pramfs: extended attributes management

diff -Nurp linux-2.6.36-orig/fs/pramfs/xattr.c linux-2.6.36/fs/pramfs/xattr.c
@@ -0,0 +1,1104 @@
+/*
+ * BRIEF DESCRIPTION
+ *
+ * Extended attributes operations.
+ *
+ * Copyright 2010 Marco Stornelli <[email protected]>
+ *
+ * based on fs/ext2/xattr.c with the following copyright:
+ *
+ * Fix by Harrison Xing <[email protected]>.
+ * Extended attributes for symlinks and special files added per
+ *  suggestion of Luka Renko <[email protected]>.
+ * xattr consolidation Copyright (c) 2004 James Morris <[email protected]>,
+ *  Red Hat Inc.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+/*
+ * Extended attributes are stored in blocks allocated outside of
+ * any inode. The i_xattr field is then made to point to this allocated
+ * block. If all extended attributes of an inode are identical, these
+ * inodes may share the same extended attribute block. Such situations
+ * are automatically detected by keeping a cache of recent attribute block
+ * numbers and hashes over the block's contents in memory.
+ *
+ *
+ * Extended attribute block layout:
+ *
+ *   +------------------+
+ *   | header           |
+ *   | entry 1          | |
+ *   | entry 2          | | growing downwards
+ *   | entry 3          | v
+ *   | four null bytes  |
+ *   | . . .            |
+ *   | value 1          | ^
+ *   | value 3          | | growing upwards
+ *   | value 2          | |
+ *   +------------------+
+ *
+ * The block header is followed by multiple entry descriptors. These entry
+ * descriptors are variable in size, and alligned to PRAM_XATTR_PAD
+ * byte boundaries. The entry descriptors are sorted by attribute name,
+ * so that two extended attribute blocks can be compared efficiently.
+ *
+ * Attribute values are aligned to the end of the block, stored in
+ * no specific order. They are also padded to PRAM_XATTR_PAD byte
+ * boundaries. No additional gaps are left between them.
+ *
+ * Locking strategy
+ * ----------------
+ * pi->i_xattr is protected by PRAM_I(inode)->xattr_sem.
+ * EA blocks are only changed if they are exclusive to an inode, so
+ * holding xattr_sem also means that nothing but the EA block's reference
+ * count will change. Multiple writers to an EA block are synchronized
+ * by the mutex in each block descriptor. Block descriptors are kept in a
+ * red black tree and the key is the absolute block number.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/mbcache.h>
+#include <linux/rwsem.h>
+#include <linux/security.h>
+#include "pram.h"
+#include "xattr.h"
+#include "acl.h"
+#include "desctree.h"
+
+#define HDR(bp) ((struct pram_xattr_header *)(bp))
+#define ENTRY(ptr) ((struct pram_xattr_entry *)(ptr))
+#define FIRST_ENTRY(bh) ENTRY(HDR(bh)+1)
+#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0)
+#define GET_DESC(sbi, blocknr) lookup_xblock_desc(sbi, blocknr, pram_xblock_desc_cache, 1)
+#define LOOKUP_DESC(sbi, blocknr) lookup_xblock_desc(sbi, blocknr, NULL, 0)
+
+#ifdef PRAM_XATTR_DEBUG
+# define ea_idebug(inode, f...) do { \
+		printk(KERN_DEBUG "inode %ld: ", inode->i_ino); \
+		printk(f); \
+		printk("\n"); \
+	} while (0)
+# define ea_bdebug(blocknr, f...) do { \
+		printk(KERN_DEBUG "block %lu: ", blocknr); \
+		printk(f); \
+		printk("\n"); \
+	} while (0)
+#else
+# define ea_idebug(f...)
+# define ea_bdebug(f...)
+#endif
+
+static int pram_xattr_set2(struct inode *, char *, struct pram_xblock_desc *, struct pram_xattr_header *);
+
+static int pram_xattr_cache_insert(struct super_block *sb, unsigned long blocknr, u32 xhash);
+static struct pram_xblock_desc *pram_xattr_cache_find(struct inode *,
+						 struct pram_xattr_header *);
+static void pram_xattr_rehash(struct pram_xattr_header *,
+			      struct pram_xattr_entry *);
+
+static struct mb_cache *pram_xattr_cache;
+static struct kmem_cache *pram_xblock_desc_cache;
+
+static const struct xattr_handler *pram_xattr_handler_map[] = {
+	[PRAM_XATTR_INDEX_USER]		     = &pram_xattr_user_handler,
+#ifdef CONFIG_PRAMFS_POSIX_ACL
+	[PRAM_XATTR_INDEX_POSIX_ACL_ACCESS]  = &pram_xattr_acl_access_handler,
+	[PRAM_XATTR_INDEX_POSIX_ACL_DEFAULT] = &pram_xattr_acl_default_handler,
+#endif
+	[PRAM_XATTR_INDEX_TRUSTED]	     = &pram_xattr_trusted_handler,
+#ifdef CONFIG_PRAMFS_SECURITY
+	[PRAM_XATTR_INDEX_SECURITY]	     = &pram_xattr_security_handler,
+#endif
+};
+
+const struct xattr_handler *pram_xattr_handlers[] = {
+	&pram_xattr_user_handler,
+	&pram_xattr_trusted_handler,
+#ifdef CONFIG_PRAMFS_POSIX_ACL
+	&pram_xattr_acl_access_handler,
+	&pram_xattr_acl_default_handler,
+#endif
+#ifdef CONFIG_PRAMFS_SECURITY
+	&pram_xattr_security_handler,
+#endif
+	NULL
+};
+
+static void desc_put(struct super_block *sb, struct pram_xblock_desc *desc)
+{
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	if (!put_xblock_desc(sbi, desc)) {
+		/* Ok we can free the block and its descriptor */
+		pram_dbg("freeing block %lu and its descriptor", desc->blocknr);
+		pram_free_block(sb, desc->blocknr);
+		kmem_cache_free(pram_xblock_desc_cache, desc);
+	}
+}
+
+static inline const struct xattr_handler *pram_xattr_handler(int name_index)
+{
+	const struct xattr_handler *handler = NULL;
+
+	if (name_index > 0 && name_index < ARRAY_SIZE(pram_xattr_handler_map))
+		handler = pram_xattr_handler_map[name_index];
+	return handler;
+}
+
+/*
+ * pram_xattr_get()
+ *
+ * Copy an extended attribute into the buffer
+ * provided, or compute the buffer size required.
