aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorSteven Whitehouse <swhiteho@redhat.com>2010-11-03 16:01:07 -0400
committerSteven Whitehouse <swhiteho@redhat.com>2010-11-15 07:44:42 -0500
commit044b9414c7caf9a26192c73a5b88fa1a8a32a1c1 (patch)
tree9596bb669a68b04eebc40864c3b3fd71d3d1e273
parent0143832cc96d0bf78486297aad5c8fb2c2ead02a (diff)
GFS2: Fix inode deallocation race
This area of the code has always been a bit delicate due to the subtleties of lock ordering. The problem is that for "normal" alloc/dealloc, we always grab the inode locks first and the rgrp lock later. In order to ensure no races in looking up the unlinked, but still allocated inodes, we need to hold the rgrp lock when we do the lookup, which means that we can't take the inode glock. The solution is to borrow the technique already used by NFS to solve what is essentially the same problem (given an inode number, look up the inode carefully, checking that it really is in the expected state). We cannot do that directly from the allocation code (lock ordering again) so we give the job to the pre-existing delete workqueue and carry on with the allocation as normal. If we find there is no space, we do a journal flush (required anyway if space from a deallocation is to be released) which should block against the pending deallocations, so we should always get the space back. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
-rw-r--r--fs/gfs2/export.c46
-rw-r--r--fs/gfs2/glock.c21
-rw-r--r--fs/gfs2/inode.c152
-rw-r--r--fs/gfs2/inode.h4
-rw-r--r--fs/gfs2/rgrp.c91
5 files changed, 98 insertions, 216 deletions
diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c
index 06d582732d34..5ab3839dfcb9 100644
--- a/fs/gfs2/export.c
+++ b/fs/gfs2/export.c
@@ -138,10 +138,8 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb,
138 struct gfs2_inum_host *inum) 138 struct gfs2_inum_host *inum)
139{ 139{
140 struct gfs2_sbd *sdp = sb->s_fs_info; 140 struct gfs2_sbd *sdp = sb->s_fs_info;
141 struct gfs2_holder i_gh;
142 struct inode *inode; 141 struct inode *inode;
143 struct dentry *dentry; 142 struct dentry *dentry;
144 int error;
145 143
146 inode = gfs2_ilookup(sb, inum->no_addr); 144 inode = gfs2_ilookup(sb, inum->no_addr);
147 if (inode) { 145 if (inode) {
@@ -152,52 +150,16 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb,
152 goto out_inode; 150 goto out_inode;
153 } 151 }
154 152
155 error = gfs2_glock_nq_num(sdp, inum->no_addr, &gfs2_inode_glops, 153 inode = gfs2_lookup_by_inum(sdp, inum->no_addr, &inum->no_formal_ino,
156 LM_ST_SHARED, LM_FLAG_ANY, &i_gh); 154 GFS2_BLKST_DINODE);
157 if (error) 155 if (IS_ERR(inode))
158 return ERR_PTR(error); 156 return ERR_CAST(inode);
159
160 error = gfs2_check_blk_type(sdp, inum->no_addr, GFS2_BLKST_DINODE);
161 if (error)
162 goto fail;
163
164 inode = gfs2_inode_lookup(sb, DT_UNKNOWN, inum->no_addr, 0);
165 if (IS_ERR(inode)) {
166 error = PTR_ERR(inode);
167 goto fail;
168 }
169
170 error = gfs2_inode_refresh(GFS2_I(inode));
171 if (error) {
172 iput(inode);
173 goto fail;
174 }
175
176 /* Pick up the works we bypass in gfs2_inode_lookup */
177 if (inode->i_state & I_NEW)
178 gfs2_set_iop(inode);
179
180 if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) {
181 iput(inode);
182 goto fail;
183 }
184
185 error = -EIO;
186 if (GFS2_I(inode)->i_diskflags & GFS2_DIF_SYSTEM) {
187 iput(inode);
188 goto fail;
189 }
190
191 gfs2_glock_dq_uninit(&i_gh);
192 157
193out_inode: 158out_inode:
194 dentry = d_obtain_alias(inode); 159 dentry = d_obtain_alias(inode);
195 if (!IS_ERR(dentry)) 160 if (!IS_ERR(dentry))
196 dentry->d_op = &gfs2_dops; 161 dentry->d_op = &gfs2_dops;
