xref: /openbmc/linux/fs/jffs2/build.c (revision 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2)
1 /*
2  * JFFS2 -- Journalling Flash File System, Version 2.
3  *
4  * Copyright (C) 2001-2003 Red Hat, Inc.
5  *
6  * Created by David Woodhouse <dwmw2@infradead.org>
7  *
8  * For licensing information, see the file 'LICENCE' in this directory.
9  *
10  * $Id: build.c,v 1.69 2004/12/16 20:22:18 dmarlin Exp $
11  *
12  */
13 
14 #include <linux/kernel.h>
15 #include <linux/sched.h>
16 #include <linux/slab.h>
17 #include <linux/vmalloc.h>
18 #include <linux/mtd/mtd.h>
19 #include "nodelist.h"
20 
21 static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *, struct jffs2_inode_cache *, struct jffs2_full_dirent **);
22 
23 static inline struct jffs2_inode_cache *
24 first_inode_chain(int *i, struct jffs2_sb_info *c)
25 {
26 	for (; *i < INOCACHE_HASHSIZE; (*i)++) {
27 		if (c->inocache_list[*i])
28 			return c->inocache_list[*i];
29 	}
30 	return NULL;
31 }
32 
33 static inline struct jffs2_inode_cache *
34 next_inode(int *i, struct jffs2_inode_cache *ic, struct jffs2_sb_info *c)
35 {
36 	/* More in this chain? */
37 	if (ic->next)
38 		return ic->next;
39 	(*i)++;
40 	return first_inode_chain(i, c);
41 }
42 
43 #define for_each_inode(i, c, ic)			\
44 	for (i = 0, ic = first_inode_chain(&i, (c));	\
45 	     ic;					\
46 	     ic = next_inode(&i, ic, (c)))
47 
48 
49 static inline void jffs2_build_inode_pass1(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic)
50 {
51 	struct jffs2_full_dirent *fd;
52 
53 	D1(printk(KERN_DEBUG "jffs2_build_inode building directory inode #%u\n", ic->ino));
54 
55 	/* For each child, increase nlink */
56 	for(fd = ic->scan_dents; fd; fd = fd->next) {
57 		struct jffs2_inode_cache *child_ic;
58 		if (!fd->ino)
59 			continue;
60 
61 		/* XXX: Can get high latency here with huge directories */
62 
63 		child_ic = jffs2_get_ino_cache(c, fd->ino);
64 		if (!child_ic) {
65 			printk(KERN_NOTICE "Eep. Child \"%s\" (ino #%u) of dir ino #%u doesn't exist!\n",
66 				  fd->name, fd->ino, ic->ino);
67 			jffs2_mark_node_obsolete(c, fd->raw);
68 			continue;
69 		}
70 
71 		if (child_ic->nlink++ && fd->type == DT_DIR) {
72 			printk(KERN_NOTICE "Child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n", fd->name, fd->ino, ic->ino);
73 			if (fd->ino == 1 && ic->ino == 1) {
74 				printk(KERN_NOTICE "This is mostly harmless, and probably caused by creating a JFFS2 image\n");
75 				printk(KERN_NOTICE "using a buggy version of mkfs.jffs2. Use at least v1.17.\n");
76 			}
77 			/* What do we do about it? */
78 		}
79 		D1(printk(KERN_DEBUG "Increased nlink for child \"%s\" (ino #%u)\n", fd->name, fd->ino));
80 		/* Can't free them. We might need them in pass 2 */
81 	}
82 }
83 
84 /* Scan plan:
85  - Scan physical nodes. Build map of inodes/dirents. Allocate inocaches as we go
86  - Scan directory tree from top down, setting nlink in inocaches
87  - Scan inocaches for inodes with nlink==0
88 */
89 static int jffs2_build_filesystem(struct jffs2_sb_info *c)
90 {
91 	int ret;
92 	int i;
93 	struct jffs2_inode_cache *ic;
94 	struct jffs2_full_dirent *fd;
95 	struct jffs2_full_dirent *dead_fds = NULL;
96 