+ * Buffer is NULL to compute the size of the buffer required.
+ *
+ * Returns a negative error number on failure, or the number of bytes
+ * used / required on success.
+ */
+int pram_xattr_get(struct inode *inode, int name_index, const char *name,
+	       void *buffer, size_t buffer_size)
+{
+	char *bp = NULL;
+	struct pram_xattr_entry *entry;
+	struct pram_xblock_desc *desc;
+	struct pram_inode *pi;
+	size_t name_len, size;
+	char *end;
+	int error = 0;
+	unsigned long blocknr;
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+
+	ea_idebug(inode, "name=%d<%s>, buffer=%p, buffer_size=%ld",
+		  name_index, name, buffer, (long)buffer_size);
+
+	pi = pram_get_inode(sb, inode->i_ino);
+	if (!pi)
+		return -EINVAL;
+	if (name == NULL)
+		return -EINVAL;
+	down_read(&PRAM_I(inode)->xattr_sem);
+	error = -ENODATA;
+	if (!pi->i_xattr)
+		goto cleanup;
+	ea_idebug(inode, "reading block %llu", be64_to_cpu(pi->i_xattr));
+	bp = pram_get_block(sb, be64_to_cpu(pi->i_xattr));
+	error = -EIO;
+	if (!bp)
+		goto cleanup;
+	end = bp + sb->s_blocksize;
+	blocknr = pram_get_blocknr(sb, be64_to_cpu(pi->i_xattr));
+	ea_bdebug(blocknr, "refcount=%d", be32_to_cpu(HDR(bp)->h_refcount));
+	if (HDR(bp)->h_magic != cpu_to_be32(PRAM_XATTR_MAGIC)) {
+bad_block:	pram_err(sb, "inode %ld: bad block %llu", inode->i_ino,
+		be64_to_cpu(pi->i_xattr));
+		error = -EIO;
+		goto cleanup;
+	}
+	/* find named attribute */
+	name_len = strlen(name);
+
+	error = -ERANGE;
+	if (name_len > 255)
+		goto cleanup;
+	entry = FIRST_ENTRY(bp);
+	while (!IS_LAST_ENTRY(entry)) {
+		struct pram_xattr_entry *next =
+			PRAM_XATTR_NEXT(entry);
+		if ((char *)next >= end)
+			goto bad_block;
+		if (name_index == entry->e_name_index &&
+		    name_len == entry->e_name_len &&
+		    memcmp(name, entry->e_name, name_len) == 0)
+			goto found;
+		entry = next;
+	}
+	/* Check the remaining name entries */
+	while (!IS_LAST_ENTRY(entry)) {
+		struct pram_xattr_entry *next =
+			PRAM_XATTR_NEXT(entry);
+		if ((char *)next >= end)
+			goto bad_block;
+		entry = next;
+	}
+
+	desc = GET_DESC(sbi, blocknr);
+	if (IS_ERR(desc)) {
+		error = -ENOMEM;
+		goto cleanup;
+	}
+	desc_put(sb, desc);
+	if (pram_xattr_cache_insert(sb, blocknr,
+					be32_to_cpu(HDR(bp)->h_hash)))
+		ea_idebug(inode, "cache insert failed");
+	error = -ENODATA;
+	goto cleanup;
+found:
+	/* check the buffer size */
+	if (entry->e_value_block != 0)
+		goto bad_block;
+	size = be32_to_cpu(entry->e_value_size);
+	if (size > inode->i_sb->s_blocksize ||
+	    be16_to_cpu(entry->e_value_offs) + size > inode->i_sb->s_blocksize)
+		goto bad_block;
+
+	desc = GET_DESC(sbi, blocknr);
+	if (IS_ERR(desc)) {
+		error = -ENOMEM;
+		goto cleanup;
+	}
+	desc_put(sb, desc);
+	if (pram_xattr_cache_insert(sb, blocknr,
+					be32_to_cpu(HDR(bp)->h_hash)))
+		ea_idebug(inode, "cache insert failed");
+	if (buffer) {
+		error = -ERANGE;
+		if (size > buffer_size)
+			goto cleanup;
+		/* return value of attribute */
+		memcpy(buffer, bp + be16_to_cpu(entry->e_value_offs),
+			size);
+	}
+	error = size;
+
+cleanup:
+	up_read(&PRAM_I(inode)->xattr_sem);
+
+	return error;
+}
+
+/*
+ * pram_xattr_list()
+ *
+ * Copy a list of attribute names into the buffer
+ * provided, or compute the buffer size required.
