
}

static void
nfs4_commit_done(struct rpc_task *task)
{
	struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
	struct inode *inode = data->inode;
	
	if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
		rpc_restart_call(task);
		return;
	}
	/* Call back common NFS writeback processing */
	nfs_commit_done(task);
}

static void
nfs4_proc_commit_setup(struct nfs_write_data *data, int how)
{
	struct rpc_task	*task = &data->task;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT],
		.rpc_argp = &data->args,
		.rpc_resp = &data->res,
		.rpc_cred = data->cred,
	};	
	struct inode *inode = data->inode;
	int flags;
	
	/* Set the initial flags for the task.  */
	flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;

	/* Finalize the task. */
	rpc_init_task(task, NFS_CLIENT(inode), nfs4_commit_done, flags);
	rpc_call_setup(task, &msg, 0);	
}

/*
 * nfs4_proc_async_renew(): This is not one of the nfs_rpc_ops; it is a special
 * standalone procedure for queueing an asynchronous RENEW.
 */
static void
renew_done(struct rpc_task *task)
{
	struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp;
	unsigned long timestamp = (unsigned long)task->tk_calldata;

	if (task->tk_status < 0) {
		switch (task->tk_status) {
			case -NFS4ERR_STALE_CLIENTID:
			case -NFS4ERR_EXPIRED:
			case -NFS4ERR_CB_PATH_DOWN:
				nfs4_schedule_state_recovery(clp);
		}
		return;
	}
	spin_lock(&clp->cl_lock);
	if (time_before(clp->cl_last_renewal,timestamp))
		clp->cl_last_renewal = timestamp;
	spin_unlock(&clp->cl_lock);
}

int
nfs4_proc_async_renew(struct nfs4_client *clp)
{
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
		.rpc_argp	= clp,
		.rpc_cred	= clp->cl_cred,
	};

	return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT,
			renew_done, (void *)jiffies);
}

int
nfs4_proc_renew(struct nfs4_client *clp)
{
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
		.rpc_argp	= clp,
		.rpc_cred	= clp->cl_cred,
	};
	unsigned long now = jiffies;
	int status;

	status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
	if (status < 0)
		return status;
	spin_lock(&clp->cl_lock);
	if (time_before(clp->cl_last_renewal,now))
		clp->cl_last_renewal = now;
	spin_unlock(&clp->cl_lock);
	return 0;
}

/*
 * We will need to arrange for the VFS layer to provide an atomic open.
 * Until then, this open method is prone to inefficiency and race conditions
 * due to the lookup, potential create, and open VFS calls from sys_open()
 * placed on the wire.
 */
static int
nfs4_proc_file_open(struct inode *inode, struct file *filp)
{
	struct dentry *dentry = filp->f_dentry;
	struct nfs_open_context *ctx;
	struct nfs4_state *state = NULL;
	struct rpc_cred *cred;
	int status = -ENOMEM;

	dprintk("nfs4_proc_file_open: starting on (%.*s/%.*s)\n",
	                       (int)dentry->d_parent->d_name.len,
	                       dentry->d_parent->d_name.name,
	                       (int)dentry->d_name.len, dentry->d_name.name);


	/* Find our open stateid */
	cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0);
	if (IS_ERR(cred))
		return PTR_ERR(cred);
	ctx = alloc_nfs_open_context(dentry, cred);
	put_rpccred(cred);
	if (unlikely(ctx == NULL))
		return -ENOMEM;
	status = -EIO; /* ERACE actually */
	state = nfs4_find_state(inode, cred, filp->f_mode);
	if (unlikely(state == NULL))
		goto no_state;
	ctx->state = state;
	nfs4_close_state(state, filp->f_mode);
	ctx->mode = filp->f_mode;
	nfs_file_set_open_context(filp, ctx);
	put_nfs_open_context(ctx);
	if (filp->f_mode & FMODE_WRITE)
		nfs_begin_data_update(inode);
	return 0;
no_state:
	printk(KERN_WARNING "NFS: v4 raced in function %s\n", __FUNCTION__);
	put_nfs_open_context(ctx);
	return status;
}

