From a0fca18835d002f276ae2a18f1b9dab6f099c72b Mon Sep 17 00:00:00 2001 From: yangsheng Date: Tue, 22 Jan 2008 08:53:38 +0000 Subject: [PATCH] Branch HEAD b=14686 i=adilger Remove SLES9 kernel patches from HEAD(ldiskfs b1_8_iam). --- ldiskfs/configure.ac | 1 - .../patches/export-ext3-2.6-suse.patch | 35 - .../patches/ext3-check-jbd-errors-2.6.5.patch | 113 - .../patches/ext3-ea-in-inode-2.6-suse.patch | 840 ------ .../patches/ext3-extents-2.6.5.patch | 2951 -------------------- ...t3-extents-multiblock-directio-2.6.5-suse.patch | 157 -- .../patches/ext3-include-fixes-2.6-suse.patch | 20 - .../patches/ext3-mballoc3-suse.patch | 397 --- .../patches/ext3-nanosecond-2.6-suse.patch | 195 -- .../patches/ext3-rename-reserve-2.6-suse.patch | 263 -- .../ext3-sector_t-overflow-2.6.5-suse.patch | 44 - .../patches/ext3-uninit-2.6-suse.patch | 653 ----- .../patches/ext3-wantedi-2.6-suse.patch | 192 -- .../kernel_patches/patches/iopen-2.6-suse.patch | 472 ---- .../kernel_patches/series/ldiskfs-2.6-suse.series | 22 - 15 files changed, 6355 deletions(-) delete mode 100644 ldiskfs/kernel_patches/patches/export-ext3-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-check-jbd-errors-2.6.5.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-ea-in-inode-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-extents-2.6.5.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-extents-multiblock-directio-2.6.5-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-include-fixes-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-mballoc3-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-nanosecond-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-rename-reserve-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-sector_t-overflow-2.6.5-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-uninit-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/ext3-wantedi-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/patches/iopen-2.6-suse.patch delete mode 100644 ldiskfs/kernel_patches/series/ldiskfs-2.6-suse.series diff --git a/ldiskfs/configure.ac b/ldiskfs/configure.ac index 760bc8e..42aed45 100644 --- a/ldiskfs/configure.ac +++ b/ldiskfs/configure.ac @@ -65,7 +65,6 @@ LB_DEFINE_LDISKFS_OPTIONS AC_MSG_CHECKING([which ldiskfs series to use]) case $LINUXRELEASE in -2.6.5*) LDISKFS_SERIES="2.6-suse.series" ;; 2.6.9*) LDISKFS_SERIES="2.6-rhel4.series" ;; 2.6.10-ac*) LDISKFS_SERIES="2.6-fc3.series" ;; 2.6.10*) LDISKFS_SERIES="2.6-rhel4.series" ;; diff --git a/ldiskfs/kernel_patches/patches/export-ext3-2.6-suse.patch b/ldiskfs/kernel_patches/patches/export-ext3-2.6-suse.patch deleted file mode 100644 index c10431b..0000000 --- a/ldiskfs/kernel_patches/patches/export-ext3-2.6-suse.patch +++ /dev/null @@ -1,35 +0,0 @@ -Index: linux-2.6.5-7.283_lustre-1.4.10.1/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-7.283_lustre-1.4.10.1.orig/fs/ext3/super.c 2007-05-30 08:48:29.000000000 +0200 -+++ linux-2.6.5-7.283_lustre-1.4.10.1/fs/ext3/super.c 2007-05-30 08:48:37.000000000 +0200 -@@ -116,6 +116,8 @@ void ext3_journal_abort_handle(const cha - handle->h_err = err; - } - -+EXPORT_SYMBOL(ext3_journal_abort_handle); -+ - static char error_buf[1024]; - - /* Deal with the reporting of failure conditions on a filesystem such as -@@ -1895,6 +1897,8 @@ int ext3_force_commit(struct super_block - return ret; - } - -+EXPORT_SYMBOL(ext3_force_commit); -+ - /* - * Ext3 always journals updates to the superblock itself, so we don't - * have to propagate any other updates to the superblock on disk at this -@@ -2334,6 +2338,12 @@ int ext3_map_inode_page(struct inode *in - unsigned long *blocks, int *created, int create); - EXPORT_SYMBOL(ext3_map_inode_page); - -+EXPORT_SYMBOL(ext3_xattr_get); -+EXPORT_SYMBOL(ext3_xattr_set_handle); -+EXPORT_SYMBOL(ext3_bread); -+EXPORT_SYMBOL(ext3_journal_start); -+EXPORT_SYMBOL(__ext3_journal_stop); -+ - MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); - MODULE_DESCRIPTION("Second Extended Filesystem with journaling extensions"); - MODULE_LICENSE("GPL"); diff --git a/ldiskfs/kernel_patches/patches/ext3-check-jbd-errors-2.6.5.patch b/ldiskfs/kernel_patches/patches/ext3-check-jbd-errors-2.6.5.patch deleted file mode 100644 index e54774f..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-check-jbd-errors-2.6.5.patch +++ /dev/null @@ -1,113 +0,0 @@ -Index: linux-2.6.5-7.201-full/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-7.201-full.orig/include/linux/ext3_fs.h 2006-08-09 17:59:34.000000000 +0400 -+++ linux-2.6.5-7.201-full/include/linux/ext3_fs.h 2006-08-22 12:35:55.000000000 +0400 -@@ -793,6 +793,7 @@ extern void ext3_put_super (struct super - extern void ext3_write_super (struct super_block *); - extern void ext3_write_super_lockfs (struct super_block *); - extern void ext3_unlockfs (struct super_block *); -+extern void ext3_commit_super (struct super_block *, struct ext3_super_block *, int); - extern int ext3_remount (struct super_block *, int *, char *); - extern int ext3_statfs (struct super_block *, struct kstatfs *); - -Index: linux-2.6.5-7.201-full/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-7.201-full.orig/fs/ext3/super.c 2006-08-09 17:59:37.000000000 +0400 -+++ linux-2.6.5-7.201-full/fs/ext3/super.c 2006-08-09 17:59:37.000000000 +0400 -@@ -39,7 +39,7 @@ - static int ext3_load_journal(struct super_block *, struct ext3_super_block *); - static int ext3_create_journal(struct super_block *, struct ext3_super_block *, - int); --static void ext3_commit_super (struct super_block * sb, -+void ext3_commit_super (struct super_block * sb, - struct ext3_super_block * es, - int sync); - static void ext3_mark_recovery_complete(struct super_block * sb, -@@ -1781,7 +1781,7 @@ static int ext3_create_journal(struct su - return 0; - } - --static void ext3_commit_super (struct super_block * sb, -+void ext3_commit_super (struct super_block * sb, - struct ext3_super_block * es, - int sync) - { -Index: linux-2.6.5-7.201-full/fs/ext3/namei.c -=================================================================== ---- linux-2.6.5-7.201-full.orig/fs/ext3/namei.c 2006-08-09 17:59:37.000000000 +0400 -+++ linux-2.6.5-7.201-full/fs/ext3/namei.c 2006-08-09 17:59:37.000000000 +0400 -@@ -1598,7 +1598,7 @@ static int ext3_delete_entry (handle_t * - struct buffer_head * bh) - { - struct ext3_dir_entry_2 * de, * pde; -- int i; -+ int i, err; - - i = 0; - pde = NULL; -@@ -1608,7 +1608,9 @@ static int ext3_delete_entry (handle_t * - return -EIO; - if (de == de_del) { - BUFFER_TRACE(bh, "get_write_access"); -- ext3_journal_get_write_access(handle, bh); -+ err = ext3_journal_get_write_access(handle, bh); -+ if (err) -+ return err; - if (pde) - pde->rec_len = - cpu_to_le16(le16_to_cpu(pde->rec_len) + -Index: linux-2.6.5-7.201-full/fs/ext3/xattr.c -=================================================================== ---- linux-2.6.5-7.201-full.orig/fs/ext3/xattr.c 2006-07-14 01:53:23.000000000 +0400 -+++ linux-2.6.5-7.201-full/fs/ext3/xattr.c 2006-08-09 17:59:37.000000000 +0400 -@@ -107,7 +107,7 @@ ext3_xattr_register(int name_index, stru - { - int error = -EINVAL; - -- if (name_index > 0 && name_index <= EXT3_XATTR_INDEX_MAX) { -+ if (name_index > 0 && name_index < EXT3_XATTR_INDEX_MAX) { - write_lock(&ext3_handler_lock); - if (!ext3_xattr_handlers[name_index-1]) { - ext3_xattr_handlers[name_index-1] = handler; -Index: linux-2.6.5-7.201-full/fs/ext3/inode.c -=================================================================== ---- linux-2.6.5-7.201-full.orig/fs/ext3/inode.c 2006-07-14 01:53:22.000000000 +0400 -+++ linux-2.6.5-7.201-full/fs/ext3/inode.c 2006-08-22 12:35:28.000000000 +0400 -@@ -1517,9 +1517,14 @@ out_stop: - if (end > inode->i_size) { - ei->i_disksize = end; - i_size_write(inode, end); -- err = ext3_mark_inode_dirty(handle, inode); -- if (!ret) -- ret = err; -+ /* -+ * We're going to return a positive `ret' -+ * here due to non-zero-length I/O, so there's -+ * no way of reporting error returns from -+ * ext3_mark_inode_dirty() to userspace. So -+ * ignore it. -+ */ -+ ext3_mark_inode_dirty(handle, inode); - } - } - err = ext3_journal_stop(handle); -@@ -1811,8 +1816,18 @@ ext3_clear_blocks(handle_t *handle, stru - ext3_mark_inode_dirty(handle, inode); - ext3_journal_test_restart(handle, inode); - if (bh) { -+ int err; - BUFFER_TRACE(bh, "retaking write access"); -- ext3_journal_get_write_access(handle, bh); -+ err = ext3_journal_get_write_access(handle, bh); -+ if (err) { -+ struct super_block *sb = inode->i_sb; -+ struct ext3_super_block *es = EXT3_SB(sb)->s_es; -+ printk (KERN_CRIT"EXT3-fs: can't continue truncate\n"); -+ EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; -+ es->s_state |= cpu_to_le16(EXT3_ERROR_FS); -+ ext3_commit_super(sb, es, 1); -+ return; -+ } - } - } - diff --git a/ldiskfs/kernel_patches/patches/ext3-ea-in-inode-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-ea-in-inode-2.6-suse.patch deleted file mode 100644 index 72c25a4..