+ * Buffer is NULL to compute the size of the buffer required.
+ *
+ * Returns a negative error number on failure, or the number of bytes
+ * used / required on success.
+ */
+static int pram_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
+{
+	struct inode *inode = dentry->d_inode;
+	char *bp = NULL;
+	struct pram_xattr_entry *entry;
+	struct pram_xblock_desc *desc;
+	struct pram_inode *pi;
+	char *end;
+	size_t rest = buffer_size;
+	int error = 0;
+	unsigned long blocknr;
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+
+	ea_idebug(inode, "buffer=%p, buffer_size=%ld",
+		  buffer, (long)buffer_size);
+
+	pi = pram_get_inode(sb, inode->i_ino);
+	if (!pi)
+		return error;
+	down_read(&PRAM_I(inode)->xattr_sem);
+	error = 0;
+	if (!pi->i_xattr)
+		goto cleanup;
+	ea_idebug(inode, "reading block %llu", be64_to_cpu(pi->i_xattr));
+	bp = pram_get_block(sb, be64_to_cpu(pi->i_xattr));
+	blocknr = pram_get_blocknr(sb, be64_to_cpu(pi->i_xattr));
+	error = -EIO;
+	if (!bp)
+		goto cleanup;
+	ea_bdebug(blocknr, "refcount=%d", be32_to_cpu(HDR(bp)->h_refcount));
+	end = bp + sb->s_blocksize;
+	if (HDR(bp)->h_magic != cpu_to_be32(PRAM_XATTR_MAGIC)) {
+bad_block:	pram_err(sb, "inode %ld: bad block %llu", inode->i_ino,
+			be64_to_cpu(pi->i_xattr));
+		error = -EIO;
+		goto cleanup;
+	}
+
+	/* check the on-disk data structure */
+	entry = FIRST_ENTRY(bp);
+	while (!IS_LAST_ENTRY(entry)) {
+		struct pram_xattr_entry *next = PRAM_XATTR_NEXT(entry);
+
+		if ((char *)next >= end)
+			goto bad_block;
+		entry = next;
+	}
+
+	desc = GET_DESC(sbi, blocknr);
+	if (IS_ERR(desc)) {
+		error = -ENOMEM;
+		goto cleanup;
+	}
+	desc_put(sb, desc);
+	if (pram_xattr_cache_insert(sb, blocknr,
+					be32_to_cpu(HDR(bp)->h_hash)))
+			ea_idebug(inode, "cache insert failed");
+
+	/* list the attribute names */
+	for (entry = FIRST_ENTRY(bp); !IS_LAST_ENTRY(entry);
+	     entry = PRAM_XATTR_NEXT(entry)) {
+		const struct xattr_handler *handler =
+			pram_xattr_handler(entry->e_name_index);
+
+		if (handler) {
+			size_t size = handler->list(dentry, buffer, rest,
+						    entry->e_name,
+						    entry->e_name_len,
+						    handler->flags);
+			if (buffer) {
+				if (size > rest) {
+					error = -ERANGE;
+					goto cleanup;
+				}
+				buffer += size;
+			}
+			rest -= size;
+		}
+	}
+	error = buffer_size - rest;  /* total size */
+
+cleanup:
+	up_read(&PRAM_I(inode)->xattr_sem);
+
+	return error;
+}
+
+/*
+ * Inode operation listxattr()
+ *
+ * dentry->d_inode->i_mutex: don't care
+ */
+ssize_t pram_listxattr(struct dentry *dentry, char *buffer, size_t size)
+{
+	return pram_xattr_list(dentry, buffer, size);
+}
+
+/*
+ * pram_xattr_set()
+ *
+ * Create, replace or remove an extended attribute for this inode. Buffer
+ * is NULL to remove an existing extended attribute, and non-NULL to
+ * either replace an existing extended attribute, or create a new extended
+ * attribute. The flags XATTR_REPLACE and XATTR_CREATE
+ * specify that an extended attribute must exist and must not exist
+ * previous to the call, respectively.
+ *
+ * Returns 0, or a negative error number on failure.
+ */
+int pram_xattr_set(struct inode *inode, int name_index, const char *name,
+	       const void *value, size_t value_len, int flags)
+{
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	struct pram_xattr_header *header = NULL;
+	struct pram_xattr_entry *here, *last;
+	struct pram_inode *pi;
+	struct pram_xblock_desc *desc = NULL;
+	size_t name_len, free, min_offs = sb->s_blocksize;
+	int not_found = 1, error;
+	char *end;
+	char *bp = NULL;
+	unsigned long blocknr = 0;
+
+	/*
+	 * header -- Points either into bp, or to a temporarily
+	 *           allocated buffer.
+	 * here -- The named entry found, or the place for inserting, within
+	 *         the block pointed to by header.
+	 * last -- Points right after the last named entry within the block
+	 *         pointed to by header.