/*
 * Release our state
 */
static int
nfs4_proc_file_release(struct inode *inode, struct file *filp)
{
	if (filp->f_mode & FMODE_WRITE)
		nfs_end_data_update(inode);
	nfs_file_clear_open_context(filp);
	return 0;
}

static inline int nfs4_server_supports_acls(struct nfs_server *server)
{
	return (server->caps & NFS_CAP_ACLS)
		&& (server->acl_bitmask & ACL4_SUPPORT_ALLOW_ACL)
		&& (server->acl_bitmask & ACL4_SUPPORT_DENY_ACL);
}

/* Assuming that XATTR_SIZE_MAX is a multiple of PAGE_CACHE_SIZE, and that
 * it's OK to put sizeof(void) * (XATTR_SIZE_MAX/PAGE_CACHE_SIZE) bytes on
 * the stack.
 */
#define NFS4ACL_MAXPAGES (XATTR_SIZE_MAX >> PAGE_CACHE_SHIFT)

static void buf_to_pages(const void *buf, size_t buflen,
		struct page **pages, unsigned int *pgbase)
{
	const void *p = buf;

	*pgbase = offset_in_page(buf);
	p -= *pgbase;
	while (p < buf + buflen) {
		*(pages++) = virt_to_page(p);
		p += PAGE_CACHE_SIZE;
	}
}

struct nfs4_cached_acl {
	int cached;
	size_t len;
	char data[0];
};

static void nfs4_set_cached_acl(struct inode *inode, struct nfs4_cached_acl *acl)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	spin_lock(&inode->i_lock);
	kfree(nfsi->nfs4_acl);
	nfsi->nfs4_acl = acl;
	spin_unlock(&inode->i_lock);
}

static void nfs4_zap_acl_attr(struct inode *inode)
{
	nfs4_set_cached_acl(inode, NULL);
}

static inline ssize_t nfs4_read_cached_acl(struct inode *inode, char *buf, size_t buflen)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs4_cached_acl *acl;
	int ret = -ENOENT;

	spin_lock(&inode->i_lock);
	acl = nfsi->nfs4_acl;
	if (acl == NULL)
		goto out;
	if (buf == NULL) /* user is just asking for length */
		goto out_len;
	if (acl->cached == 0)
		goto out;
	ret = -ERANGE; /* see getxattr(2) man page */
	if (acl->len > buflen)
		goto out;
	memcpy(buf, acl->data, acl->len);
out_len:
	ret = acl->len;
out:
	spin_unlock(&inode->i_lock);
	return ret;
}

static void nfs4_write_cached_acl(struct inode *inode, const char *buf, size_t acl_len)
{
	struct nfs4_cached_acl *acl;

	if (buf && acl_len <= PAGE_SIZE) {
		acl = kmalloc(sizeof(*acl) + acl_len, GFP_KERNEL);
		if (acl == NULL)
			goto out;
		acl->cached = 1;
		memcpy(acl->data, buf, acl_len);
	} else {
		acl = kmalloc(sizeof(*acl), GFP_KERNEL);
		if (acl == NULL)
			goto out;
		acl->cached = 0;
	}
	acl->len = acl_len;
out:
	nfs4_set_cached_acl(inode, acl);
}

static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen)
{
	struct page *pages[NFS4ACL_MAXPAGES];
	struct nfs_getaclargs args = {
		.fh = NFS_FH(inode),
		.acl_pages = pages,
		.acl_len = buflen,
	};
	size_t resp_len = buflen;
	void *resp_buf;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL],
		.rpc_argp = &args,
		.rpc_resp = &resp_len,
	};
	struct page *localpage = NULL;
	int ret;

	if (buflen < PAGE_SIZE) {
		/* As long as we're doing a round trip to the server anyway,
		 * let's be prepared for a page of acl data. */
		localpage = alloc_page(GFP_KERNEL);
		resp_buf = page_address(localpage);
		if (localpage == NULL)
			return -ENOMEM;
		args.acl_pages[0] = localpage;
		args.acl_pgbase = 0;
		args.acl_len = PAGE_SIZE;
	} else {
		resp_buf = buf;
		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
	}
	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
	if (ret)
		goto out_free;
	if (resp_len > args.acl_len)
		nfs4_write_cached_acl(inode, NULL, resp_len);
	else
		nfs4_write_cached_acl(inode, resp_buf, resp_len);
	if (buf) {
		ret = -ERANGE;
		if (resp_len > buflen)
			goto out_free;
		if (localpage)
			memcpy(buf, resp_buf, resp_len);
	}
	ret = resp_len;
out_free:
	if (localpage)
		__free_page(localpage);
	return ret;
}