97 	/* First, scan the medium and build all the inode caches with
98 	   lists of physical nodes */
99 
100 	c->flags |= JFFS2_SB_FLAG_MOUNTING;
101 	ret = jffs2_scan_medium(c);
102 	if (ret)
103 		goto exit;
104 
105 	D1(printk(KERN_DEBUG "Scanned flash completely\n"));
106 	D2(jffs2_dump_block_lists(c));
107 
108 	/* Now scan the directory tree, increasing nlink according to every dirent found. */
109 	for_each_inode(i, c, ic) {
110 		D1(printk(KERN_DEBUG "Pass 1: ino #%u\n", ic->ino));
111 
112 		D1(BUG_ON(ic->ino > c->highest_ino));
113 
114 		if (ic->scan_dents) {
115 			jffs2_build_inode_pass1(c, ic);
116 			cond_resched();
117 		}
118 	}
119 	c->flags &= ~JFFS2_SB_FLAG_MOUNTING;
120 
121 	D1(printk(KERN_DEBUG "Pass 1 complete\n"));
122 
123 	/* Next, scan for inodes with nlink == 0 and remove them. If
124 	   they were directories, then decrement the nlink of their
125 	   children too, and repeat the scan. As that's going to be
126 	   a fairly uncommon occurrence, it's not so evil to do it this
127 	   way. Recursion bad. */
128 	D1(printk(KERN_DEBUG "Pass 2 starting\n"));
129 
130 	for_each_inode(i, c, ic) {
131 		D1(printk(KERN_DEBUG "Pass 2: ino #%u, nlink %d, ic %p, nodes %p\n", ic->ino, ic->nlink, ic, ic->nodes));
132 		if (ic->nlink)
133 			continue;
134 
135 		jffs2_build_remove_unlinked_inode(c, ic, &dead_fds);
136 		cond_resched();
137 	}
138 
139 	D1(printk(KERN_DEBUG "Pass 2a starting\n"));
140 
141 	while (dead_fds) {
142 		fd = dead_fds;
143 		dead_fds = fd->next;
144 
145 		ic = jffs2_get_ino_cache(c, fd->ino);
146 		D1(printk(KERN_DEBUG "Removing dead_fd ino #%u (\"%s\"), ic at %p\n", fd->ino, fd->name, ic));
147 
148 		if (ic)
149 			jffs2_build_remove_unlinked_inode(c, ic, &dead_fds);
150 		jffs2_free_full_dirent(fd);
151 	}
152 
153 	D1(printk(KERN_DEBUG "Pass 2 complete\n"));
154 
155 	/* Finally, we can scan again and free the dirent structs */
156 	for_each_inode(i, c, ic) {
157 		D1(printk(KERN_DEBUG "Pass 3: ino #%u, ic %p, nodes %p\n", ic->ino, ic, ic->nodes));
158 
159 		while(ic->scan_dents) {
160 			fd = ic->scan_dents;
161 			ic->scan_dents = fd->next;
162 			jffs2_free_full_dirent(fd);
163 		}
164 		ic->scan_dents = NULL;
165 		cond_resched();
166 	}
167 	D1(printk(KERN_DEBUG "Pass 3 complete\n"));
168 	D2(jffs2_dump_block_lists(c));
169 
170 	/* Rotate the lists by some number to ensure wear levelling */
171 	jffs2_rotate_lists(c);
172 
173 	ret = 0;
174 
175 exit:
176 	if (ret) {
177 		for_each_inode(i, c, ic) {
178 			while(ic->scan_dents) {
179 				fd = ic->scan_dents;
180 				ic->scan_dents = fd->next;
181 				jffs2_free_full_dirent(fd);
182 			}
183 		}
184 	}
185 
186 	return ret;
187 }
188 
189 static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, struct jffs2_full_dirent **dead_fds)
190 {
191 	struct jffs2_raw_node_ref *raw;
192 	struct jffs2_full_dirent *fd;
193 
194 	D1(printk(KERN_DEBUG "JFFS2: Removing ino #%u with nlink == zero.\n", ic->ino));
195 
196 	raw = ic->nodes;
197 	while (raw != (void *)ic) {
198 		struct jffs2_raw_node_ref *next = raw->next_in_ino;
199 		D1(printk(KERN_DEBUG "obsoleting node at 0x%08x\n", ref_offset(raw)));
200 		jffs2_mark_node_obsolete(c, raw);
201 		raw = next;
202 	}
203 
204 	if (ic->scan_dents) {