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-ea-in-inode-2.6-suse.patch +++ /dev/null @@ -1,840 +0,0 @@ -%patch -Index: linux-2.6.0/fs/ext3/ialloc.c -=================================================================== ---- linux-2.6.0.orig/fs/ext3/ialloc.c 2004-01-14 18:54:11.000000000 +0300 -+++ linux-2.6.0/fs/ext3/ialloc.c 2004-01-14 18:54:12.000000000 +0300 -@@ -627,6 +627,9 @@ - inode->i_generation = EXT3_SB(sb)->s_next_generation++; - - ei->i_state = EXT3_STATE_NEW; -+ ei->i_extra_isize = -+ (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) ? -+ sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE : 0; - - ret = inode; - if(DQUOT_ALLOC_INODE(inode)) { -Index: linux-2.6.0/fs/ext3/inode.c -=================================================================== ---- linux-2.6.0.orig/fs/ext3/inode.c 2004-01-14 18:54:12.000000000 +0300 -+++ linux-2.6.0/fs/ext3/inode.c 2004-01-14 19:09:46.000000000 +0300 -@@ -2339,7 +2339,7 @@ - * trying to determine the inode's location on-disk and no read need be - * performed. - */ --static int ext3_get_inode_loc(struct inode *inode, -+int ext3_get_inode_loc(struct inode *inode, - struct ext3_iloc *iloc, int in_mem) - { - unsigned long block; -@@ -2547,6 +2547,11 @@ - ei->i_data[block] = raw_inode->i_block[block]; - INIT_LIST_HEAD(&ei->i_orphan); - -+ if (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) -+ ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); -+ else -+ ei->i_extra_isize = 0; -+ - if (S_ISREG(inode->i_mode)) { - inode->i_op = &ext3_file_inode_operations; - inode->i_fop = &ext3_file_operations; -@@ -2682,6 +2687,9 @@ - } else for (block = 0; block < EXT3_N_BLOCKS; block++) - raw_inode->i_block[block] = ei->i_data[block]; - -+ if (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) -+ raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); -+ - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - rc = ext3_journal_dirty_metadata(handle, bh); - if (!err) -@@ -2849,7 +2857,8 @@ ext3_reserve_inode_write(handle_t *handl - { - int err = 0; - if (handle) { -- err = ext3_get_inode_loc(inode, iloc, 1); -+ err = ext3_get_inode_loc(inode, iloc, EXT3_I(inode)->i_state & -+ EXT3_STATE_NEW); - if (!err) { - BUFFER_TRACE(iloc->bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, iloc->bh); -Index: linux-2.6.0/fs/ext3/xattr.c -=================================================================== ---- linux-2.6.0.orig/fs/ext3/xattr.c 2003-12-30 08:33:13.000000000 +0300 -+++ linux-2.6.0/fs/ext3/xattr.c 2004-01-14 18:54:12.000000000 +0300 -@@ -246,17 +246,12 @@ - } - - /* -- * ext3_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. -+ * ext3_xattr_block_get() - * -- * Returns a negative error number on failure, or the number of bytes -- * used / required on success. -+ * routine looks for attribute in EA block and returns it's value and size - */ - int --ext3_xattr_get(struct inode *inode, int name_index, const char *name, -+ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, - void *buffer, size_t buffer_size) - { - struct buffer_head *bh = NULL; -@@ -270,7 +265,6 @@ - - if (name == NULL) - return -EINVAL; -- down_read(&EXT3_I(inode)->xattr_sem); - error = -ENODATA; - if (!EXT3_I(inode)->i_file_acl) - goto cleanup; -@@ -343,15 +337,87 @@ - - cleanup: - brelse(bh); -- up_read(&EXT3_I(inode)->xattr_sem); - - return error; - } - - /* -- * ext3_xattr_list() -+ * ext3_xattr_ibody_get() - * -- * Copy a list of attribute names into the buffer -+ * routine looks for attribute in inode body and returns it's value and size -+ */ -+int -+ext3_xattr_ibody_get(struct inode *inode, int name_index, const char *name, -+ void *buffer, size_t buffer_size) -+{ -+ int size, name_len = strlen(name), storage_size; -+ struct ext3_xattr_entry *last; -+ struct ext3_inode *raw_inode; -+ struct ext3_iloc iloc; -+ char *start, *end; -+ int ret = -ENOENT; -+ -+ if (EXT3_SB(inode->i_sb)->s_inode_size <= EXT3_GOOD_OLD_INODE_SIZE) -+ return -ENOENT; -+ -+ ret = ext3_get_inode_loc(inode, &iloc, 0); -+ if (ret) -+ return ret; -+ raw_inode = ext3_raw_inode(&iloc); -+ -+ storage_size = EXT3_SB(inode->i_sb)->s_inode_size - -+ EXT3_GOOD_OLD_INODE_SIZE - -+ EXT3_I(inode)->i_extra_isize - -+ sizeof(__u32); -+ start = (char *) raw_inode + EXT3_GOOD_OLD_INODE_SIZE + -+ EXT3_I(inode)->i_extra_isize; -+ if (le32_to_cpu((*(__u32*) start)) != EXT3_XATTR_MAGIC) { -+ brelse(iloc.bh); -+ return -ENOENT; -+ } -+ start += sizeof(__u32); -+ end = (char *) raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; -+ -+ last = (struct ext3_xattr_entry *) start; -+ while (!IS_LAST_ENTRY(last)) { -+ struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(last); -+ if (le32_to_cpu(last->e_value_size) > storage_size || -+ (char *) next >= end) { -+ ext3_error(inode->i_sb, "ext3_xattr_ibody_get", -+ "inode %ld", inode->i_ino); -+ brelse(iloc.bh); -+ return -EIO; -+ } -+ if (name_index == last->e_name_index && -+ name_len == last->e_name_len && -+ !memcmp(name, last->e_name, name_len)) -+ goto found; -+ last = next; -+ } -+ -+ /* can't find EA */ -+ brelse(iloc.bh); -+ return -ENOENT; -+ -+found: -+ size = le32_to_cpu(last->e_value_size); -+ if (buffer) { -+ ret = -ERANGE; -+ if (buffer_size >= size) { -+ memcpy(buffer, start + le16_to_cpu(last->e_value_offs), -+ size); -+ ret = size; -+ } -+ } else -+ ret = size; -+ brelse(iloc.bh); -+ return ret; -+} -+ -+/* -+ * ext3_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. - * -@@ -359,7 +425,31 @@ - * used / required on success. - */ - int --ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) -+ext3_xattr_get(struct inode *inode, int name_index, const char *name, -+ void *buffer, size_t buffer_size) -+{ -+ int err; -+ -+ down_read(&EXT3_I(inode)->xattr_sem); -+ -+ /* try to find attribute in inode body */ -+ err = ext3_xattr_ibody_get(inode, name_index, name, -+ buffer, buffer_size); -+ if (err < 0) -+ /* search was unsuccessful, try to find EA in dedicated block */ -+ err = ext3_xattr_block_get(inode, name_index, name, -+ buffer, buffer_size); -+ up_read(&EXT3_I(inode)->xattr_sem); -+ -+ return err; -+} -+ -+/* ext3_xattr_ibody_list() -+ * -+ * generate list of attributes stored in EA block -+ */ -+int -+ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) - { - struct buffer_head *bh = NULL; - struct ext3_xattr_entry *entry; -@@ -370,7 +460,6 @@ - ea_idebug(inode, "buffer=%p, buffer_size=%ld", - buffer, (long)buffer_size); - -- down_read(&EXT3_I(inode)->xattr_sem); - error = 0; - if (!EXT3_I(inode)->i_file_acl) - goto cleanup; -@@ -431,11 +520,138 @@ - - cleanup: - brelse(bh); -- up_read(&EXT3_I(inode)->xattr_sem); - - return error; - } - -+/* ext3_xattr_ibody_list() -+ * -+ * generate list of attributes stored in inode body -+ */ -+int -+ext3_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size) -+{ -+ struct ext3_xattr_entry *last; -+ struct ext3_inode *raw_inode; -+ char *start, *end, *buf; -+ struct ext3_iloc iloc; -+ int storage_size; -+ int ret; -+ int size = 0; -+ -+ if (EXT3_SB(inode->i_sb)->s_inode_size <= EXT3_GOOD_OLD_INODE_SIZE) -+ return 0; -+ -+ ret = ext3_get_inode_loc(inode, &iloc, 0); -+ if (ret) -+ return ret; -+ raw_inode = ext3_raw_inode(&iloc); -+ -+ storage_size = EXT3_SB(inode->i_sb)->s_inode_size - -+ EXT3_GOOD_OLD_INODE_SIZE - -+ EXT3_I(inode)->i_extra_isize - -+ sizeof(__u32); -+ start = (char *) raw_inode + EXT3_GOOD_OLD_INODE_SIZE + -+ EXT3_I(inode)->i_extra_isize; -+ if (le32_to_cpu((*(__u32*) start)) != EXT3_XATTR_MAGIC) { -+ brelse(iloc.bh); -+ return 0; -+ } -+ start += sizeof(__u32); -+ end = (char *) raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; -+ -+ last = (struct ext3_xattr_entry *) start; -+ while (!IS_LAST_ENTRY(last)) { -+ struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(last); -+ struct ext3_xattr_handler *handler; -+ if (le32_to_cpu(last->e_value_size) > storage_size || -+ (char *) next >= end) { -+ ext3_error(inode->i_sb, "ext3_xattr_ibody_list", -+ "inode %ld", inode->i_ino); -+ brelse(iloc.bh); -+ return -EIO; -+ } -+ handler = ext3_xattr_handler(last->e_name_index); -+ if (handler) -+ size += handler->list(NULL, inode, last->e_name, -+ last->e_name_len); -+ last = next; -+ } -+ -+ if (!buffer) { -+ ret = size; -+ goto cleanup; -+ } else { -+ ret = -ERANGE; -+ if (size > buffer_size) -+ goto cleanup; -+ } -+ -+ last = (struct ext3_xattr_entry *) start; -+ buf = buffer; -+ while (!IS_LAST_ENTRY(last)) { -+ struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(last); -+ struct ext3_xattr_handler *handler; -+ handler = ext3_xattr_handler(last->e_name_index); -+ if (handler) -+ buf += handler->list(buf, inode, last->e_name, -+ last->e_name_len); -+ last = next; -+ } -+ ret = size; -+cleanup: -+ brelse(iloc.bh); -+ return ret; -+} -+ -+/* -+ * ext3_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. -+ */ -+int -+ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) -+{ -+ int error; -+ int size = buffer_size; -+ -+ down_read(&EXT3_I(inode)->xattr_sem); -+ -+ /* get list of attributes stored in inode body */ -+ error = ext3_xattr_ibody_list(inode, buffer, buffer_size); -+ if (error < 0) { -+ /* some error occured while collecting -+ * attributes in inode body */ -+ size = 0; -+ goto cleanup; -+ } -+ size = error; -+ -+ /* get list of attributes stored in dedicated block */ -+ if (buffer) { -+ buffer_size -= error; -+ if (buffer_size <= 0) { -+ buffer = NULL; -+ buffer_size = 0; -+ } else -+ buffer += error; -+ } -+ -+ error = ext3_xattr_block_list(inode, buffer, buffer_size); -+ if (error < 0) -+ /* listing was successful, so we return len */ -+ size = 0; -+ -+cleanup: -+ up_read(&EXT3_I(inode)->xattr_sem); -+ return error + size; -+} -+ - /* - * If the EXT3_FEATURE_COMPAT_EXT_ATTR feature of this file system is - * not set, set it. -@@ -457,6 +673,279 @@ - } - - /* -+ * ext3_xattr_ibody_find() -+ * -+ * search attribute and calculate free space in inode body -+ * NOTE: free space includes space our attribute hold -+ */ -+int -+ext3_xattr_ibody_find(struct inode *inode, int name_index, -+ const char *name, struct ext3_xattr_entry *rentry, int *free) -+{ -+ struct ext3_xattr_entry *last; -+ struct ext3_inode *raw_inode; -+ int name_len = strlen(name); -+ int err, storage_size; -+ struct ext3_iloc iloc; -+ char *start, *end; -+ int ret = -ENOENT; -+ -+ if (EXT3_SB(inode->i_sb)->s_inode_size <= EXT3_GOOD_OLD_INODE_SIZE) -+ return ret; -+ -+ err = ext3_get_inode_loc(inode, &iloc, 0); -+ if (err) -+ return -EIO; -+ raw_inode = ext3_raw_inode(&iloc); -+ -+ storage_size = EXT3_SB(inode->i_sb)->s_inode_size - -+ EXT3_GOOD_OLD_INODE_SIZE - -+ EXT3_I(inode)->i_extra_isize - -+ sizeof(__u32); -+ *free = storage_size - sizeof(__u32); -+ start = (char *) raw_inode + EXT3_GOOD_OLD_INODE_SIZE + -+ EXT3_I(inode)->i_extra_isize; -+ if (le32_to_cpu((*(__u32*) start)) != EXT3_XATTR_MAGIC) { -+ brelse(iloc.bh); -+ return -ENOENT; -+ } -+ start += sizeof(__u32); -+ end = (char *) raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; -+ -+ last = (struct ext3_xattr_entry *) start; -+ while (!IS_LAST_ENTRY(last)) { -+ struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(last); -+ if (le32_to_cpu(last->e_value_size) > storage_size || -+ (char *) next >= end) { -+ ext3_error(inode->i_sb, "ext3_xattr_ibody_find", -+ "inode %ld", inode->i_ino); -+ brelse(iloc.bh); -+ return -EIO; -+ } -+ -+ if (name_index == last->e_name_index && -+ name_len == last->e_name_len && -+ !memcmp(name, last->e_name, name_len)) { -+ memcpy(rentry, last, sizeof(struct ext3_xattr_entry)); -+ ret = 0; -+ } else { -+ *free -= EXT3_XATTR_LEN(last->e_name_len); -+ *free -= le32_to_cpu(last->e_value_size); -+ } -+ last = next; -+ } -+ -+ brelse(iloc.bh); -+ return ret; -+} -+ -+/* -+ * ext3_xattr_block_find() -+ * -+ * search attribute and calculate free space in EA block (if it allocated) -+ * NOTE: free space includes space our attribute hold -+ */ -+int -+ext3_xattr_block_find(struct inode *inode, int name_index, const char *name, -+ struct ext3_xattr_entry *rentry, int *free) -+{ -+ struct buffer_head *bh = NULL; -+ struct ext3_xattr_entry *entry; -+ char *end; -+ int name_len, error = -ENOENT; -+ -+ if (!EXT3_I(inode)->i_file_acl) { -+ *free = inode->i_sb->s_blocksize - -+ sizeof(struct ext3_xattr_header) - -+ sizeof(__u32); -+ return -ENOENT; -+ } -+ ea_idebug(inode, "reading block %d", EXT3_I(inode)->i_file_acl); -+ bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); -+ if (!bh) -+ return -EIO; -+ ea_bdebug(bh, "b_count=%d, refcount=%d", -+ atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount)); -+ end = bh->b_data + bh->b_size; -+ if (HDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || -+ HDR(bh)->h_blocks != cpu_to_le32(1)) { -+bad_block: ext3_error(inode->i_sb, "ext3_xattr_get", -+ "inode %ld: bad block %d", inode->i_ino, -+ EXT3_I(inode)->i_file_acl); -+ brelse(bh); -+ return -EIO; -+ } -+ /* find named attribute */ -+ name_len = strlen(name); -+ *free = bh->b_size - sizeof(__u32); -+ -+ entry = FIRST_ENTRY(bh); -+ while (!IS_LAST_ENTRY(entry)) { -+ struct ext3_xattr_entry *next = -+ EXT3_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) { -+ memcpy(rentry, entry, sizeof(struct ext3_xattr_entry)); -+ error = 0; -+ } else { -+ *free -= EXT3_XATTR_LEN(entry->e_name_len); -+ *free -= le32_to_cpu(entry->e_value_size); -+ } -+ entry = next; -+ } -+ brelse(bh); -+ -+ return error; -+} -+ -+/* -+ * ext3_xattr_inode_set() -+ * -+ * this routine add/remove/replace attribute in inode body -+ */ -+int -+ext3_xattr_ibody_set(handle_t *handle, struct inode *inode, int name_index, -+ const char *name, const void *value, size_t value_len, -+ int flags) -+{ -+ struct ext3_xattr_entry *last, *next, *here = NULL; -+ struct ext3_inode *raw_inode; -+ int name_len = strlen(name); -+ int esize = EXT3_XATTR_LEN(name_len); -+ struct buffer_head *bh; -+ int err, storage_size; -+ struct ext3_iloc iloc; -+ int free, min_offs; -+ char *start, *end; -+ -+ if (EXT3_SB(inode->i_sb)->s_inode_size <= EXT3_GOOD_OLD_INODE_SIZE) -+ return -ENOSPC; -+ -+ err = ext3_get_inode_loc(inode, &iloc, 0); -+ if (err) -+ return err; -+ raw_inode = ext3_raw_inode(&iloc); -+ bh = iloc.bh; -+ -+ storage_size = EXT3_SB(inode->i_sb)->s_inode_size - -+ EXT3_GOOD_OLD_INODE_SIZE - -+ EXT3_I(inode)->i_extra_isize - -+ sizeof(__u32); -+ start = (char *) raw_inode + EXT3_GOOD_OLD_INODE_SIZE + -+ EXT3_I(inode)->i_extra_isize; -+ if ((*(__u32*) start) != EXT3_XATTR_MAGIC) { -+ /* inode had no attributes before */ -+ *((__u32*) start) = cpu_to_le32(EXT3_XATTR_MAGIC); -+ } -+ start += sizeof(__u32); -+ end = (char *) raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; -+ min_offs = storage_size; -+ free = storage_size - sizeof(__u32); -+ -+ last = (struct ext3_xattr_entry *) start; -+ while (!IS_LAST_ENTRY(last)) { -+ next = EXT3_XATTR_NEXT(last); -+ if (le32_to_cpu(last->e_value_size) > storage_size || -+ (char *) next >= end) { -+ ext3_error(inode->i_sb, "ext3_xattr_ibody_set", -+ "inode %ld", inode->i_ino); -+ brelse(bh); -+ return -EIO; -+ } -+ -+ if (last->e_value_size) { -+ int offs = le16_to_cpu(last->e_value_offs); -+ if (offs < min_offs) -+ min_offs = offs; -+ } -+ if (name_index == last->e_name_index && -+ name_len == last->e_name_len && -+ !memcmp(name, last->e_name, name_len)) -+ here = last; -+ else { -+ /* we calculate all but our attribute -+ * because it will be removed before changing */ -+ free -= EXT3_XATTR_LEN(last->e_name_len); -+ free -= le32_to_cpu(last->e_value_size); -+ } -+ last = next; -+ } -+ -+ if (value && (esize + value_len > free)) { -+ brelse(bh); -+ return -ENOSPC; -+ } -+ -+ err = ext3_reserve_inode_write(handle, inode, &iloc); -+ if (err) { -+ brelse(bh); -+ return err; -+ } -+ -+ if (here) { -+ /* time to remove old value */ -+ struct ext3_xattr_entry *e; -+ int size = le32_to_cpu(here->e_value_size); -+ int border = le16_to_cpu(here->e_value_offs); -+ char *src; -+ -+ /* move tail */ -+ memmove(start + min_offs + size, start + min_offs, -+ border - min_offs); -+ -+ /* recalculate offsets */ -+ e = (struct ext3_xattr_entry *) start; -+ while (!IS_LAST_ENTRY(e)) { -+ struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(e); -+ int offs = le16_to_cpu(e->e_value_offs); -+ if (offs < border) -+ e->e_value_offs = -+ cpu_to_le16(offs + size); -+ e = next; -+ } -+ min_offs += size; -+ -+ /* remove entry */ -+ border = EXT3_XATTR_LEN(here->e_name_len); -+ src = (char *) here + EXT3_XATTR_LEN(here->e_name_len); -+ size = (char *) last - src; -+ if ((char *) here + size > end) -+ printk("ALERT at %s:%d: 0x%p + %d > 0x%p\n", -+ __FILE__, __LINE__, here, size, end); -+ memmove(here, src, size); -+ last = (struct ext3_xattr_entry *) ((char *) last - border); -+ *((__u32 *) last) = 0; -+ } -+ -+ if (value) { -+ int offs = min_offs - value_len; -+ /* use last to create new entry */ -+ last->e_name_len = strlen(name); -+ last->e_name_index = name_index; -+ last->e_value_offs = cpu_to_le16(offs); -+ last->e_value_size = cpu_to_le32(value_len); -+ last->e_hash = last->e_value_block = 0; -+ memset(last->e_name, 0, esize); -+ memcpy(last->e_name, name, last->e_name_len); -+ if (start + offs + value_len > end) -+ printk("ALERT at %s:%d: 0x%p + %d + %zd > 0x%p\n", -+ __FILE__, __LINE__, start, offs, -+ value_len, end); -+ memcpy(start + offs, value, value_len); -+ last = EXT3_XATTR_NEXT(last); -+ *((__u32 *) last) = 0; -+ } -+ -+ ext3_mark_iloc_dirty(handle, inode, &iloc); -+ brelse(bh); -+ -+ return 0; -+} -+ -+/* - * ext3_xattr_set_handle() - * - * Create, replace or remove an extended attribute for this inode. Buffer -@@ -470,6 +959,104 @@ - */ - int - ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, -+ const char *name, const void *value, size_t value_len, -+ int flags) -+{ -+ struct ext3_xattr_entry entry; -+ int err, where = 0, found = 0, total; -+ int free1 = -1, free2 = -1; -+ int name_len; -+ -+ ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", -+ name_index, name, value, (long)value_len); -+ -+ if (IS_RDONLY(inode)) -+ return -EROFS; -+ if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) -+ return -EPERM; -+ if (value == NULL) -+ value_len = 0; -+ if (name == NULL) -+ return -EINVAL; -+ name_len = strlen(name); -+ if (name_len > 255 || value_len > inode->i_sb->s_blocksize) -+ return -ERANGE; -+ down_write(&EXT3_I(inode)->xattr_sem); -+ -+ /* try to find attribute in inode body */ -+ err = ext3_xattr_ibody_find(inode, name_index, name, &entry, &free1); -+ if (err == 0) { -+ /* found EA in inode */ -+ found = 1; -+ where = 0; -+ } else if (err == -ENOENT) { -+ /* there is no such attribute in inode body */ -+ /* try to find attribute in dedicated block */ -+ err = ext3_xattr_block_find(inode, name_index, name, -+ &entry, &free2); -+ if (err != 0 && err != -ENOENT) { -+ /* not found EA in block */ -+ goto finish; -+ } else if (err == 0) { -+ /* found EA in block */ -+ where = 1; -+ found = 1; -+ } -+ } else -+ goto finish; -+ -+ /* check flags: may replace? may create ? */ -+ if (found && (flags & XATTR_CREATE)) { -+ err = -EEXIST; -+ goto finish; -+ } else if (!found && (flags & XATTR_REPLACE)) { -+ err = -ENODATA; -+ goto finish; -+ } -+ -+ /* check if we have enough space to store attribute */ -+ total = EXT3_XATTR_LEN(strlen(name)) + value_len; -+ if (free1 >= 0 && total > free1 && free2 >= 0 && total > free2) { -+ /* have no enough space */ -+ err = -ENOSPC; -+ goto finish; -+ } -+ -+ /* time to remove attribute */ -+ if (found) { -+ if (where == 0) { -+ /* EA is stored in inode body */ -+ ext3_xattr_ibody_set(handle, inode, name_index, name, -+ NULL, 0, flags); -+ } else { -+ /* EA is stored in separated block */ -+ ext3_xattr_block_set(handle, inode, name_index, name, -+ NULL, 0, flags); -+ } -+ } -+ -+ /* try to store EA in inode body */ -+ err = ext3_xattr_ibody_set(handle, inode, name_index, name, -+ value, value_len, flags); -+ if (err) { -+ /* can't store EA in inode body */ -+ /* try to store in block */ -+ err = ext3_xattr_block_set(handle, inode, name_index, -+ name, value, value_len, flags); -+ } -+ -+finish: -+ up_write(&EXT3_I(inode)->xattr_sem); -+ return err; -+} -+ -+/* -+ * ext3_xattr_block_set() -+ * -+ * this routine add/remove/replace attribute in EA block -+ */ -+int -+ext3_xattr_block_set(handle_t *handle, struct inode *inode, int name_index, - const char *name, const void *value, size_t value_len, - int flags) - { -@@ -492,22 +1078,7 @@ - * 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 (IS_RDONLY(inode)) -- return -EROFS; -- if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) -- return -EPERM; -- 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; -- down_write(&EXT3_I(inode)->xattr_sem); - if (EXT3_I(inode)->i_file_acl) { - /* The inode already has an extended attribute block. */ - bh = sb_bread(sb, EXT3_I(inode)->i_file_acl); -@@ -733,7 +1304,6 @@ - brelse(bh); - if (!(bh && header == HDR(bh))) - kfree(header); -- up_write(&EXT3_I(inode)->xattr_sem); - - return error; - } -Index: linux-2.6.0/fs/ext3/xattr.h -=================================================================== ---- linux-2.6.0.orig/fs/ext3/xattr.h 2003-06-24 18:04:43.000000000 +0400 -+++ linux-2.6.0/fs/ext3/xattr.h 2004-01-14 18:54:12.000000000 +0300 -@@ -77,7 +77,8 @@ - extern int ext3_xattr_get(struct inode *, int, const char *, void *, size_t); - extern int ext3_xattr_list(struct inode *, char *, size_t); - extern int ext3_xattr_set(struct inode *, int, const char *, const void *, size_t, int); --extern int ext3_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); -+extern int ext3_xattr_set_handle(handle_t *, struct inode *, int, const char *,const void *,size_t,int); -+extern int ext3_xattr_block_set(handle_t *, struct inode *, int, const char *,const void *,size_t,int); - - extern void ext3_xattr_delete_inode(handle_t *, struct inode *); - extern void ext3_xattr_put_super(struct super_block *); -Index: linux-2.6.0/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.0.orig/include/linux/ext3_fs.h 2004-01-14 18:54:11.000000000 +0300 -+++ linux-2.6.0/include/linux/ext3_fs.h 2004-01-14 18:54:12.000000000 +0300 -@@ -265,6 +265,8 @@ - __u32 m_i_reserved2[2]; - } masix2; - } osd2; /* OS dependent 2 */ -+ __u16 i_extra_isize; -+ __u16 i_pad1; - }; - - #define i_size_high i_dir_acl -@@ -721,6 +723,7 @@ - extern int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int); - extern struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *); - extern struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *); -+int ext3_get_inode_loc(struct inode *inode, struct ext3_iloc *iloc, int in_mem); - - extern void ext3_read_inode (struct inode *); - extern void ext3_write_inode (struct inode *, int); -Index: linux-2.6.0/include/linux/ext3_fs_i.h -=================================================================== ---- linux-2.6.0.orig/include/linux/ext3_fs_i.h 2003-12-30 08:32:44.000000000 +0300 -+++ linux-2.6.0/include/linux/ext3_fs_i.h 2004-01-14 18:54:12.000000000 +0300 -@@ -96,6 +96,9 @@ - */ - loff_t i_disksize; - -+ /* on-disk additional length */ -+ __u16 i_extra_isize; -+ - /* - * truncate_sem is for serialising ext3_truncate() against - * ext3_getblock(). In the 2.4 ext2 design, great chunks of inode's - -%diffstat - fs/ext3/ialloc.c | 5 - fs/ext3/inode.c | 10 - fs/ext3/xattr.c | 634 +++++++++++++++++++++++++++++++++++++++++++--- - fs/ext3/xattr.h | 3 - include/linux/ext3_fs.h | 2 - include/linux/ext3_fs_i.h | 3 - 6 files changed, 623 insertions(+), 34 deletions(-) - diff --git a/ldiskfs/kernel_patches/patches/ext3-extents-2.6.5.patch b/ldiskfs/kernel_patches/patches/ext3-extents-2.6.5.patch deleted file mode 100644 index b6c37c1..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-extents-2.6.5.patch +++ /dev/null @@ -1,2951 +0,0 @@ -%patch -Index: linux-2.6.5-sles9/fs/ext3/extents.