+	 * min_offs -- The offset of the first value (values are aligned
+	 *             towards the end of the block).
+	 * end -- Points right after the block pointed to by header.
+	 */
+
+	ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld",
+		  name_index, name, value, (long)value_len);
+
+	if (value == NULL)
+		value_len = 0;
+	if (name == NULL)
+		return -EINVAL;
+	name_len = strlen(name);
+	if (name_len > 255 || value_len > sb->s_blocksize)
+		return -ERANGE;
+	pi = pram_get_inode(sb, inode->i_ino);
+	if (!pi)
+		return -EINVAL;
+	down_write(&PRAM_I(inode)->xattr_sem);
+	if (pi->i_xattr) {
+		/* The inode already has an extended attribute block. */
+		bp = pram_get_block(sb, be64_to_cpu(pi->i_xattr));
+		error = -EIO;
+		if (!bp)
+			goto cleanup;
+		blocknr = pram_get_blocknr(sb, be64_to_cpu(pi->i_xattr));
+		ea_bdebug(blocknr, "refcount=%d", be32_to_cpu(HDR(bp)->h_refcount));
+		header = HDR(bp);
+		end = bp + sb->s_blocksize;
+		if (header->h_magic != cpu_to_be32(PRAM_XATTR_MAGIC)) {
+bad_block:		pram_err(sb, "inode %ld: bad block %llu", inode->i_ino,
+				   be64_to_cpu(pi->i_xattr));
+			error = -EIO;
+			goto cleanup;
+		}
+		/* Find the named attribute. */
+		here = FIRST_ENTRY(bp);
+		while (!IS_LAST_ENTRY(here)) {
+			struct pram_xattr_entry *next = PRAM_XATTR_NEXT(here);
+			if ((char *)next >= end)
+				goto bad_block;
+			if (!here->e_value_block && here->e_value_size) {
+				size_t offs = be16_to_cpu(here->e_value_offs);
+				if (offs < min_offs)
+					min_offs = offs;
+			}
+			not_found = name_index - here->e_name_index;
+			if (!not_found)
+				not_found = name_len - here->e_name_len;
+			if (!not_found)
+				not_found = memcmp(name, here->e_name, name_len);
+			if (not_found <= 0)
+				break;
+			here = next;
+		}
+		last = here;
+		/* We still need to compute min_offs and last. */
+		while (!IS_LAST_ENTRY(last)) {
+			struct pram_xattr_entry *next = PRAM_XATTR_NEXT(last);
+			if ((char *)next >= end)
+				goto bad_block;
+			if (!last->e_value_block && last->e_value_size) {
+				size_t offs = be16_to_cpu(last->e_value_offs);
+				if (offs < min_offs)
+					min_offs = offs;
+			}
+			last = next;
+		}
+
+		/* Check whether we have enough space left. */
+		free = min_offs - ((char *)last - (char *)header) - sizeof(__u32);
+	} else {
+		/* We will use a new extended attribute block. */
+		free = sb->s_blocksize -
+			sizeof(struct pram_xattr_header) - sizeof(__u32);
+		here = last = NULL;  /* avoid gcc uninitialized warning. */
+	}
+
+	if (not_found) {
+		/* Request to remove a nonexistent attribute? */
+		error = -ENODATA;
+		if (flags & XATTR_REPLACE)
+			goto cleanup;
+		error = 0;
+		if (value == NULL)
+			goto cleanup;
+	} else {
+		/* Request to create an existing attribute? */
+		error = -EEXIST;
+		if (flags & XATTR_CREATE)
+			goto cleanup;
+		if (!here->e_value_block && here->e_value_size) {
+			size_t size = be32_to_cpu(here->e_value_size);
+
+			if (be16_to_cpu(here->e_value_offs) + size >
+			    sb->s_blocksize || size > sb->s_blocksize)
+				goto bad_block;
+			free += PRAM_XATTR_SIZE(size);
+		}
+		free += PRAM_XATTR_LEN(name_len);
+	}
+	error = -ENOSPC;
+	if (free < PRAM_XATTR_LEN(name_len) + PRAM_XATTR_SIZE(value_len))
+		goto cleanup;
+
+	/* Here we know that we can set the new attribute. */
+
+	if (header) {
+		struct mb_cache_entry *ce;
+
+		desc = GET_DESC(sbi, blocknr);
+		if (IS_ERR(desc)) {
+			error = -ENOMEM;
+			goto cleanup;
+		}
+
+		/* assert(header == HDR(bp)); */
+		ce = mb_cache_entry_get(pram_xattr_cache, (struct block_device *)sbi,
+							blocknr);
+		mutex_lock(&desc->lock);
+		pram_memunlock_block(sb, bp);
+		if (header->h_refcount == cpu_to_be32(1)) {
+			ea_bdebug(blocknr, "modifying in-place");
+			if (ce)
+				mb_cache_entry_free(ce);
+			/* keep it locked while modifying it. */