197 return dentry; 162 return dentry;
198fail:
199 gfs2_glock_dq_uninit(&i_gh);
200 return ERR_PTR(error);
201} 163}
202 164
203static struct dentry *gfs2_fh_to_dentry(struct super_block *sb, struct fid *fid, 165static struct dentry *gfs2_fh_to_dentry(struct super_block *sb, struct fid *fid,
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 87778857f099..f92c17704169 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -686,21 +686,20 @@ static void delete_work_func(struct work_struct *work)
686{ 686{
687 struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_delete); 687 struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_delete);
688 struct gfs2_sbd *sdp = gl->gl_sbd; 688 struct gfs2_sbd *sdp = gl->gl_sbd;
689 struct gfs2_inode *ip = NULL; 689 struct gfs2_inode *ip;
690 struct inode *inode; 690 struct inode *inode;
691 u64 no_addr = 0; 691 u64 no_addr = gl->gl_name.ln_number;
692
693 ip = gl->gl_object;
694 /* Note: Unsafe to dereference ip as we don't hold right refs/locks */
692 695
693 spin_lock(&gl->gl_spin);
694 ip = (struct gfs2_inode *)gl->gl_object;
695 if (ip) 696 if (ip)
696 no_addr = ip->i_no_addr;
697 spin_unlock(&gl->gl_spin);
698 if (ip) {
699 inode = gfs2_ilookup(sdp->sd_vfs, no_addr); 697 inode = gfs2_ilookup(sdp->sd_vfs, no_addr);
700 if (inode) { 698 else
701 d_prune_aliases(inode); 699 inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED);
702 iput(inode); 700 if (inode && !IS_ERR(inode)) {
703 } 701 d_prune_aliases(inode);
702 iput(inode);
704 } 703 }
705 gfs2_glock_put(gl); 704 gfs2_glock_put(gl);
706} 705}
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 06370f8bd8cf..e1213f7f9217 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -73,49 +73,6 @@ static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr)
73 return iget5_locked(sb, hash, iget_test, iget_set, &no_addr); 73 return iget5_locked(sb, hash, iget_test, iget_set, &no_addr);
74} 74}
75 75
76struct gfs2_skip_data {
77 u64 no_addr;
78 int skipped;
79};
80
81static int iget_skip_test(struct inode *inode, void *opaque)
82{
83 struct gfs2_inode *ip = GFS2_I(inode);
84 struct gfs2_skip_data *data = opaque;
85
86 if (ip->i_no_addr == data->no_addr) {
87 if (inode->i_state & (I_FREEING|I_WILL_FREE)){
88 data->skipped = 1;
89 return 0;
90 }
91 return 1;
92 }
93 return 0;
94}
95
96static int iget_skip_set(struct inode *inode, void *opaque)
97{
98 struct gfs2_inode *ip = GFS2_I(inode);
99 struct gfs2_skip_data *data = opaque;
100
101 if (data->skipped)
102 return 1;
103 inode->i_ino = (unsigned long)(data->no_addr);
104 ip->i_no_addr = data->no_addr;
105 return 0;
106}
107
108static struct inode *gfs2_iget_skip(struct super_block *sb,
109 u64 no_addr)
110{
111 struct gfs2_skip_data data;
112 unsigned long hash = (unsigned long)no_addr;
113
114 data.no_addr = no_addr;
115 data.skipped = 0;
116 return iget5_locked(sb, hash, iget_skip_test, iget_skip_set, &data);
117}
118
119/** 76/**
120 * GFS2 lookup code fills in vfs inode contents based on info obtained 77 * GFS2 lookup code fills in vfs inode contents based on info obtained
121 * from directory entry inside gfs2_inode_lookup(). This has caused issues 78 * from directory entry inside gfs2_inode_lookup(). This has caused issues
@@ -243,93 +200,54 @@ fail:
243 return ERR_PTR(error); 200 return ERR_PTR(error);
244} 201}
245 202
246/** 203struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
247 * gfs2_process_unlinked_inode - Lookup an unlinked inode for reclamation 204 u64 *no_formal_ino, unsigned int blktype)
248 * and try to reclaim it by doing iput.
249 *
250 * This function assumes no rgrp locks are currently held.