static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen)
{
	struct nfs_server *server = NFS_SERVER(inode);
	int ret;

	if (!nfs4_server_supports_acls(server))
		return -EOPNOTSUPP;
	ret = nfs_revalidate_inode(server, inode);
	if (ret < 0)
		return ret;
	ret = nfs4_read_cached_acl(inode, buf, buflen);
	if (ret != -ENOENT)
		return ret;
	return nfs4_get_acl_uncached(inode, buf, buflen);
}

static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen)
{
	struct nfs_server *server = NFS_SERVER(inode);
	struct page *pages[NFS4ACL_MAXPAGES];
	struct nfs_setaclargs arg = {
		.fh		= NFS_FH(inode),
		.acl_pages	= pages,
		.acl_len	= buflen,
	};
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_SETACL],
		.rpc_argp	= &arg,
		.rpc_resp	= NULL,
	};
	int ret;

	if (!nfs4_server_supports_acls(server))
		return -EOPNOTSUPP;
	buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
	ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0);
	if (ret == 0)
		nfs4_write_cached_acl(inode, buf, buflen);
	return ret;
}

static int
nfs4_async_handle_error(struct rpc_task *task, struct nfs_server *server)
{
	struct nfs4_client *clp = server->nfs4_state;

	if (!clp || task->tk_status >= 0)
		return 0;
	switch(task->tk_status) {
		case -NFS4ERR_STALE_CLIENTID:
		case -NFS4ERR_STALE_STATEID:
		case -NFS4ERR_EXPIRED:
			rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL, NULL);
			nfs4_schedule_state_recovery(clp);
			if (test_bit(NFS4CLNT_OK, &clp->cl_state))
				rpc_wake_up_task(task);
			task->tk_status = 0;
			return -EAGAIN;
		case -NFS4ERR_GRACE:
		case -NFS4ERR_DELAY:
			rpc_delay(task, NFS4_POLL_RETRY_MAX);
			task->tk_status = 0;
			return -EAGAIN;
		case -NFS4ERR_OLD_STATEID:
			task->tk_status = 0;
			return -EAGAIN;
	}
	task->tk_status = nfs4_map_errors(task->tk_status);
	return 0;
}

static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp)
{
	DEFINE_WAIT(wait);
	sigset_t oldset;
	int interruptible, res = 0;

	might_sleep();

	rpc_clnt_sigmask(clnt, &oldset);
	interruptible = TASK_UNINTERRUPTIBLE;
	if (clnt->cl_intr)
		interruptible = TASK_INTERRUPTIBLE;
	prepare_to_wait(&clp->cl_waitq, &wait, interruptible);
	nfs4_schedule_state_recovery(clp);
	if (clnt->cl_intr && signalled())
		res = -ERESTARTSYS;
	else if (!test_bit(NFS4CLNT_OK, &clp->cl_state))
		schedule();
	finish_wait(&clp->cl_waitq, &wait);
	rpc_clnt_sigunmask(clnt, &oldset);
	return res;
}

static int nfs4_delay(struct rpc_clnt *clnt, long *timeout)
{
	sigset_t oldset;
	int res = 0;

	might_sleep();

	if (*timeout <= 0)
		*timeout = NFS4_POLL_RETRY_MIN;
	if (*timeout > NFS4_POLL_RETRY_MAX)
		*timeout = NFS4_POLL_RETRY_MAX;
	rpc_clnt_sigmask(clnt, &oldset);
	if (clnt->cl_intr) {
		set_current_state(TASK_INTERRUPTIBLE);
		schedule_timeout(*timeout);
		if (signalled())
			res = -ERESTARTSYS;
	} else {
		set_current_state(TASK_UNINTERRUPTIBLE);
		schedule_timeout(*timeout);
	}
	rpc_clnt_sigunmask(clnt, &oldset);
	*timeout <<= 1;
	return res;
}