+		} else {
+			int offset;
+
+			if (ce)
+				mb_cache_entry_release(ce);
+			pram_memlock_block(sb, bp);
+			mutex_unlock(&desc->lock);
+			ea_bdebug(desc->blocknr, "cloning");
+			header = kmalloc(inode->i_sb->s_blocksize, GFP_KERNEL);
+			error = -ENOMEM;
+			if (header == NULL)
+				goto cleanup;
+			memcpy(header, HDR(bp), inode->i_sb->s_blocksize);
+			header->h_refcount = cpu_to_be32(1);
+
+			offset = (char *)here - bp;
+			here = ENTRY((char *)header + offset);
+			offset = (char *)last - bp;
+			last = ENTRY((char *)header + offset);
+		}
+	} else {
+		/* Allocate a buffer where we construct the new block. */
+		header = kzalloc(sb->s_blocksize, GFP_KERNEL);
+		error = -ENOMEM;
+		if (header == NULL)
+			goto cleanup;
+		end = (char *)header + sb->s_blocksize;
+		header->h_magic = cpu_to_be32(PRAM_XATTR_MAGIC);
+		header->h_refcount = cpu_to_be32(1);
+		last = here = ENTRY(header+1);
+	}
+
+	/* Iff we are modifying the block in-place, the block is locked here. */
+
+	if (not_found) {
+		/* Insert the new name. */
+		size_t size = PRAM_XATTR_LEN(name_len);
+		size_t rest = (char *)last - (char *)here;
+		memmove((char *)here + size, here, rest);
+		memset(here, 0, size);
+		here->e_name_index = name_index;
+		here->e_name_len = name_len;
+		memcpy(here->e_name, name, name_len);
+	} else {
+		if (!here->e_value_block && here->e_value_size) {
+			char *first_val = (char *)header + min_offs;
+			size_t offs = be16_to_cpu(here->e_value_offs);
+			char *val = (char *)header + offs;
+			size_t size = PRAM_XATTR_SIZE(
+				be32_to_cpu(here->e_value_size));
+
+			if (size == PRAM_XATTR_SIZE(value_len)) {
+				/* The old and the new value have the same
+				   size. Just replace. */
+				here->e_value_size = cpu_to_be32(value_len);
+				memset(val + size - PRAM_XATTR_PAD, 0,
+				       PRAM_XATTR_PAD); /* Clear pad bytes. */
+				memcpy(val, value, value_len);
+				goto skip_replace;
+			}
+
+			/* Remove the old value. */
+			memmove(first_val + size, first_val, val - first_val);
+			memset(first_val, 0, size);
+			here->e_value_offs = 0;
+			min_offs += size;
+
+			/* Adjust all value offsets. */
+			last = ENTRY(header+1);
+			while (!IS_LAST_ENTRY(last)) {
+				size_t o = be16_to_cpu(last->e_value_offs);
+				if (!last->e_value_block && o < offs)
+					last->e_value_offs =
+						cpu_to_be16(o + size);
+				last = PRAM_XATTR_NEXT(last);
+			}
+		}
+		if (value == NULL) {
+			/* Remove the old name. */
+			size_t size = PRAM_XATTR_LEN(name_len);
+			last = ENTRY((char *)last - size);
+			memmove(here, (char *)here + size,
+				(char *)last - (char *)here);
+			memset(last, 0, size);
+		}
+	}
+
+	if (value != NULL) {
+		/* Insert the new value. */
+		here->e_value_size = cpu_to_be32(value_len);
+		if (value_len) {
+			size_t size = PRAM_XATTR_SIZE(value_len);
+			char *val = (char *)header + min_offs - size;
+			here->e_value_offs =
+				cpu_to_be16((char *)val - (char *)header);
+			memset(val + size - PRAM_XATTR_PAD, 0,
+			       PRAM_XATTR_PAD); /* Clear the pad bytes. */
+			memcpy(val, value, value_len);
+		}
+	}
+
+skip_replace:
+	if (IS_LAST_ENTRY(ENTRY(header+1))) {
+		/* This block is now empty. */
+		if (bp && header == HDR(bp)) {
+			/* we were modifying in-place. */
+			pram_memlock_block(sb, bp);
+			mutex_unlock(&desc->lock);
+		}
+		error = pram_xattr_set2(inode, bp, desc, NULL);
+	} else {
+		pram_xattr_rehash(header, here);
+		if (bp && header == HDR(bp)) {
+			/* we were modifying in-place. */
+			pram_memlock_block(sb, bp);
+			mutex_unlock(&desc->lock);
+		}
+		error = pram_xattr_set2(inode, bp, desc, header);
+	}
+
+cleanup:
+	desc_put(sb, desc);
+	if (!(bp && header == HDR(bp)))
+		kfree(header);
+	up_write(&PRAM_I(inode)->xattr_sem);
+
+	return error;
+}
+
+/*
+ * Second half of pram_xattr_set(): Update the file system.