251 *
252 * @sb: The super block
253 * no_addr: The inode number
254 *
255 */
256
257void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr)
258{ 205{
259 struct gfs2_sbd *sdp; 206 struct super_block *sb = sdp->sd_vfs;
260 struct gfs2_inode *ip; 207 struct gfs2_holder i_gh;
261 struct gfs2_glock *io_gl = NULL;
262 int error;
263 struct gfs2_holder gh;
264 struct inode *inode; 208 struct inode *inode;
209 int error;
265 210
266 inode = gfs2_iget_skip(sb, no_addr); 211 error = gfs2_glock_nq_num(sdp, no_addr, &gfs2_inode_glops,
267 212 LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
268 if (!inode) 213 if (error)
269 return; 214 return ERR_PTR(error);
270
271 /* If it's not a new inode, someone's using it, so leave it alone. */
272 if (!(inode->i_state & I_NEW)) {
273 iput(inode);
274 return;
275 }
276
277 ip = GFS2_I(inode);
278 sdp = GFS2_SB(inode);
279 ip->i_no_formal_ino = -1;
280 215
281 error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &ip->i_gl); 216 error = gfs2_check_blk_type(sdp, no_addr, blktype);
282 if (unlikely(error)) 217 if (error)
283 goto fail; 218 goto fail;
284 ip->i_gl->gl_object = ip;
285 219
286 error = gfs2_glock_get(sdp, no_addr, &gfs2_iopen_glops, CREATE, &io_gl); 220 inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0);
287 if (unlikely(error)) 221 if (IS_ERR(inode))
288 goto fail_put; 222 goto fail;
289
290 set_bit(GIF_INVALID, &ip->i_flags);
291 error = gfs2_glock_nq_init(io_gl, LM_ST_SHARED, LM_FLAG_TRY | GL_EXACT,
292 &ip->i_iopen_gh);
293 if (unlikely(error))
294 goto fail_iopen;
295 223
296 ip->i_iopen_gh.gh_gl->gl_object = ip; 224 error = gfs2_inode_refresh(GFS2_I(inode));
297 gfs2_glock_put(io_gl); 225 if (error)
298 io_gl = NULL; 226 goto fail_iput;
299 227
300 inode->i_mode = DT2IF(DT_UNKNOWN); 228 /* Pick up the works we bypass in gfs2_inode_lookup */
229 if (inode->i_state & I_NEW)
230 gfs2_set_iop(inode);
301 231
302 /* 232 /* Two extra checks for NFS only */
303 * We must read the inode in order to work out its type in 233 if (no_formal_ino) {
304 * this case. Note that this doesn't happen often as we normally 234 error = -ESTALE;
305 * know the type beforehand. This code path only occurs during 235 if (GFS2_I(inode)->i_no_formal_ino != *no_formal_ino)
306 * unlinked inode recovery (where it is safe to do this glock, 236 goto fail_iput;
307 * which is not true in the general case).
308 */
309 error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, LM_FLAG_TRY,
310 &gh);
311 if (unlikely(error))
312 goto fail_glock;
313 237
314 /* Inode is now uptodate */ 238 error = -EIO;
315 gfs2_glock_dq_uninit(&gh); 239 if (GFS2_I(inode)->i_diskflags & GFS2_DIF_SYSTEM)
316 gfs2_set_iop(inode); 240 goto fail_iput;
317 241
318 /* The iput will cause it to be deleted. */ 242 error = 0;
319 iput(inode); 243 }
320 return;
321 244
322fail_glock:
323 gfs2_glock_dq(&ip->i_iopen_gh);
324fail_iopen:
325 if (io_gl)
326 gfs2_glock_put(io_gl);
327fail_put:
328 ip->i_gl->gl_object = NULL;
329 gfs2_glock_put(ip->i_gl);
330fail: 245fail:
331 iget_failed(inode); 246 gfs2_glock_dq_uninit(&i_gh);
332 return; 247 return error ? ERR_PTR(error) : inode;
248fail_iput:
249 iput(inode);
250 goto fail;
333} 251}
334 252
335static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) 253static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 6720d7d5fbc6..d8499fadcc53 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -99,7 +99,9 @@ err:
99extern void gfs2_set_iop(struct inode *inode); 99extern void gfs2_set_iop(struct inode *inode);
100extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 100extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type,
101 u64 no_addr, u64 no_formal_ino); 101 u64 no_addr, u64 no_formal_ino);
102extern void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr); 102extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
103 u64 *no_formal_ino,
104 unsigned int blktype);
103extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr); 105extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr);
104 106
105extern int gfs2_inode_refresh(struct gfs2_inode *ip); 107extern int gfs2_inode_refresh(struct gfs2_inode *ip);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index bef3ab6cf5c1..33c8407b876f 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -963,17 +963,18 @@ static int try_rgrp_fit(struct gfs2_rgrpd *rgd, struct gfs2_alloc *al)
963 * The inode, if one has been found, in inode. 963 * The inode, if one has been found, in inode.