/* This is the error handling routine for processes that are allowed
 * to sleep.
 */
int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
{
	struct nfs4_client *clp = server->nfs4_state;
	int ret = errorcode;

	exception->retry = 0;
	switch(errorcode) {
		case 0:
			return 0;
		case -NFS4ERR_STALE_CLIENTID:
		case -NFS4ERR_STALE_STATEID:
		case -NFS4ERR_EXPIRED:
			ret = nfs4_wait_clnt_recover(server->client, clp);
			if (ret == 0)
				exception->retry = 1;
			break;
		case -NFS4ERR_GRACE:
		case -NFS4ERR_DELAY:
			ret = nfs4_delay(server->client, &exception->timeout);
			if (ret == 0)
				exception->retry = 1;
			break;
		case -NFS4ERR_OLD_STATEID:
			if (ret == 0)
				exception->retry = 1;
	}
	/* We failed to handle the error */
	return nfs4_map_errors(ret);
}

int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short port)
{
	nfs4_verifier sc_verifier;
	struct nfs4_setclientid setclientid = {
		.sc_verifier = &sc_verifier,
		.sc_prog = program,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID],
		.rpc_argp = &setclientid,
		.rpc_resp = clp,
		.rpc_cred = clp->cl_cred,
	};
	u32 *p;
	int loop = 0;
	int status;

	p = (u32*)sc_verifier.data;
	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
	*p = htonl((u32)clp->cl_boot_time.tv_nsec);

	for(;;) {
		setclientid.sc_name_len = scnprintf(setclientid.sc_name,
				sizeof(setclientid.sc_name), "%s/%u.%u.%u.%u %s %u",
				clp->cl_ipaddr, NIPQUAD(clp->cl_addr.s_addr),
				clp->cl_cred->cr_ops->cr_name,
				clp->cl_id_uniquifier);
		setclientid.sc_netid_len = scnprintf(setclientid.sc_netid,
				sizeof(setclientid.sc_netid), "tcp");
		setclientid.sc_uaddr_len = scnprintf(setclientid.sc_uaddr,
				sizeof(setclientid.sc_uaddr), "%s.%d.%d",
				clp->cl_ipaddr, port >> 8, port & 255);

		status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
		if (status != -NFS4ERR_CLID_INUSE)
			break;
		if (signalled())
			break;
		if (loop++ & 1)
			ssleep(clp->cl_lease_time + 1);
		else
			if (++clp->cl_id_uniquifier == 0)
				break;
	}
	return status;
}

int
nfs4_proc_setclientid_confirm(struct nfs4_client *clp)
{
	struct nfs_fsinfo fsinfo;
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SETCLIENTID_CONFIRM],
		.rpc_argp = clp,
		.rpc_resp = &fsinfo,
		.rpc_cred = clp->cl_cred,
	};
	unsigned long now;
	int status;

	now = jiffies;
	status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
	if (status == 0) {
		spin_lock(&clp->cl_lock);
		clp->cl_lease_time = fsinfo.lease_time * HZ;
		clp->cl_last_renewal = now;
		spin_unlock(&clp->cl_lock);
	}
	return status;
}

static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid)
{
	struct nfs4_delegreturnargs args = {
		.fhandle = NFS_FH(inode),
		.stateid = stateid,
	};
	struct rpc_message msg = {
		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DELEGRETURN],
		.rpc_argp = &args,
		.rpc_cred = cred,
	};

	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
}

int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid)
{
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_exception exception = { };
	int err;
	do {
		err = _nfs4_proc_delegreturn(inode, cred, stateid);
		switch (err) {
			case -NFS4ERR_STALE_STATEID:
			case -NFS4ERR_EXPIRED:
				nfs4_schedule_state_recovery(server->nfs4_state);
			case 0:
				return 0;
		}
		err = nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

#define NFS4_LOCK_MINTIMEOUT (1 * HZ)
#define NFS4_LOCK_MAXTIMEOUT (30 * HZ)

/* 
 * sleep, with exponential backoff, and retry the LOCK operation. 
 */
static unsigned long
nfs4_set_lock_task_retry(unsigned long timeout)
{
	current->state = TASK_INTERRUPTIBLE;
	schedule_timeout(timeout);
	timeout <<= 1;
	if (timeout > NFS4_LOCK_MAXTIMEOUT)
		return NFS4_LOCK_MAXTIMEOUT;
	return timeout;
}

static inline int
nfs4_lck_type(int cmd, struct file_lock *request)
{
	/* set lock type */
	switch (request->fl_type) {
		case F_RDLCK:
			return IS_SETLKW(cmd) ? NFS4_READW_LT : NFS4_READ_LT;
		case F_WRLCK:
			return IS_SETLKW(cmd) ? NFS4_WRITEW_LT : NFS4_WRITE_LT;
		case F_UNLCK:
			return NFS4_WRITE_LT; 
	}
	BUG();
	return 0;
}