+ */
+static int pram_xattr_set2(struct inode *inode, char *old_bp,
+			   struct pram_xblock_desc *old_desc,
+			   struct pram_xattr_header *header)
+{
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	struct pram_xblock_desc *new_desc = NULL;
+	unsigned long blocknr;
+	struct pram_inode *pi;
+	int error;
+	char *new_bp = NULL;
+
+	if (header) {
+		new_desc = pram_xattr_cache_find(inode, header);
+		if (new_desc) {
+			new_bp = pram_get_block(sb, pram_get_block_off(sb, new_desc->blocknr));
+			/* We found an identical block in the cache. */
+			if (new_bp == old_bp) {
+				ea_bdebug(new_desc->blocknr, "keeping this block");
+			} else {
+				/* The old block is released after updating
+				   the inode.  */
+				ea_bdebug(new_desc->blocknr, "reusing block");
+				pram_memunlock_block(sb, new_bp);
+				be32_add_cpu(&HDR(new_bp)->h_refcount, 1);
+				pram_memlock_block(sb, new_bp);
+				ea_bdebug(new_desc->blocknr, "refcount now=%d",
+					be32_to_cpu(HDR(new_bp)->h_refcount));
+			}
+			blocknr = new_desc->blocknr;
+			mutex_unlock(&new_desc->lock);
+			desc_put(sb, new_desc);
+		} else if (old_bp && header == HDR(old_bp)) {
+			/* Keep this block. No need to lock the block as we
+			   don't need to change the reference count. */
+			new_bp = old_bp;
+			pram_xattr_cache_insert(sb, old_desc->blocknr, HDR(new_bp)->h_hash);
+			blocknr = old_desc->blocknr;
+		} else {
+			/* We need to allocate a new block */
+			struct pram_xblock_desc *new_desc;
+
+			error = pram_new_block(sb, &blocknr, 1);
+			if (error)
+				goto out;
+			ea_idebug(inode, "creating block %lu", blocknr);
+			new_desc = kmem_cache_alloc(pram_xblock_desc_cache, GFP_KERNEL);
+			if (!new_desc) {
+				pram_free_block(sb, blocknr);
+				error = -EIO;
+				goto out;
+			}
+			xblock_desc_init_always(new_desc);
+			new_desc->blocknr = blocknr;
+			new_bp = pram_get_block(sb, pram_get_block_off(sb, blocknr));
+			if (!new_bp) {
+				pram_free_block(sb, blocknr);
+				kmem_cache_free(pram_xblock_desc_cache, new_desc);
+				error = -EIO;
+				goto out;
+			}
+			pram_memunlock_block(sb, new_bp);
+			memcpy(new_bp, header, sb->s_blocksize);
+			pram_memlock_block(sb, new_bp);
+			insert_xblock_desc(sbi, new_desc);
+			pram_xattr_cache_insert(sb, new_desc->blocknr, HDR(new_bp)->h_hash);
+		}
+	}
+
+	/* Update the inode. */
+	pi = pram_get_inode(sb, inode->i_ino);
+	pram_memunlock_inode(sb, pi);
+	pi->i_xattr = new_bp ? be64_to_cpu(pram_get_block_off(sb, blocknr)) : 0;
+	inode->i_ctime = CURRENT_TIME_SEC;
+	pi->i_ctime = cpu_to_be32(inode->i_ctime.tv_sec);
+	pram_memlock_inode(sb, pi);
+
+	error = 0;
+	if (old_bp && old_bp != new_bp) {
+		struct mb_cache_entry *ce;
+
+		/* Here old_desc MUST be valid or we have a bug */
+		BUG_ON(!old_desc);
+
+		/*
+		 * If there was an old block and we are no longer using it,
+		 * release the old block.
+		 */
+		ce = mb_cache_entry_get(pram_xattr_cache, (struct block_device *)sbi,
+					old_desc->blocknr);
+		mutex_lock(&old_desc->lock);
+		if (HDR(old_bp)->h_refcount == cpu_to_be32(1)) {
+			/* Free the old block. */
+			if (ce)
+				mb_cache_entry_free(ce);
+			ea_bdebug(old_desc->blocknr, "freeing");
+			mutex_unlock(&old_desc->lock);
+			/* Caller will call desc_put later */
+			mark_free_desc(old_desc);
+		} else {
+			/* Decrement the refcount only. */
+			pram_memunlock_block(sb, old_bp);
+			be32_add_cpu(&HDR(old_bp)->h_refcount, -1);
+			pram_memlock_block(sb, old_bp);
+			if (ce)
+				mb_cache_entry_release(ce);
+			ea_bdebug(old_desc->blocknr, "refcount now=%d",
+			be32_to_cpu(HDR(old_bp)->h_refcount));
+			mutex_unlock(&old_desc->lock);
+		}
+	}
+
+out:
+	return error;
+}
+
+/*
+ * pram_xattr_delete_inode()
+ *
+ * Free extended attribute resources associated with this inode. This
+ * is called immediately before an inode is freed.