205 		int whinged = 0;
206 		D1(printk(KERN_DEBUG "Inode #%u was a directory which may have children...\n", ic->ino));
207 
208 		while(ic->scan_dents) {
209 			struct jffs2_inode_cache *child_ic;
210 
211 			fd = ic->scan_dents;
212 			ic->scan_dents = fd->next;
213 
214 			if (!fd->ino) {
215 				/* It's a deletion dirent. Ignore it */
216 				D1(printk(KERN_DEBUG "Child \"%s\" is a deletion dirent, skipping...\n", fd->name));
217 				jffs2_free_full_dirent(fd);
218 				continue;
219 			}
220 			if (!whinged) {
221 				whinged = 1;
222 				printk(KERN_NOTICE "Inode #%u was a directory with children - removing those too...\n", ic->ino);
223 			}
224 
225 			D1(printk(KERN_DEBUG "Removing child \"%s\", ino #%u\n",
226 				  fd->name, fd->ino));
227 
228 			child_ic = jffs2_get_ino_cache(c, fd->ino);
229 			if (!child_ic) {
230 				printk(KERN_NOTICE "Cannot remove child \"%s\", ino #%u, because it doesn't exist\n", fd->name, fd->ino);
231 				jffs2_free_full_dirent(fd);
232 				continue;
233 			}
234 
235 			/* Reduce nlink of the child. If it's now zero, stick it on the
236 			   dead_fds list to be cleaned up later. Else just free the fd */
237 
238 			child_ic->nlink--;
239 
240 			if (!child_ic->nlink) {
241 				D1(printk(KERN_DEBUG "Inode #%u (\"%s\") has now got zero nlink. Adding to dead_fds list.\n",
242 					  fd->ino, fd->name));
243 				fd->next = *dead_fds;
244 				*dead_fds = fd;
245 			} else {
246 				D1(printk(KERN_DEBUG "Inode #%u (\"%s\") has now got nlink %d. Ignoring.\n",
247 					  fd->ino, fd->name, child_ic->nlink));
248 				jffs2_free_full_dirent(fd);
249 			}
250 		}
251 	}
252 
253 	/*
254 	   We don't delete the inocache from the hash list and free it yet.
255 	   The erase code will do that, when all the nodes are completely gone.
256 	*/
257 }
258 
259 static void jffs2_calc_trigger_levels(struct jffs2_sb_info *c)
260 {
261 	uint32_t size;
262 
263 	/* Deletion should almost _always_ be allowed. We're fairly
264 	   buggered once we stop allowing people to delete stuff
265 	   because there's not enough free space... */
266 	c->resv_blocks_deletion = 2;
267 
268 	/* Be conservative about how much space we need before we allow writes.
269 	   On top of that which is required for deletia, require an extra 2%
270 	   of the medium to be available, for overhead caused by nodes being
271 	   split across blocks, etc. */
272 
273 	size = c->flash_size / 50; /* 2% of flash size */
274 	size += c->nr_blocks * 100; /* And 100 bytes per eraseblock */
275 	size += c->sector_size - 1; /* ... and round up */
276 
277 	c->resv_blocks_write = c->resv_blocks_deletion + (size / c->sector_size);
278 
279 	/* When do we let the GC thread run in the background */
280 
281 	c->resv_blocks_gctrigger = c->resv_blocks_write + 1;
282 
283 	/* When do we allow garbage collection to merge nodes to make
284 	   long-term progress at the expense of short-term space exhaustion? */
285 	c->resv_blocks_gcmerge = c->resv_blocks_deletion + 1;
286 
287 	/* When do we allow garbage collection to eat from bad blocks rather
288 	   than actually making progress? */
289 	c->resv_blocks_gcbad = 0;//c->resv_blocks_deletion + 2;
290 