static inline uint64_t
nfs4_lck_length(struct file_lock *request)
{
	if (request->fl_end == OFFSET_MAX)
		return ~(uint64_t)0;
	return request->fl_end - request->fl_start + 1;
}

static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct inode *inode = state->inode;
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_client *clp = server->nfs4_state;
	struct nfs_lockargs arg = {
		.fh = NFS_FH(inode),
		.type = nfs4_lck_type(cmd, request),
		.offset = request->fl_start,
		.length = nfs4_lck_length(request),
	};
	struct nfs_lockres res = {
		.server = server,
	};
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_LOCKT],
		.rpc_argp       = &arg,
		.rpc_resp       = &res,
		.rpc_cred	= state->owner->so_cred,
	};
	struct nfs_lowner nlo;
	struct nfs4_lock_state *lsp;
	int status;

	down_read(&clp->cl_sem);
	nlo.clientid = clp->cl_clientid;
	down(&state->lock_sema);
	status = nfs4_set_lock_state(state, request);
	if (status != 0)
		goto out;
	lsp = request->fl_u.nfs4_fl.owner;
	nlo.id = lsp->ls_id; 
	arg.u.lockt = &nlo;
	status = rpc_call_sync(server->client, &msg, 0);
	if (!status) {
		request->fl_type = F_UNLCK;
	} else if (status == -NFS4ERR_DENIED) {
		int64_t len, start, end;
		start = res.u.denied.offset;
		len = res.u.denied.length;
		end = start + len - 1;
		if (end < 0 || len == 0)
			request->fl_end = OFFSET_MAX;
		else
			request->fl_end = (loff_t)end;
		request->fl_start = (loff_t)start;
		request->fl_type = F_WRLCK;
		if (res.u.denied.type & 1)
			request->fl_type = F_RDLCK;
		request->fl_pid = 0;
		status = 0;
	}
out:
	up(&state->lock_sema);
	up_read(&clp->cl_sem);
	return status;
}

static int nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct nfs4_exception exception = { };
	int err;

	do {
		err = nfs4_handle_exception(NFS_SERVER(state->inode),
				_nfs4_proc_getlk(state, cmd, request),
				&exception);
	} while (exception.retry);
	return err;
}

static int do_vfs_lock(struct file *file, struct file_lock *fl)
{
	int res = 0;
	switch (fl->fl_flags & (FL_POSIX|FL_FLOCK)) {
		case FL_POSIX:
			res = posix_lock_file_wait(file, fl);
			break;
		case FL_FLOCK:
			res = flock_lock_file_wait(file, fl);
			break;
		default:
			BUG();
	}
	if (res < 0)
		printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n",
				__FUNCTION__);
	return res;
}

static int _nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct inode *inode = state->inode;
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_client *clp = server->nfs4_state;
	struct nfs_lockargs arg = {
		.fh = NFS_FH(inode),
		.type = nfs4_lck_type(cmd, request),
		.offset = request->fl_start,
		.length = nfs4_lck_length(request),
	};
	struct nfs_lockres res = {
		.server = server,
	};
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_LOCKU],
		.rpc_argp       = &arg,
		.rpc_resp       = &res,
		.rpc_cred	= state->owner->so_cred,
	};
	struct nfs4_lock_state *lsp;
	struct nfs_locku_opargs luargs;
	int status;
			
	down_read(&clp->cl_sem);
	down(&state->lock_sema);
	status = nfs4_set_lock_state(state, request);
	if (status != 0)
		goto out;
	lsp = request->fl_u.nfs4_fl.owner;
	/* We might have lost the locks! */
	if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0)
		goto out;
	luargs.seqid = lsp->ls_seqid;
	memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid));
	arg.u.locku = &luargs;
	status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
	nfs4_increment_lock_seqid(status, lsp);

	if (status == 0)
		memcpy(&lsp->ls_stateid,  &res.u.stateid, 
				sizeof(lsp->ls_stateid));
out:
	up(&state->lock_sema);
	if (status == 0)
		do_vfs_lock(request->fl_file, request);
	up_read(&clp->cl_sem);
	return status;
}

static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct nfs4_exception exception = { };
	int err;