+ */
+void pram_xattr_delete_inode(struct inode *inode)
+{
+	char *bp = NULL;
+	struct mb_cache_entry *ce;
+	struct pram_inode *pi;
+	struct pram_xblock_desc *desc;
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	unsigned long blocknr;
+
+	pi = pram_get_inode(sb, inode->i_ino);
+	if (!pi)
+		goto cleanup;
+	down_write(&PRAM_I(inode)->xattr_sem);
+	if (!pi->i_xattr)
+		goto cleanup;
+	bp = pram_get_block(sb, be64_to_cpu(pi->i_xattr));
+	if (!bp) {
+		pram_err(sb, "inode %ld: block %llu read error", inode->i_ino,
+			be64_to_cpu(pi->i_xattr));
+		goto cleanup;
+	}
+	blocknr = pram_get_blocknr(sb, be64_to_cpu(pi->i_xattr));
+	if (HDR(bp)->h_magic != cpu_to_be32(PRAM_XATTR_MAGIC)) {
+		pram_err(sb, "inode %ld: bad block %llu", inode->i_ino,
+			be64_to_cpu(pi->i_xattr));
+		goto cleanup;
+	}
+	ce = mb_cache_entry_get(pram_xattr_cache, (struct block_device *)sbi, blocknr);
+	desc = GET_DESC(sbi, blocknr);
+	if (IS_ERR(desc))
+		goto cleanup;
+	mutex_lock(&desc->lock);
+	if (HDR(bp)->h_refcount == cpu_to_be32(1)) {
+		if (ce)
+			mb_cache_entry_free(ce);
+		mark_free_desc(desc);
+	} else {
+		be32_add_cpu(&HDR(bp)->h_refcount, -1);
+		if (ce)
+			mb_cache_entry_release(ce);
+		ea_bdebug(blocknr, "refcount now=%d",
+			be32_to_cpu(HDR(bp)->h_refcount));
+		mutex_unlock(&desc->lock);
+	}
+	desc_put(sb, desc);
+
+cleanup:
+	up_write(&PRAM_I(inode)->xattr_sem);
+}
+
+/*
+ * pram_xattr_put_super()
+ *
+ * This is called when a file system is unmounted.
+ */
+void pram_xattr_put_super(struct super_block *sb)
+{
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	/*
+	 * NOTE: we haven't got any block device to use with mb. Mb code
+	 * doesn't use the pointer but it uses only the address as unique
+	 * key so it's safe to use a "general purpose" address. We use
+	 * super block info data as unique key. Maybe it'd be better to
+	 * change mb code in order to use a generic void pointer to a
+	 * generic id.
+	 */
+	mb_cache_shrink((struct block_device *)sbi);
+	erase_tree(sbi, pram_xblock_desc_cache);
+	kmem_cache_shrink(pram_xblock_desc_cache);
+}
+
+
+/*
+ * pram_xattr_cache_insert()
+ *
+ * Create a new entry in the extended attribute cache, and insert
+ * it unless such an entry is already in the cache.
+ *
+ * Returns 0, or a negative error number on failure.
+ */
+static int pram_xattr_cache_insert(struct super_block *sb, unsigned long blocknr, u32 xhash)
+{
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+	__u32 hash = be32_to_cpu(xhash);
+	struct mb_cache_entry *ce;
+	int error;
+
+	ce = mb_cache_entry_alloc(pram_xattr_cache, GFP_NOFS);
+	if (!ce)
+		return -ENOMEM;
+	error = mb_cache_entry_insert(ce, (struct block_device *)sbi, blocknr, hash);
+	if (error) {
+		mb_cache_entry_free(ce);
+		if (error == -EBUSY) {
+			ea_bdebug(blocknr, "already in cache");
+			error = 0;
+		}
+	} else {
+		ea_bdebug(blocknr, "inserting [%x]", (int)hash);
+		mb_cache_entry_release(ce);
+	}
+	return error;
+}
+
+/*
+ * pram_xattr_cmp()
+ *
+ * Compare two extended attribute blocks for equality.
+ *
+ * Returns 0 if the blocks are equal, 1 if they differ, and
+ * a negative error number on errors.
+ */
+static int pram_xattr_cmp(struct pram_xattr_header *header1,
+			  struct pram_xattr_header *header2)
+{
+	struct pram_xattr_entry *entry1, *entry2;
+
+	entry1 = ENTRY(header1+1);
+	entry2 = ENTRY(header2+1);
+	while (!IS_LAST_ENTRY(entry1)) {
+		if (IS_LAST_ENTRY(entry2))
+			return 1;
+		if (entry1->e_hash != entry2->e_hash ||
+		    entry1->e_name_index != entry2->e_name_index ||
+		    entry1->e_name_len != entry2->e_name_len ||
+		    entry1->e_value_size != entry2->e_value_size ||
+		    memcmp(entry1->e_name, entry2->e_name, entry1->e_name_len))
+			return 1;
+		if (entry1->e_value_block != 0 || entry2->e_value_block != 0)
+			return -EIO;
+		if (memcmp((char *)header1 + be16_to_cpu(entry1->e_value_offs),
+			   (char *)header2 + be16_to_cpu(entry2->e_value_offs),
+			   be32_to_cpu(entry1->e_value_size)))
+			return 1;
+
+		entry1 = PRAM_XATTR_NEXT(entry1);
+		entry2 = PRAM_XATTR_NEXT(entry2);
+	}
+	if (!IS_LAST_ENTRY(entry2))
+		return 1;
+	return 0;
+}
+
+/*
+ * pram_xattr_cache_find()
+ *
+ * Find an identical extended attribute block.
+ *
+ * Returns a locked extended block descriptor for the block found, or
+ * NULL if such a block was not found or an error occurred.
+ * The block, however, is not memory unlocked.