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/extents.c 2005-02-17 22:07:57.023609040 +0300 -+++ linux-2.6.5-sles9/fs/ext3/extents.c 2005-02-23 01:02:37.396435640 +0300 -@@ -0,0 +1,2361 @@ -+/* -+ * Copyright(c) 2003, 2004, 2005, Cluster File Systems, Inc, info@clusterfs.com -+ * Written by Alex Tomas -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public Licens -+ * along with this program; if not, write to the Free Software -+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- -+ */ -+ -+/* -+ * Extents support for EXT3 -+ * -+ * TODO: -+ * - ext3_ext_walk_space() sould not use ext3_ext_find_extent() -+ * - ext3_ext_calc_credits() could take 'mergable' into account -+ * - ext3*_error() should be used in some situations -+ * - find_goal() [to be tested and improved] -+ * - smart tree reduction -+ * - arch-independence -+ * common on-disk format for big/little-endian arch -+ */ -+ -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+ -+ -+static inline int ext3_ext_check_header(struct ext3_extent_header *eh) -+{ -+ if (eh->eh_magic != EXT3_EXT_MAGIC) { -+ printk(KERN_ERR "EXT3-fs: invalid magic = 0x%x\n", -+ (unsigned)eh->eh_magic); -+ return -EIO; -+ } -+ if (eh->eh_max == 0) { -+ printk(KERN_ERR "EXT3-fs: invalid eh_max = %u\n", -+ (unsigned)eh->eh_max); -+ return -EIO; -+ } -+ if (eh->eh_entries > eh->eh_max) { -+ printk(KERN_ERR "EXT3-fs: invalid eh_entries = %u\n", -+ (unsigned)eh->eh_entries); -+ return -EIO; -+ } -+ return 0; -+} -+ -+static handle_t *ext3_ext_journal_restart(handle_t *handle, int needed) -+{ -+ int err; -+ -+ if (handle->h_buffer_credits > needed) -+ return handle; -+ if (!ext3_journal_extend(handle, needed)) -+ return handle; -+ err = ext3_journal_restart(handle, needed); -+ -+ return handle; -+} -+ -+static int inline -+ext3_ext_get_access_for_root(handle_t *h, struct ext3_extents_tree *tree) -+{ -+ if (tree->ops->get_write_access) -+ return tree->ops->get_write_access(h,tree->buffer); -+ else -+ return 0; -+} -+ -+static int inline -+ext3_ext_mark_root_dirty(handle_t *h, struct ext3_extents_tree *tree) -+{ -+ if (tree->ops->mark_buffer_dirty) -+ return tree->ops->mark_buffer_dirty(h,tree->buffer); -+ else -+ return 0; -+} -+ -+/* -+ * could return: -+ * - EROFS -+ * - ENOMEM -+ */ -+static int ext3_ext_get_access(handle_t *handle, -+ struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ int err; -+ -+ if (path->p_bh) { -+ /* path points to block */ -+ err = ext3_journal_get_write_access(handle, path->p_bh); -+ } else { -+ /* path points to leaf/index in inode body */ -+ err = ext3_ext_get_access_for_root(handle, tree); -+ } -+ return err; -+} -+ -+/* -+ * could return: -+ * - EROFS -+ * - ENOMEM -+ * - EIO -+ */ -+static int ext3_ext_dirty(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ int err; -+ if (path->p_bh) { -+ /* path points to block */ -+ err =ext3_journal_dirty_metadata(handle, path->p_bh); -+ } else { -+ /* path points to leaf/index in inode body */ -+ err = ext3_ext_mark_root_dirty(handle, tree); -+ } -+ return err; -+} -+ -+static int inline -+ext3_ext_new_block(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, struct ext3_extent *ex, -+ int *err) -+{ -+ int goal, depth, newblock; -+ struct inode *inode; -+ -+ EXT_ASSERT(tree); -+ if (tree->ops->new_block) -+ return tree->ops->new_block(handle, tree, path, ex, err); -+ -+ inode = tree->inode; -+ depth = EXT_DEPTH(tree); -+ if (path && depth > 0) { -+ goal = path[depth-1].p_block; -+ } else { -+ struct ext3_inode_info *ei = EXT3_I(inode); -+ unsigned long bg_start; -+ unsigned long colour; -+ -+ bg_start = (ei->i_block_group * -+ EXT3_BLOCKS_PER_GROUP(inode->i_sb)) + -+ le32_to_cpu(EXT3_SB(inode->i_sb)->s_es->s_first_data_block); -+ colour = (current->pid % 16) * -+ (EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16); -+ goal = bg_start + colour; -+ } -+ -+ newblock = ext3_new_block(handle, inode, goal, err); -+ return newblock; -+} -+ -+static inline void ext3_ext_tree_changed(struct ext3_extents_tree *tree) -+{ -+ struct ext3_extent_header *neh = EXT_ROOT_HDR(tree); -+ neh->eh_generation = ((EXT_FLAGS(neh) & ~EXT_FLAGS_CLR_UNKNOWN) << 24) | -+ (EXT_HDR_GEN(neh) + 1); -+} -+ -+static inline int ext3_ext_space_block(struct ext3_extents_tree *tree) -+{ -+ int size; -+ -+ size = (tree->inode->i_sb->s_blocksize - -+ sizeof(struct ext3_extent_header)) / -+ sizeof(struct ext3_extent); -+#ifdef AGRESSIVE_TEST -+ size = 6; -+#endif -+ return size; -+} -+ -+static inline int ext3_ext_space_block_idx(struct ext3_extents_tree *tree) -+{ -+ int size; -+ -+ size = (tree->inode->i_sb->s_blocksize - -+ sizeof(struct ext3_extent_header)) / -+ sizeof(struct ext3_extent_idx); -+#ifdef AGRESSIVE_TEST -+ size = 5; -+#endif -+ return size; -+} -+ -+static inline int ext3_ext_space_root(struct ext3_extents_tree *tree) -+{ -+ int size; -+ -+ size = (tree->buffer_len - sizeof(struct ext3_extent_header)) / -+ sizeof(struct ext3_extent); -+#ifdef AGRESSIVE_TEST -+ size = 3; -+#endif -+ return size; -+} -+ -+static inline int ext3_ext_space_root_idx(struct ext3_extents_tree *tree) -+{ -+ int size; -+ -+ size = (tree->buffer_len - sizeof(struct ext3_extent_header)) / -+ sizeof(struct ext3_extent_idx); -+#ifdef AGRESSIVE_TEST -+ size = 4; -+#endif -+ return size; -+} -+ -+static void ext3_ext_show_path(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+#ifdef EXT_DEBUG -+ int k, l = path->p_depth; -+ -+ ext_debug(tree, "path:"); -+ for (k = 0; k <= l; k++, path++) { -+ if (path->p_idx) { -+ ext_debug(tree, " %d->%d", path->p_idx->ei_block, -+ path->p_idx->ei_leaf); -+ } else if (path->p_ext) { -+ ext_debug(tree, " %d:%d:%d", -+ path->p_ext->ee_block, -+ path->p_ext->ee_len, -+ path->p_ext->ee_start); -+ } else -+ ext_debug(tree, " []"); -+ } -+ ext_debug(tree, "\n"); -+#endif -+} -+ -+static void ext3_ext_show_leaf(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+#ifdef EXT_DEBUG -+ int depth = EXT_DEPTH(tree); -+ struct ext3_extent_header *eh; -+ struct ext3_extent *ex; -+ int i; -+ -+ if (!path) -+ return; -+ -+ eh = path[depth].p_hdr; -+ ex = EXT_FIRST_EXTENT(eh); -+ -+ for (i = 0; i < eh->eh_entries; i++, ex++) { -+ ext_debug(tree, "%d:%d:%d ", -+ ex->ee_block, ex->ee_len, ex->ee_start); -+ } -+ ext_debug(tree, "\n"); -+#endif -+} -+ -+static void ext3_ext_drop_refs(struct ext3_ext_path *path) -+{ -+ int depth = path->p_depth; -+ int i; -+ -+ for (i = 0; i <= depth; i++, path++) { -+ if (path->p_bh) { -+ brelse(path->p_bh); -+ path->p_bh = NULL; -+ } -+ } -+} -+ -+/* -+ * binary search for closest index by given block -+ */ -+static inline void -+ext3_ext_binsearch_idx(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, int block) -+{ -+ struct ext3_extent_header *eh = path->p_hdr; -+ struct ext3_extent_idx *ix; -+ int l = 0, k, r; -+ -+ EXT_ASSERT(eh->eh_magic == EXT3_EXT_MAGIC); -+ EXT_ASSERT(eh->eh_entries <= eh->eh_max); -+ EXT_ASSERT(eh->eh_entries > 0); -+ -+ ext_debug(tree, "binsearch for %d(idx): ", block); -+ -+ path->p_idx = ix = EXT_FIRST_INDEX(eh); -+ -+ r = k = eh->eh_entries; -+ while (k > 1) { -+ k = (r - l) / 2; -+ if (block < ix[l + k].ei_block) -+ r -= k; -+ else -+ l += k; -+ ext_debug(tree, "%d:%d:%d ", k, l, r); -+ } -+ -+ ix += l; -+ path->p_idx = ix; -+ ext_debug(tree," -> %d->%d ",path->p_idx->ei_block,path->p_idx->ei_leaf); -+ -+ while (l++ < r) { -+ if (block < ix->ei_block) -+ break; -+ path->p_idx = ix++; -+ } -+ ext_debug(tree, " -> %d->%d\n", path->p_idx->ei_block, -+ path->p_idx->ei_leaf); -+ -+#ifdef CHECK_BINSEARCH -+ { -+ struct ext3_extent_idx *chix; -+ -+ chix = ix = EXT_FIRST_INDEX(eh); -+ for (k = 0; k < eh->eh_entries; k++, ix++) { -+ if (k != 0 && ix->ei_block <= ix[-1].ei_block) { -+ printk("k=%d, ix=0x%p, first=0x%p\n", k, -+ ix, EXT_FIRST_INDEX(eh)); -+ printk("%u <= %u\n", -+ ix->ei_block,ix[-1].ei_block); -+ } -+ EXT_ASSERT(k == 0 || ix->ei_block > ix[-1].ei_block); -+ if (block < ix->ei_block) -+ break; -+ chix = ix; -+ } -+ EXT_ASSERT(chix == path->p_idx); -+ } -+#endif -+} -+ -+/* -+ * binary search for closest extent by given block -+ */ -+static inline void -+ext3_ext_binsearch(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, int block) -+{ -+ struct ext3_extent_header *eh = path->p_hdr; -+ struct ext3_extent *ex; -+ int l = 0, k, r; -+ -+ EXT_ASSERT(eh->eh_magic == EXT3_EXT_MAGIC); -+ EXT_ASSERT(eh->eh_entries <= eh->eh_max); -+ -+ if (eh->eh_entries == 0) { -+ /* -+ * this leaf is empty yet: -+ * we get such a leaf in split/add case -+ */ -+ return; -+ } -+ -+ ext_debug(tree, "binsearch for %d: ", block); -+ -+ path->p_ext = ex = EXT_FIRST_EXTENT(eh); -+ -+ r = k = eh->eh_entries; -+ while (k > 1) { -+ k = (r - l) / 2; -+ if (block < ex[l + k].ee_block) -+ r -= k; -+ else -+ l += k; -+ ext_debug(tree, "%d:%d:%d ", k, l, r); -+ } -+ -+ ex += l; -+ path->p_ext = ex; -+ ext_debug(tree, " -> %d:%d:%d ", path->p_ext->ee_block, -+ path->p_ext->ee_start, path->p_ext->ee_len); -+ -+ while (l++ < r) { -+ if (block < ex->ee_block) -+ break; -+ path->p_ext = ex++; -+ } -+ ext_debug(tree, " -> %d:%d:%d\n", path->p_ext->ee_block, -+ path->p_ext->ee_start, path->p_ext->ee_len); -+ -+#ifdef CHECK_BINSEARCH -+ { -+ struct ext3_extent *chex; -+ -+ chex = ex = EXT_FIRST_EXTENT(eh); -+ for (k = 0; k < eh->eh_entries; k++, ex++) { -+ EXT_ASSERT(k == 0 || ex->ee_block > ex[-1].ee_block); -+ if (block < ex->ee_block) -+ break; -+ chex = ex; -+ } -+ EXT_ASSERT(chex == path->p_ext); -+ } -+#endif -+} -+ -+int ext3_extent_tree_init(handle_t *handle, struct ext3_extents_tree *tree) -+{ -+ struct ext3_extent_header *eh; -+ -+ BUG_ON(tree->buffer_len == 0); -+ ext3_ext_get_access_for_root(handle, tree); -+ eh = EXT_ROOT_HDR(tree); -+ eh->eh_depth = 0; -+ eh->eh_entries = 0; -+ eh->eh_magic = EXT3_EXT_MAGIC; -+ eh->eh_max = ext3_ext_space_root(tree); -+ ext3_ext_mark_root_dirty(handle, tree); -+ ext3_ext_invalidate_cache(tree); -+ return 0; -+} -+ -+struct ext3_ext_path * -+ext3_ext_find_extent(struct ext3_extents_tree *tree, int block, -+ struct ext3_ext_path *path) -+{ -+ struct ext3_extent_header *eh; -+ struct buffer_head *bh; -+ int depth, i, ppos = 0; -+ -+ EXT_ASSERT(tree); -+ EXT_ASSERT(tree->inode); -+ EXT_ASSERT(tree->root); -+ -+ eh = EXT_ROOT_HDR(tree); -+ EXT_ASSERT(eh); -+ if (ext3_ext_check_header(eh)) { -+ /* don't free previously allocated path -+ * -- caller should take care */ -+ path = NULL; -+ goto err; -+ } -+ -+ i = depth = EXT_DEPTH(tree); -+ EXT_ASSERT(eh->eh_max); -+ EXT_ASSERT(eh->eh_magic == EXT3_EXT_MAGIC); -+ -+ /* account possible depth increase */ -+ if (!path) { -+ path = kmalloc(sizeof(struct ext3_ext_path) * (depth + 2), -+ GFP_NOFS); -+ if (!path) -+ return ERR_PTR(-ENOMEM); -+ } -+ memset(path, 0, sizeof(struct ext3_ext_path) * (depth + 1)); -+ path[0].p_hdr = eh; -+ -+ /* walk through the tree */ -+ while (i) { -+ ext_debug(tree, "depth %d: num %d, max %d\n", -+ ppos, eh->eh_entries, eh->eh_max); -+ ext3_ext_binsearch_idx(tree, path + ppos, block); -+ path[ppos].p_block = path[ppos].p_idx->ei_leaf; -+ path[ppos].p_depth = i; -+ path[ppos].p_ext = NULL; -+ -+ bh = sb_bread(tree->inode->i_sb, path[ppos].p_block); -+ if (!bh) -+ goto err; -+ -+ eh = EXT_BLOCK_HDR(bh); -+ ppos++; -+ EXT_ASSERT(ppos <= depth); -+ path[ppos].p_bh = bh; -+ path[ppos].p_hdr = eh; -+ i--; -+ -+ if (ext3_ext_check_header(eh)) -+ goto err; -+ } -+ -+ path[ppos].p_depth = i; -+ path[ppos].p_hdr = eh; -+ path[ppos].p_ext = NULL; -+ path[ppos].p_idx = NULL; -+ -+ if (ext3_ext_check_header(eh)) -+ goto err; -+ -+ /* find extent */ -+ ext3_ext_binsearch(tree, path + ppos, block); -+ -+ ext3_ext_show_path(tree, path); -+ -+ return path; -+ -+err: -+ printk(KERN_ERR "EXT3-fs: header is corrupted!\n"); -+ if (path) { -+ ext3_ext_drop_refs(path); -+ kfree(path); -+ } -+ return ERR_PTR(-EIO); -+} -+ -+/* -+ * insert new index [logical;ptr] into the block at cupr -+ * it check where to insert: before curp or after curp -+ */ -+static int ext3_ext_insert_index(handle_t *handle, -+ struct ext3_extents_tree *tree, -+ struct ext3_ext_path *curp, -+ int logical, int ptr) -+{ -+ struct ext3_extent_idx *ix; -+ int len, err; -+ -+ if ((err = ext3_ext_get_access(handle, tree, curp))) -+ return err; -+ -+ EXT_ASSERT(logical != curp->p_idx->ei_block); -+ len = EXT_MAX_INDEX(curp->p_hdr) - curp->p_idx; -+ if (logical > curp->p_idx->ei_block) { -+ /* insert after */ -+ if (curp->p_idx != EXT_LAST_INDEX(curp->p_hdr)) { -+ len = (len - 1) * sizeof(struct ext3_extent_idx); -+ len = len < 0 ? 0 : len; -+ ext_debug(tree, "insert new index %d after: %d. " -+ "move %d from 0x%p to 0x%p\n", -+ logical, ptr, len, -+ (curp->p_idx + 1), (curp->p_idx + 2)); -+ memmove(curp->p_idx + 2, curp->p_idx + 1, len); -+ } -+ ix = curp->p_idx + 1; -+ } else { -+ /* insert before */ -+ len = len * sizeof(struct ext3_extent_idx); -+ len = len < 0 ? 0 : len; -+ ext_debug(tree, "insert new index %d before: %d. " -+ "move %d from 0x%p to 0x%p\n", -+ logical, ptr, len, -+ curp->p_idx, (curp->p_idx + 1)); -+ memmove(curp->p_idx + 1, curp->p_idx, len); -+ ix = curp->p_idx; -+ } -+ -+ ix->ei_block = logical; -+ ix->ei_leaf = ptr; -+ ix->ei_leaf_hi = ix->ei_unused = 0; -+ curp->p_hdr->eh_entries++; -+ -+ EXT_ASSERT(curp->p_hdr->eh_entries <= curp->p_hdr->eh_max); -+ EXT_ASSERT(ix <= EXT_LAST_INDEX(curp->p_hdr)); -+ -+ err = ext3_ext_dirty(handle, tree, curp); -+ ext3_std_error(tree->inode->i_sb, err); -+ -+ return err; -+} -+ -+/* -+ * routine inserts new subtree into the path, using free index entry -+ * at depth 'at: -+ * - allocates all needed blocks (new leaf and all intermediate index blocks) -+ * - makes decision where to split -+ * - moves remaining extens and index entries (right to the split point) -+ * into the newly allocated blocks -+ * - initialize subtree -+ */ -+static int ext3_ext_split(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_extent *newext, int at) -+{ -+ struct buffer_head *bh = NULL; -+ int depth = EXT_DEPTH(tree); -+ struct ext3_extent_header *neh; -+ struct ext3_extent_idx *fidx; -+ struct ext3_extent *ex; -+ int i = at, k, m, a; -+ unsigned long newblock, oldblock, border; -+ int *ablocks = NULL; /* array of allocated blocks */ -+ int err = 0; -+ -+ /* make decision: where to split? */ -+ /* FIXME: now desicion is simplest: at current extent */ -+ -+ /* if current leaf will be splitted, then we should use -+ * border from split point */ -+ EXT_ASSERT(path[depth].p_ext <= EXT_MAX_EXTENT(path[depth].p_hdr)); -+ if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) { -+ border = path[depth].p_ext[1].ee_block; -+ ext_debug(tree, "leaf will be splitted." -+ " next leaf starts at %d\n", -+ (int)border); -+ } else { -+ border = newext->ee_block; -+ ext_debug(tree, "leaf will be added." -+ " next leaf starts at %d\n", -+ (int)border); -+ } -+ -+ /* -+ * if error occurs, then we break processing -+ * and turn filesystem read-only. so, index won't -+ * be inserted and tree will be in consistent -+ * state. next mount will repair buffers too -+ */ -+ -+ /* -+ * get array to track all allocated blocks -+ * we need this to handle errors and free blocks -+ * upon them -+ */ -+ ablocks = kmalloc(sizeof(unsigned long) * depth, GFP_NOFS); -+ if (!ablocks) -+ return -ENOMEM; -+ memset(ablocks, 0, sizeof(unsigned long) * depth); -+ -+ /* allocate all needed blocks */ -+ ext_debug(tree, "allocate %d blocks for indexes/leaf\n", depth - at); -+ for (a = 0; a < depth - at; a++) { -+ newblock = ext3_ext_new_block(handle, tree, path, newext, &err); -+ if (newblock == 0) -+ goto cleanup; -+ ablocks[a] = newblock; -+ } -+ -+ /* initialize new leaf */ -+ newblock = ablocks[--a]; -+ EXT_ASSERT(newblock); -+ bh = sb_getblk(tree->inode->i_sb, newblock); -+ if (!bh) { -+ err = -EIO; -+ goto cleanup; -+ } -+ lock_buffer(bh); -+ -+ if ((err = ext3_journal_get_create_access(handle, bh))) -+ goto cleanup; -+ -+ neh = EXT_BLOCK_HDR(bh); -+ neh->eh_entries = 0; -+ neh->eh_max = ext3_ext_space_block(tree); -+ neh->eh_magic = EXT3_EXT_MAGIC; -+ neh->eh_depth = 0; -+ ex = EXT_FIRST_EXTENT(neh); -+ -+ /* move remain of path[depth] to the new leaf */ -+ EXT_ASSERT(path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max); -+ /* start copy from next extent */ -+ /* TODO: we could do it by single memmove */ -+ m = 0; -+ path[depth].p_ext++; -+ while (path[depth].p_ext <= -+ EXT_MAX_EXTENT(path[depth].p_hdr)) { -+ ext_debug(tree, "move %d:%d:%d in new leaf %lu\n", -+ path[depth].p_ext->ee_block, -+ path[depth].p_ext->ee_start, -+ path[depth].p_ext->ee_len, -+ newblock); -+ memmove(ex++, path[depth].p_ext++, sizeof(struct ext3_extent)); -+ neh->eh_entries++; -+ m++; -+ } -+ set_buffer_uptodate(bh); -+ unlock_buffer(bh); -+ -+ if ((err = ext3_journal_dirty_metadata(handle, bh))) -+ goto cleanup; -+ brelse(bh); -+ bh = NULL; -+ -+ /* correct old leaf */ -+ if (m) { -+ if ((err = ext3_ext_get_access(handle, tree, path + depth))) -+ goto cleanup; -+ path[depth].p_hdr->eh_entries -= m; -+ if ((err = ext3_ext_dirty(handle, tree, path + depth))) -+ goto cleanup; -+ -+ } -+ -+ /* create intermediate indexes */ -+ k = depth - at - 1; -+ EXT_ASSERT(k >= 0); -+ if (k) -+ ext_debug(tree, "create %d intermediate indices\n", k); -+ /* insert new index into current index block */ -+ /* current depth stored in i var */ -+ i = depth - 1; -+ while (k--) { -+ oldblock = newblock; -+ newblock = ablocks[--a]; -+ bh = sb_getblk(tree->inode->i_sb, newblock); -+ if (!bh) { -+ err = -EIO; -+ goto cleanup; -+ } -+ lock_buffer(bh); -+ -+ if ((err = ext3_journal_get_create_access(handle, bh))) -+ goto cleanup; -+ -+ neh = EXT_BLOCK_HDR(bh); -+ neh->eh_entries = 1; -+ neh->eh_magic = EXT3_EXT_MAGIC; -+ neh->eh_max = ext3_ext_space_block_idx(tree); -+ neh->eh_depth = depth - i; -+ fidx = EXT_FIRST_INDEX(neh); -+ fidx->ei_block = border; -+ fidx->ei_leaf = oldblock; -+ fidx->ei_leaf_hi = fidx->ei_unused = 0; -+ -+ ext_debug(tree, "int.index at %d (block %lu): %lu -> %lu\n", -+ i, newblock, border, oldblock); -+ /* copy indexes */ -+ m = 0; -+ path[i].p_idx++; -+ -+ ext_debug(tree, "cur 0x%p, last 0x%p\n", path[i].p_idx, -+ EXT_MAX_INDEX(path[i].p_hdr)); -+ EXT_ASSERT(EXT_MAX_INDEX(path[i].p_hdr) == -+ EXT_LAST_INDEX(path[i].p_hdr)); -+ while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { -+ ext_debug(tree, "%d: move %d:%d in new index %lu\n", -+ i, path[i].p_idx->ei_block, -+ path[i].p_idx->ei_leaf, newblock); -+ memmove(++fidx, path[i].p_idx++, -+ sizeof(struct ext3_extent_idx)); -+ neh->eh_entries++; -+ EXT_ASSERT(neh->eh_entries <= neh->eh_max); -+ m++; -+ } -+ set_buffer_uptodate(bh); -+ unlock_buffer(bh); -+ -+ if ((err = ext3_journal_dirty_metadata(handle, bh))) -+ goto cleanup; -+ brelse(bh); -+ bh = NULL; -+ -+ /* correct old index */ -+ if (m) { -+ err = ext3_ext_get_access(handle, tree, path + i); -+ if (err) -+ goto cleanup; -+ path[i].p_hdr->eh_entries -= m; -+ err = ext3_ext_dirty(handle, tree, path + i); -+ if (err) -+ goto cleanup; -+ } -+ -+ i--; -+ } -+ -+ /* insert new index */ -+ if (!err) -+ err = ext3_ext_insert_index(handle, tree, path + at, -+ border, newblock); -+ -+cleanup: -+ if (bh) { -+ if (buffer_locked(bh)) -+ unlock_buffer(bh); -+ brelse(bh); -+ } -+ -+ if (err) { -+ /* free all allocated blocks in error case */ -+ for (i = 0; i < depth; i++) { -+ if (!ablocks[i]) -+ continue; -+ ext3_free_blocks(handle, tree->inode, ablocks[i], 1); -+ } -+ } -+ kfree(ablocks); -+ -+ return err; -+} -+ -+/* -+ * routine implements tree growing procedure: -+ * - allocates new block -+ * - moves top-level data (index block or leaf) into the new block -+ * - initialize new top-level, creating index that points to the -+ * just created block -+ */ -+static int ext3_ext_grow_indepth(handle_t *handle, -+ struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_extent *newext) -+{ -+ struct ext3_ext_path *curp = path; -+ struct ext3_extent_header *neh; -+ struct ext3_extent_idx *fidx; -+ struct buffer_head *bh; -+ unsigned long newblock; -+ int err = 0; -+ -+ newblock = ext3_ext_new_block(handle, tree, path, newext, &err); -+ if (newblock == 0) -+ return err; -+ -+ bh = sb_getblk(tree->inode->i_sb, newblock); -+ if (!bh) { -+ err = -EIO; -+ ext3_std_error(tree->inode->i_sb, err); -+ return err; -+ } -+ lock_buffer(bh); -+ -+ if ((err = ext3_journal_get_create_access(handle, bh))) { -+ unlock_buffer(bh); -+ goto out; -+ } -+ -+ /* move top-level index/leaf into new block */ -+ memmove(bh->b_data, curp->p_hdr, tree->buffer_len); -+ -+ /* set size of new block */ -+ neh = EXT_BLOCK_HDR(bh); -+ /* old root could have indexes or leaves -+ * so calculate eh_max right way */ -+ if (EXT_DEPTH(tree)) -+ neh->eh_max = ext3_ext_space_block_idx(tree); -+ else -+ neh->eh_max = ext3_ext_space_block(tree); -+ neh->eh_magic = EXT3_EXT_MAGIC; -+ set_buffer_uptodate(bh); -+ unlock_buffer(bh); -+ -+ if ((err = ext3_journal_dirty_metadata(handle, bh))) -+ goto out; -+ -+ /* create index in new top-level index: num,max,pointer */ -+ if ((err = ext3_ext_get_access(handle, tree, curp))) -+ goto out; -+ -+ curp->p_hdr->eh_magic = EXT3_EXT_MAGIC; -+ curp->p_hdr->eh_max = ext3_ext_space_root_idx(tree); -+ curp->p_hdr->eh_entries = 1; -+ curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr); -+ /* FIXME: it works, but actually path[0] can be index */ -+ curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block; -+ curp->p_idx->ei_leaf = newblock; -+ curp->p_idx->ei_leaf_hi = curp->p_idx->ei_unused = 0; -+ -+ neh = EXT_ROOT_HDR(tree); -+ fidx = EXT_FIRST_INDEX(neh); -+ ext_debug(tree, "new root: num %d(%d), lblock %d, ptr %d\n", -+ neh->eh_entries, neh->eh_max, fidx->ei_block, fidx->ei_leaf); -+ -+ neh->eh_depth = path->p_depth + 1; -+ err = ext3_ext_dirty(handle, tree, curp); -+out: -+ brelse(bh); -+ -+ return err; -+} -+ -+/* -+ * routine finds empty index and adds new leaf. if no free index found -+ * then it requests in-depth growing -+ */ -+static int ext3_ext_create_new_leaf(handle_t *handle, -+ struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_extent *newext) -+{ -+ struct ext3_ext_path *curp; -+ int depth, i, err = 0; -+ -+repeat: -+ i = depth = EXT_DEPTH(tree); -+ -+ /* walk up to the tree and look for free index entry */ -+ curp = path + depth; -+ while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) { -+ i--; -+ curp--; -+ } -+ -+ /* we use already allocated block for index block -+ * so, subsequent data blocks should be contigoues */ -+ if (EXT_HAS_FREE_INDEX(curp)) { -+ /* if we found index with free entry, then use that -+ * entry: create all needed subtree and add new leaf */ -+ err = ext3_ext_split(handle, tree, path, newext, i); -+ -+ /* refill path */ -+ ext3_ext_drop_refs(path); -+ path = ext3_ext_find_extent(tree, newext->ee_block, path); -+ if (IS_ERR(path)) -+ err = PTR_ERR(path); -+ } else { -+ /* tree is full, time to grow in depth */ -+ err = ext3_ext_grow_indepth(handle, tree, path, newext); -+ -+ /* refill path */ -+ ext3_ext_drop_refs(path); -+ path = ext3_ext_find_extent(tree, newext->ee_block, path); -+ if (IS_ERR(path)) -+ err = PTR_ERR(path); -+ -+ /* -+ * only first (depth 0 -> 1) produces free space -+ * in all other cases we have to split growed tree -+ */ -+ depth = EXT_DEPTH(tree); -+ if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) { -+ /* now we need split */ -+ goto repeat; -+ } -+ } -+ -+ if (err) -+ return err; -+ -+ return 0; -+} -+ -+/* -+ * returns allocated block in subsequent extent or EXT_MAX_BLOCK -+ * NOTE: it consider block number from index entry as -+ * allocated block. thus, index entries have to be consistent -+ * with leafs -+ */ -+static unsigned long -+ext3_ext_next_allocated_block(struct ext3_ext_path *path) -+{ -+ int depth; -+ -+ EXT_ASSERT(path != NULL); -+ depth = path->p_depth; -+ -+ if (depth == 0 && path->p_ext == NULL) -+ return EXT_MAX_BLOCK; -+ -+ /* FIXME: what if index isn't full ?! */ -+ while (depth >= 0) { -+ if (depth == path->p_depth) { -+ /* leaf */ -+ if (path[depth].p_ext != -+ EXT_LAST_EXTENT(path[depth].p_hdr)) -+ return path[depth].p_ext[1].ee_block; -+ } else { -+ /* index */ -+ if (path[depth].p_idx != -+ EXT_LAST_INDEX(path[depth].p_hdr)) -+ return path[depth].p_idx[1].ei_block; -+ } -+ depth--; -+ } -+ -+ return EXT_MAX_BLOCK; -+} -+ -+/* -+ * returns first allocated block from next leaf or EXT_MAX_BLOCK -+ */ -+static unsigned ext3_ext_next_leaf_block(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ int depth; -+ -+ EXT_ASSERT(path != NULL); -+ depth = path->p_depth; -+ -+ /* zero-tree has no leaf blocks at all */ -+ if (depth == 0) -+ return EXT_MAX_BLOCK; -+ -+ /* go to