	do {
		err = nfs4_handle_exception(NFS_SERVER(state->inode),
				_nfs4_proc_unlck(state, cmd, request),
				&exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *request, int reclaim)
{
	struct inode *inode = state->inode;
	struct nfs_server *server = NFS_SERVER(inode);
	struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner;
	struct nfs_lockargs arg = {
		.fh = NFS_FH(inode),
		.type = nfs4_lck_type(cmd, request),
		.offset = request->fl_start,
		.length = nfs4_lck_length(request),
	};
	struct nfs_lockres res = {
		.server = server,
	};
	struct rpc_message msg = {
		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_LOCK],
		.rpc_argp       = &arg,
		.rpc_resp       = &res,
		.rpc_cred	= state->owner->so_cred,
	};
	struct nfs_lock_opargs largs = {
		.reclaim = reclaim,
		.new_lock_owner = 0,
	};
	int status;

	if (!(lsp->ls_flags & NFS_LOCK_INITIALIZED)) {
		struct nfs4_state_owner *owner = state->owner;
		struct nfs_open_to_lock otl = {
			.lock_owner = {
				.clientid = server->nfs4_state->cl_clientid,
			},
		};

		otl.lock_seqid = lsp->ls_seqid;
		otl.lock_owner.id = lsp->ls_id;
		memcpy(&otl.open_stateid, &state->stateid, sizeof(otl.open_stateid));
		largs.u.open_lock = &otl;
		largs.new_lock_owner = 1;
		arg.u.lock = &largs;
		down(&owner->so_sema);
		otl.open_seqid = owner->so_seqid;
		status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
		/* increment open_owner seqid on success, and 
		* seqid mutating errors */
		nfs4_increment_seqid(status, owner);
		up(&owner->so_sema);
		if (status == 0) {
			lsp->ls_flags |= NFS_LOCK_INITIALIZED;
			lsp->ls_seqid++;
		}
	} else {
		struct nfs_exist_lock el = {
			.seqid = lsp->ls_seqid,
		};
		memcpy(&el.stateid, &lsp->ls_stateid, sizeof(el.stateid));
		largs.u.exist_lock = &el;
		arg.u.lock = &largs;
		status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR);
		/* increment seqid on success, and * seqid mutating errors*/
		nfs4_increment_lock_seqid(status, lsp);
	}
	/* save the returned stateid. */
	if (status == 0)
		memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(nfs4_stateid));
	else if (status == -NFS4ERR_DENIED)
		status = -EAGAIN;
	return status;
}

static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_exception exception = { };
	int err;

	do {
		err = _nfs4_do_setlk(state, F_SETLK, request, 1);
		if (err != -NFS4ERR_DELAY)
			break;
		nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request)
{
	struct nfs_server *server = NFS_SERVER(state->inode);
	struct nfs4_exception exception = { };
	int err;

	do {
		err = _nfs4_do_setlk(state, F_SETLK, request, 0);
		if (err != -NFS4ERR_DELAY)
			break;
		nfs4_handle_exception(server, err, &exception);
	} while (exception.retry);
	return err;
}

static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct nfs4_client *clp = state->owner->so_client;
	int status;

	down_read(&clp->cl_sem);
	down(&state->lock_sema);
	status = nfs4_set_lock_state(state, request);
	if (status == 0)
		status = _nfs4_do_setlk(state, cmd, request, 0);
	up(&state->lock_sema);
	if (status == 0) {
		/* Note: we always want to sleep here! */
		request->fl_flags |= FL_SLEEP;
		if (do_vfs_lock(request->fl_file, request) < 0)
			printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__);
	}
	up_read(&clp->cl_sem);
	return status;
}

static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
{
	struct nfs4_exception exception = { };
	int err;

	do {
		err = nfs4_handle_exception(NFS_SERVER(state->inode),
				_nfs4_proc_setlk(state, cmd, request),
				&exception);
	} while (exception.retry);
	return err;
}

static int
nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request)
{
	struct nfs_open_context *ctx;
	struct nfs4_state *state;
	unsigned long timeout = NFS4_LOCK_MINTIMEOUT;
	int status;

	/* verify open state */
	ctx = (struct nfs_open_context *)filp->private_data;
	state = ctx->state;

	if (request->fl_start < 0 || request->fl_end < 0)
		return -EINVAL;

	if (IS_GETLK(cmd))
		return nfs4_proc_getlk(state, F_GETLK, request);

	if (!(IS_SETLK(cmd) || IS_SETLKW(cmd)))
		return -EINVAL;

	if (request->fl_type == F_UNLCK)
		return nfs4_proc_unlck(state, cmd, request);

	do {
		status = nfs4_proc_setlk(state, cmd, request);
		if ((status != -EAGAIN) || IS_SETLK(cmd))
			break;
		timeout = nfs4_set_lock_task_retry(timeout);
		status = -ERESTARTSYS;
		if (signalled())
			break;
	} while(status < 0);
	return status;
}