964 */ 964 */
965 965
966static u64 try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, 966static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip)
967 u64 skip)
968{ 967{
969 u32 goal = 0, block; 968 u32 goal = 0, block;
970 u64 no_addr; 969 u64 no_addr;
971 struct gfs2_sbd *sdp = rgd->rd_sbd; 970 struct gfs2_sbd *sdp = rgd->rd_sbd;
972 unsigned int n; 971 unsigned int n;
972 struct gfs2_glock *gl;
973 struct gfs2_inode *ip;
974 int error;
975 int found = 0;
973 976
974 for(;;) { 977 while (goal < rgd->rd_data) {
975 if (goal >= rgd->rd_data)
976 break;
977 down_write(&sdp->sd_log_flush_lock); 978 down_write(&sdp->sd_log_flush_lock);
978 n = 1; 979 n = 1;
979 block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED, 980 block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED,
@@ -990,11 +991,32 @@ static u64 try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked,
990 if (no_addr == skip) 991 if (no_addr == skip)
991 continue; 992 continue;
992 *last_unlinked = no_addr; 993 *last_unlinked = no_addr;
993 return no_addr; 994
995 error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &gl);
996 if (error)
997 continue;
998
999 /* If the inode is already in cache, we can ignore it here
1000 * because the existing inode disposal code will deal with
1001 * it when all refs have gone away. Accessing gl_object like
1002 * this is not safe in general. Here it is ok because we do
1003 * not dereference the pointer, and we only need an approx
1004 * answer to whether it is NULL or not.
1005 */
1006 ip = gl->gl_object;
1007
1008 if (ip || queue_work(gfs2_delete_workqueue, &gl->gl_delete) == 0)
1009 gfs2_glock_put(gl);
1010 else
1011 found++;
1012
1013 /* Limit reclaim to sensible number of tasks */
1014 if (found > 2*NR_CPUS)
1015 return;
994 } 1016 }
995 1017
996 rgd->rd_flags &= ~GFS2_RDF_CHECK; 1018 rgd->rd_flags &= ~GFS2_RDF_CHECK;
997 return 0; 1019 return;
998} 1020}
999 1021
1000/** 1022/**
@@ -1075,11 +1097,9 @@ static void forward_rgrp_set(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd)
1075 * Try to acquire rgrp in way which avoids contending with others. 1097 * Try to acquire rgrp in way which avoids contending with others.
1076 * 1098 *
1077 * Returns: errno 1099 * Returns: errno
1078 * unlinked: the block address of an unlinked block to be reclaimed
1079 */ 1100 */
1080 1101
1081static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked, 1102static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked)
1082 u64 *last_unlinked)
1083{ 1103{
1084 struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); 1104 struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1085 struct gfs2_rgrpd *rgd, *begin = NULL; 1105 struct gfs2_rgrpd *rgd, *begin = NULL;
@@ -1089,7 +1109,6 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked,
1089 int loops = 0; 1109 int loops = 0;
1090 int error, rg_locked; 1110 int error, rg_locked;
1091 1111
1092 *unlinked = 0;
1093 rgd = gfs2_blk2rgrpd(sdp, ip->i_goal); 1112 rgd = gfs2_blk2rgrpd(sdp, ip->i_goal);
1094 1113
1095 while (rgd) { 1114 while (rgd) {
@@ -1106,17 +1125,10 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked,
1106 case 0: 1125 case 0:
1107 if (try_rgrp_fit(rgd, al)) 1126 if (try_rgrp_fit(rgd, al))
1108 goto out; 1127 goto out;
1109 /* If the rg came in already locked, there's no 1128 if (rgd->rd_flags & GFS2_RDF_CHECK)
1110 way we can recover from a failed try_rgrp_unlink 1129 try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr);
1111 because that would require an iput which can only
1112 happen after the rgrp is unlocked. */
1113 if (!rg_locked && rgd->rd_flags & GFS2_RDF_CHECK)
1114 *unlinked = try_rgrp_unlink(rgd, last_unlinked,
1115 ip->i_no_addr);