291 	/* If there's less than this amount of dirty space, don't bother
292 	   trying to GC to make more space. It'll be a fruitless task */
293 	c->nospc_dirty_size = c->sector_size + (c->flash_size / 100);
294 
295 	D1(printk(KERN_DEBUG "JFFS2 trigger levels (size %d KiB, block size %d KiB, %d blocks)\n",
296 		  c->flash_size / 1024, c->sector_size / 1024, c->nr_blocks));
297 	D1(printk(KERN_DEBUG "Blocks required to allow deletion:    %d (%d KiB)\n",
298 		  c->resv_blocks_deletion, c->resv_blocks_deletion*c->sector_size/1024));
299 	D1(printk(KERN_DEBUG "Blocks required to allow writes:      %d (%d KiB)\n",
300 		  c->resv_blocks_write, c->resv_blocks_write*c->sector_size/1024));
301 	D1(printk(KERN_DEBUG "Blocks required to quiesce GC thread: %d (%d KiB)\n",
302 		  c->resv_blocks_gctrigger, c->resv_blocks_gctrigger*c->sector_size/1024));
303 	D1(printk(KERN_DEBUG "Blocks required to allow GC merges:   %d (%d KiB)\n",
304 		  c->resv_blocks_gcmerge, c->resv_blocks_gcmerge*c->sector_size/1024));
305 	D1(printk(KERN_DEBUG "Blocks required to GC bad blocks:     %d (%d KiB)\n",
306 		  c->resv_blocks_gcbad, c->resv_blocks_gcbad*c->sector_size/1024));
307 	D1(printk(KERN_DEBUG "Amount of dirty space required to GC: %d bytes\n",
308 		  c->nospc_dirty_size));
309 }
310 
311 int jffs2_do_mount_fs(struct jffs2_sb_info *c)
312 {
313 	int i;
314 
315 	c->free_size = c->flash_size;
316 	c->nr_blocks = c->flash_size / c->sector_size;
317  	if (c->mtd->flags & MTD_NO_VIRTBLOCKS)
318 		c->blocks = vmalloc(sizeof(struct jffs2_eraseblock) * c->nr_blocks);
319 	else
320 		c->blocks = kmalloc(sizeof(struct jffs2_eraseblock) * c->nr_blocks, GFP_KERNEL);
321 	if (!c->blocks)
322 		return -ENOMEM;
323 	for (i=0; i<c->nr_blocks; i++) {
324 		INIT_LIST_HEAD(&c->blocks[i].list);
325 		c->blocks[i].offset = i * c->sector_size;
326 		c->blocks[i].free_size = c->sector_size;
327 		c->blocks[i].dirty_size = 0;
328 		c->blocks[i].wasted_size = 0;
329 		c->blocks[i].unchecked_size = 0;
330 		c->blocks[i].used_size = 0;
331 		c->blocks[i].first_node = NULL;
332 		c->blocks[i].last_node = NULL;
333 		c->blocks[i].bad_count = 0;
334 	}
335 
336 	init_MUTEX(&c->alloc_sem);
337 	init_MUTEX(&c->erase_free_sem);
338 	init_waitqueue_head(&c->erase_wait);
339 	init_waitqueue_head(&c->inocache_wq);
340 	spin_lock_init(&c->erase_completion_lock);
341 	spin_lock_init(&c->inocache_lock);
342 
343 	INIT_LIST_HEAD(&c->clean_list);
344 	INIT_LIST_HEAD(&c->very_dirty_list);
345 	INIT_LIST_HEAD(&c->dirty_list);
346 	INIT_LIST_HEAD(&c->erasable_list);
347 	INIT_LIST_HEAD(&c->erasing_list);
348 	INIT_LIST_HEAD(&c->erase_pending_list);
349 	INIT_LIST_HEAD(&c->erasable_pending_wbuf_list);
350 	INIT_LIST_HEAD(&c->erase_complete_list);
351 	INIT_LIST_HEAD(&c->free_list);
352 	INIT_LIST_HEAD(&c->bad_list);
353 	INIT_LIST_HEAD(&c->bad_used_list);
354 	c->highest_ino = 1;
355 
356 	if (jffs2_build_filesystem(c)) {
357 		D1(printk(KERN_DEBUG "build_fs failed\n"));
358 		jffs2_free_ino_caches(c);
359 		jffs2_free_raw_node_refs(c);
360 		if (c->mtd->flags & MTD_NO_VIRTBLOCKS) {
361 			vfree(c->blocks);
362 		} else {
363 			kfree(c->blocks);
364 		}
365 		return -EIO;
366 	}
367 
368 	jffs2_calc_trigger_levels(c);
369 
370 	return 0;
371 }
372