+ */
+static struct pram_xblock_desc *pram_xattr_cache_find(struct inode *inode, struct pram_xattr_header *header)
+{
+	__u32 hash = be32_to_cpu(header->h_hash);
+	struct mb_cache_entry *ce;
+	struct pram_xblock_desc *desc;
+	struct super_block *sb = inode->i_sb;
+	struct pram_sb_info *sbi = (struct pram_sb_info *)sb->s_fs_info;
+
+	if (!header->h_hash)
+		return NULL;  /* never share */
+	ea_idebug(inode, "looking for cached blocks [%x]", (int)hash);
+again:
+	ce = mb_cache_entry_find_first(pram_xattr_cache, (struct block_device *)sbi, hash);
+	while (ce) {
+		char *bp;
+
+		if (IS_ERR(ce)) {
+			if (PTR_ERR(ce) == -EAGAIN)
+				goto again;
+			break;
+		}
+
+		bp = pram_get_block(sb, pram_get_block_off(sb, (unsigned long)ce->e_block));
+		if (!bp) {
+			pram_err(sb, "inode %ld: block %ld read error",
+				inode->i_ino, (unsigned long) ce->e_block);
+		} else {
+			desc = LOOKUP_DESC(sbi, ce->e_block);
+			if (!desc) {
+				mb_cache_entry_release(ce);
+				return NULL;
+			}
+			mutex_lock(&desc->lock);
+			if (be32_to_cpu(HDR(bp)->h_refcount) >
+				   PRAM_XATTR_REFCOUNT_MAX) {
+				ea_idebug(inode, "block %ld refcount %d>%d",
+					  (unsigned long) ce->e_block,
+					  be32_to_cpu(HDR(bp)->h_refcount),
+					  PRAM_XATTR_REFCOUNT_MAX);
+			} else if (!pram_xattr_cmp(header, HDR(bp))) {
+				mb_cache_entry_release(ce);
+				return desc;
+			}
+			mutex_unlock(&desc->lock);
+		}
+		ce = mb_cache_entry_find_next(ce, (struct block_device *)sbi, hash);
+	}
+	return NULL;
+}
+
+#define NAME_HASH_SHIFT 5
+#define VALUE_HASH_SHIFT 16
+
+/*
+ * pram_xattr_hash_entry()
+ *
+ * Compute the hash of an extended attribute.
+ */
+static inline void pram_xattr_hash_entry(struct pram_xattr_header *header,
+					 struct pram_xattr_entry *entry)
+{
+	__u32 hash = 0;
+	char *name = entry->e_name;
+	int n;
+
+	for (n = 0; n < entry->e_name_len; n++) {
+		hash = (hash << NAME_HASH_SHIFT) ^
+		       (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^
+		       *name++;
+	}
+
+	if (entry->e_value_block == 0 && entry->e_value_size != 0) {
+		__be32 *value = (__be32 *)((char *)header +
+			be16_to_cpu(entry->e_value_offs));
+		for (n = (be32_to_cpu(entry->e_value_size) +
+		     PRAM_XATTR_ROUND) >> PRAM_XATTR_PAD_BITS; n; n--) {
+			hash = (hash << VALUE_HASH_SHIFT) ^
+			       (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^
+			       be32_to_cpu(*value++);
+		}
+	}
+	entry->e_hash = cpu_to_be32(hash);
+}
+
+#undef NAME_HASH_SHIFT
+#undef VALUE_HASH_SHIFT
+
+#define BLOCK_HASH_SHIFT 16
+
+/*
+ * pram_xattr_rehash()
+ *
+ * Re-compute the extended attribute hash value after an entry has changed.
+ */
+static void pram_xattr_rehash(struct pram_xattr_header *header,
+			      struct pram_xattr_entry *entry)
+{
+	struct pram_xattr_entry *here;
+	__u32 hash = 0;
+
+	pram_xattr_hash_entry(header, entry);
+	here = ENTRY(header+1);
+	while (!IS_LAST_ENTRY(here)) {
+		if (!here->e_hash) {
+			/* Block is not shared if an entry's hash value == 0 */
+			hash = 0;
+			break;
+		}
+		hash = (hash << BLOCK_HASH_SHIFT) ^
+		       (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^
+		       be32_to_cpu(here->e_hash);
+		here = PRAM_XATTR_NEXT(here);
+	}
+	header->h_hash = cpu_to_be32(hash);
+}
+
+#undef BLOCK_HASH_SHIFT
+
+static void init_xblock_desc_once(void *foo)
+{
+	struct pram_xblock_desc *desc = (struct pram_xblock_desc *) foo;
+
+	xblock_desc_init_once(desc);
+}
+
+int __init init_pram_xattr(void)
+{
+	int ret = 0;
+	pram_xattr_cache = mb_cache_create("pram_xattr", 6);
+	if (!pram_xattr_cache) {
+		ret = -ENOMEM;
+		goto fail1;
+	}
+
+	pram_xblock_desc_cache = kmem_cache_create("pram_xblock_desc",
+					     sizeof(struct pram_xblock_desc),
+					     0, (SLAB_RECLAIM_ACCOUNT|
+						SLAB_MEM_SPREAD),
+					     init_xblock_desc_once);
+	if (!pram_xblock_desc_cache) {
+		ret = -ENOMEM;
+		goto fail2;
+	}
+
+	return 0;
+fail2:
+	mb_cache_destroy(pram_xattr_cache);
+fail1:
+	return ret;
+}
+
+void exit_pram_xattr(void)
+{
+	mb_cache_destroy(pram_xattr_cache);
+	kmem_cache_destroy(pram_xblock_desc_cache);
+}

131 lines not on this page · raw message

78 lines not on this page · raw message

65 lines not on this page · raw message

68 lines not on this page · raw message