#define XATTR_NAME_NFSV4_ACL "system.nfs4_acl"

int nfs4_setxattr(struct dentry *dentry, const char *key, const void *buf,
		size_t buflen, int flags)
{
	struct inode *inode = dentry->d_inode;

	if (strcmp(key, XATTR_NAME_NFSV4_ACL) != 0)
		return -EOPNOTSUPP;

	if (!S_ISREG(inode->i_mode) &&
	    (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX))
		return -EPERM;

	return nfs4_proc_set_acl(inode, buf, buflen);
}

/* The getxattr man page suggests returning -ENODATA for unknown attributes,
 * and that's what we'll do for e.g. user attributes that haven't been set.
 * But we'll follow ext2/ext3's lead by returning -EOPNOTSUPP for unsupported
 * attributes in kernel-managed attribute namespaces. */
ssize_t nfs4_getxattr(struct dentry *dentry, const char *key, void *buf,
		size_t buflen)
{
	struct inode *inode = dentry->d_inode;

	if (strcmp(key, XATTR_NAME_NFSV4_ACL) != 0)
		return -EOPNOTSUPP;

	return nfs4_proc_get_acl(inode, buf, buflen);
}

ssize_t nfs4_listxattr(struct dentry *dentry, char *buf, size_t buflen)
{
	size_t len = strlen(XATTR_NAME_NFSV4_ACL) + 1;

	if (buf && buflen < len)
		return -ERANGE;
	if (buf)
		memcpy(buf, XATTR_NAME_NFSV4_ACL, len);
	return len;
}

struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
	.recover_open	= nfs4_open_reclaim,
	.recover_lock	= nfs4_lock_reclaim,
};

struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops = {
	.recover_open	= nfs4_open_expired,
	.recover_lock	= nfs4_lock_expired,
};

static struct inode_operations nfs4_file_inode_operations = {
	.permission	= nfs_permission,
	.getattr	= nfs_getattr,
	.setattr	= nfs_setattr,
	.getxattr	= nfs4_getxattr,
	.setxattr	= nfs4_setxattr,
	.listxattr	= nfs4_listxattr,
};

struct nfs_rpc_ops	nfs_v4_clientops = {
	.version	= 4,			/* protocol version */
	.dentry_ops	= &nfs4_dentry_operations,
	.dir_inode_ops	= &nfs4_dir_inode_operations,
	.file_inode_ops	= &nfs4_file_inode_operations,
	.getroot	= nfs4_proc_get_root,
	.getattr	= nfs4_proc_getattr,
	.setattr	= nfs4_proc_setattr,
	.lookup		= nfs4_proc_lookup,
	.access		= nfs4_proc_access,
	.readlink	= nfs4_proc_readlink,
	.read		= nfs4_proc_read,
	.write		= nfs4_proc_write,
	.commit		= nfs4_proc_commit,
	.create		= nfs4_proc_create,
	.remove		= nfs4_proc_remove,
	.unlink_setup	= nfs4_proc_unlink_setup,
	.unlink_done	= nfs4_proc_unlink_done,
	.rename		= nfs4_proc_rename,
	.link		= nfs4_proc_link,
	.symlink	= nfs4_proc_symlink,
	.mkdir		= nfs4_proc_mkdir,
	.rmdir		= nfs4_proc_remove,
	.readdir	= nfs4_proc_readdir,
	.mknod		= nfs4_proc_mknod,
	.statfs		= nfs4_proc_statfs,
	.fsinfo		= nfs4_proc_fsinfo,
	.pathconf	= nfs4_proc_pathconf,
	.decode_dirent	= nfs4_decode_dirent,
	.read_setup	= nfs4_proc_read_setup,
	.write_setup	= nfs4_proc_write_setup,
	.commit_setup	= nfs4_proc_commit_setup,
	.file_open      = nfs4_proc_file_open,
	.file_release   = nfs4_proc_file_release,
	.lock		= nfs4_proc_lock,
	.clear_acl_cache = nfs4_zap_acl_attr,
};

/*
 * Local variables:
 *  c-basic-offset: 8
 * End:
 */