1116 if (!rg_locked) 1130 if (!rg_locked)
1117 gfs2_glock_dq_uninit(&al->al_rgd_gh); 1131 gfs2_glock_dq_uninit(&al->al_rgd_gh);
1118 if (*unlinked)
1119 return -EAGAIN;
1120 /* fall through */ 1132 /* fall through */
1121 case GLR_TRYFAILED: 1133 case GLR_TRYFAILED:
1122 rgd = recent_rgrp_next(rgd); 1134 rgd = recent_rgrp_next(rgd);
@@ -1145,13 +1157,10 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked,
1145 case 0: 1157 case 0:
1146 if (try_rgrp_fit(rgd, al)) 1158 if (try_rgrp_fit(rgd, al))
1147 goto out; 1159 goto out;
1148 if (!rg_locked && rgd->rd_flags & GFS2_RDF_CHECK) 1160 if (rgd->rd_flags & GFS2_RDF_CHECK)
1149 *unlinked = try_rgrp_unlink(rgd, last_unlinked, 1161 try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr);
1150 ip->i_no_addr);
1151 if (!rg_locked) 1162 if (!rg_locked)
1152 gfs2_glock_dq_uninit(&al->al_rgd_gh); 1163 gfs2_glock_dq_uninit(&al->al_rgd_gh);
1153 if (*unlinked)
1154 return -EAGAIN;
1155 break; 1164 break;
1156 1165
1157 case GLR_TRYFAILED: 1166 case GLR_TRYFAILED:
@@ -1204,12 +1213,12 @@ int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex,
1204 struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); 1213 struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1205 struct gfs2_alloc *al = ip->i_alloc; 1214 struct gfs2_alloc *al = ip->i_alloc;
1206 int error = 0; 1215 int error = 0;
1207 u64 last_unlinked = NO_BLOCK, unlinked; 1216 u64 last_unlinked = NO_BLOCK;
1217 int tries = 0;
1208 1218
1209 if (gfs2_assert_warn(sdp, al->al_requested)) 1219 if (gfs2_assert_warn(sdp, al->al_requested))
1210 return -EINVAL; 1220 return -EINVAL;
1211 1221
1212try_again:
1213 if (hold_rindex) { 1222 if (hold_rindex) {
1214 /* We need to hold the rindex unless the inode we're using is 1223 /* We need to hold the rindex unless the inode we're using is
1215 the rindex itself, in which case it's already held. */ 1224 the rindex itself, in which case it's already held. */
@@ -1218,31 +1227,23 @@ try_again:
1218 else if (!sdp->sd_rgrps) /* We may not have the rindex read 1227 else if (!sdp->sd_rgrps) /* We may not have the rindex read
1219 in, so: */ 1228 in, so: */
1220 error = gfs2_ri_update_special(ip); 1229 error = gfs2_ri_update_special(ip);
1230 if (error)
1231 return error;
1221 } 1232 }
1222 1233
1223 if (error) 1234 do {
1224 return error; 1235 error = get_local_rgrp(ip, &last_unlinked);
1236 /* If there is no space, flushing the log may release some */
1237 if (error)
1238 gfs2_log_flush(sdp, NULL);
1239 } while (error && tries++ < 3);
1225 1240
1226 /* Find an rgrp suitable for allocation. If it encounters any unlinked
1227 dinodes along the way, error will equal -EAGAIN and unlinked will
1228 contains it block address. We then need to look up that inode and
1229 try to free it, and try the allocation again. */
1230 error = get_local_rgrp(ip, &unlinked, &last_unlinked);
1231 if (error) { 1241 if (error) {
1232 if (hold_rindex && ip != GFS2_I(sdp->sd_rindex)) 1242 if (hold_rindex && ip != GFS2_I(sdp->sd_rindex))
1233 gfs2_glock_dq_uninit(&al->al_ri_gh); 1243 gfs2_glock_dq_uninit(&al->al_ri_gh);
1234 if (error != -EAGAIN) 1244 return error;
1235 return error;
1236
1237 gfs2_process_unlinked_inode(ip->i_inode.i_sb, unlinked);
1238 /* regardless of whether or not gfs2_process_unlinked_inode
1239 was successful, we don't want to repeat it again. */
1240 last_unlinked = unlinked;
1241 gfs2_log_flush(sdp, NULL);
1242 error = 0;
1243
1244 goto try_again;
1245 } 1245 }
1246
1246 /* no error, so we have the rgrp set in the inode's allocation. */ 1247 /* no error, so we have the rgrp set in the inode's allocation. */
1247 al->al_file = file; 1248 al->al_file = file;
1248 al->al_line = line; 1249 al->al_line = line;