index block */ -+ depth--; -+ -+ while (depth >= 0) { -+ if (path[depth].p_idx != -+ EXT_LAST_INDEX(path[depth].p_hdr)) -+ return path[depth].p_idx[1].ei_block; -+ depth--; -+ } -+ -+ return EXT_MAX_BLOCK; -+} -+ -+/* -+ * if leaf gets modified and modified extent is first in the leaf -+ * then we have to correct all indexes above -+ * TODO: do we need to correct tree in all cases? -+ */ -+int ext3_ext_correct_indexes(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ struct ext3_extent_header *eh; -+ int depth = EXT_DEPTH(tree); -+ struct ext3_extent *ex; -+ unsigned long border; -+ int k, err = 0; -+ -+ eh = path[depth].p_hdr; -+ ex = path[depth].p_ext; -+ EXT_ASSERT(ex); -+ EXT_ASSERT(eh); -+ -+ if (depth == 0) { -+ /* there is no tree at all */ -+ return 0; -+ } -+ -+ if (ex != EXT_FIRST_EXTENT(eh)) { -+ /* we correct tree if first leaf got modified only */ -+ return 0; -+ } -+ -+ /* -+ * TODO: we need correction if border is smaller then current one -+ */ -+ k = depth - 1; -+ border = path[depth].p_ext->ee_block; -+ if ((err = ext3_ext_get_access(handle, tree, path + k))) -+ return err; -+ path[k].p_idx->ei_block = border; -+ if ((err = ext3_ext_dirty(handle, tree, path + k))) -+ return err; -+ -+ while (k--) { -+ /* change all left-side indexes */ -+ if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr)) -+ break; -+ if ((err = ext3_ext_get_access(handle, tree, path + k))) -+ break; -+ path[k].p_idx->ei_block = border; -+ if ((err = ext3_ext_dirty(handle, tree, path + k))) -+ break; -+ } -+ -+ return err; -+} -+ -+static int inline -+ext3_can_extents_be_merged(struct ext3_extents_tree *tree, -+ struct ext3_extent *ex1, -+ struct ext3_extent *ex2) -+{ -+ if (ex1->ee_block + ex1->ee_len != ex2->ee_block) -+ return 0; -+ -+#ifdef AGRESSIVE_TEST -+ if (ex1->ee_len >= 4) -+ return 0; -+#endif -+ -+ if (!tree->ops->mergable) -+ return 1; -+ -+ return tree->ops->mergable(ex1, ex2); -+} -+ -+/* -+ * this routine tries to merge requsted extent into the existing -+ * extent or inserts requested extent as new one into the tree, -+ * creating new leaf in no-space case -+ */ -+int ext3_ext_insert_extent(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_extent *newext) -+{ -+ struct ext3_extent_header * eh; -+ struct ext3_extent *ex, *fex; -+ struct ext3_extent *nearex; /* nearest extent */ -+ struct ext3_ext_path *npath = NULL; -+ int depth, len, err, next; -+ -+ EXT_ASSERT(newext->ee_len > 0); -+ depth = EXT_DEPTH(tree); -+ ex = path[depth].p_ext; -+ EXT_ASSERT(path[depth].p_hdr); -+ -+ /* try to insert block into found extent and return */ -+ if (ex && ext3_can_extents_be_merged(tree, ex, newext)) { -+ ext_debug(tree, "append %d block to %d:%d (from %d)\n", -+ newext->ee_len, ex->ee_block, ex->ee_len, -+ ex->ee_start); -+ if ((err = ext3_ext_get_access(handle, tree, path + depth))) -+ return err; -+ ex->ee_len += newext->ee_len; -+ eh = path[depth].p_hdr; -+ nearex = ex; -+ goto merge; -+ } -+ -+repeat: -+ depth = EXT_DEPTH(tree); -+ eh = path[depth].p_hdr; -+ if (eh->eh_entries < eh->eh_max) -+ goto has_space; -+ -+ /* probably next leaf has space for us? */ -+ fex = EXT_LAST_EXTENT(eh); -+ next = ext3_ext_next_leaf_block(tree, path); -+ if (newext->ee_block > fex->ee_block && next != EXT_MAX_BLOCK) { -+ ext_debug(tree, "next leaf block - %d\n", next); -+ EXT_ASSERT(!npath); -+ npath = ext3_ext_find_extent(tree, next, NULL); -+ if (IS_ERR(npath)) -+ return PTR_ERR(npath); -+ EXT_ASSERT(npath->p_depth == path->p_depth); -+ eh = npath[depth].p_hdr; -+ if (eh->eh_entries < eh->eh_max) { -+ ext_debug(tree, "next leaf isnt full(%d)\n", -+ eh->eh_entries); -+ path = npath; -+ goto repeat; -+ } -+ ext_debug(tree, "next leaf hasno free space(%d,%d)\n", -+ eh->eh_entries, eh->eh_max); -+ } -+ -+ /* -+ * there is no free space in found leaf -+ * we're gonna add new leaf in the tree -+ */ -+ err = ext3_ext_create_new_leaf(handle, tree, path, newext); -+ if (err) -+ goto cleanup; -+ depth = EXT_DEPTH(tree); -+ eh = path[depth].p_hdr; -+ -+has_space: -+ nearex = path[depth].p_ext; -+ -+ if ((err = ext3_ext_get_access(handle, tree, path + depth))) -+ goto cleanup; -+ -+ if (!nearex) { -+ /* there is no extent in this leaf, create first one */ -+ ext_debug(tree, "first extent in the leaf: %d:%d:%d\n", -+ newext->ee_block, newext->ee_start, -+ newext->ee_len); -+ path[depth].p_ext = EXT_FIRST_EXTENT(eh); -+ } else if (newext->ee_block > nearex->ee_block) { -+ EXT_ASSERT(newext->ee_block != nearex->ee_block); -+ if (nearex != EXT_LAST_EXTENT(eh)) { -+ len = EXT_MAX_EXTENT(eh) - nearex; -+ len = (len - 1) * sizeof(struct ext3_extent); -+ len = len < 0 ? 0 : len; -+ ext_debug(tree, "insert %d:%d:%d after: nearest 0x%p, " -+ "move %d from 0x%p to 0x%p\n", -+ newext->ee_block, newext->ee_start, -+ newext->ee_len, -+ nearex, len, nearex + 1, nearex + 2); -+ memmove(nearex + 2, nearex + 1, len); -+ } -+ path[depth].p_ext = nearex + 1; -+ } else { -+ EXT_ASSERT(newext->ee_block != nearex->ee_block); -+ len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext3_extent); -+ len = len < 0 ? 0 : len; -+ ext_debug(tree, "insert %d:%d:%d before: nearest 0x%p, " -+ "move %d from 0x%p to 0x%p\n", -+ newext->ee_block, newext->ee_start, newext->ee_len, -+ nearex, len, nearex + 1, nearex + 2); -+ memmove(nearex + 1, nearex, len); -+ path[depth].p_ext = nearex; -+ } -+ -+ eh->eh_entries++; -+ nearex = path[depth].p_ext; -+ nearex->ee_block = newext->ee_block; -+ nearex->ee_start = newext->ee_start; -+ nearex->ee_len = newext->ee_len; -+ /* FIXME: support for large fs */ -+ nearex->ee_start_hi = 0; -+ -+merge: -+ /* try to merge extents to the right */ -+ while (nearex < EXT_LAST_EXTENT(eh)) { -+ if (!ext3_can_extents_be_merged(tree, nearex, nearex + 1)) -+ break; -+ /* merge with next extent! */ -+ nearex->ee_len += nearex[1].ee_len; -+ if (nearex + 1 < EXT_LAST_EXTENT(eh)) { -+ len = (EXT_LAST_EXTENT(eh) - nearex - 1) * -+ sizeof(struct ext3_extent); -+ memmove(nearex + 1, nearex + 2, len); -+ } -+ eh->eh_entries--; -+ EXT_ASSERT(eh->eh_entries > 0); -+ } -+ -+ /* try to merge extents to the left */ -+ -+ /* time to correct all indexes above */ -+ err = ext3_ext_correct_indexes(handle, tree, path); -+ if (err) -+ goto cleanup; -+ -+ err = ext3_ext_dirty(handle, tree, path + depth); -+ -+cleanup: -+ if (npath) { -+ ext3_ext_drop_refs(npath); -+ kfree(npath); -+ } -+ ext3_ext_tree_changed(tree); -+ ext3_ext_invalidate_cache(tree); -+ return err; -+} -+ -+int ext3_ext_walk_space(struct ext3_extents_tree *tree, unsigned long block, -+ unsigned long num, ext_prepare_callback func) -+{ -+ struct ext3_ext_path *path = NULL; -+ struct ext3_ext_cache cbex; -+ struct ext3_extent *ex; -+ unsigned long next, start = 0, end = 0; -+ unsigned long last = block + num; -+ int depth, exists, err = 0; -+ -+ EXT_ASSERT(tree); -+ EXT_ASSERT(func); -+ EXT_ASSERT(tree->inode); -+ EXT_ASSERT(tree->root); -+ -+ while (block < last && block != EXT_MAX_BLOCK) { -+ num = last - block; -+ /* find extent for this block */ -+ path = ext3_ext_find_extent(tree, block, path); -+ if (IS_ERR(path)) { -+ err = PTR_ERR(path); -+ path = NULL; -+ break; -+ } -+ -+ depth = EXT_DEPTH(tree); -+ EXT_ASSERT(path[depth].p_hdr); -+ ex = path[depth].p_ext; -+ next = ext3_ext_next_allocated_block(path); -+ -+ exists = 0; -+ if (!ex) { -+ /* there is no extent yet, so try to allocate -+ * all requested space */ -+ start = block; -+ end = block + num; -+ } else if (ex->ee_block > block) { -+ /* need to allocate space before found extent */ -+ start = block; -+ end = ex->ee_block; -+ if (block + num < end) -+ end = block + num; -+ } else if (block >= ex->ee_block + ex->ee_len) { -+ /* need to allocate space after found extent */ -+ start = block; -+ end = block + num; -+ if (end >= next) -+ end = next; -+ } else if (block >= ex->ee_block) { -+ /* -+ * some part of requested space is covered -+ * by found extent -+ */ -+ start = block; -+ end = ex->ee_block + ex->ee_len; -+ if (block + num < end) -+ end = block + num; -+ exists = 1; -+ } else { -+ BUG(); -+ } -+ EXT_ASSERT(end > start); -+ -+ if (!exists) { -+ cbex.ec_block = start; -+ cbex.ec_len = end - start; -+ cbex.ec_start = 0; -+ cbex.ec_type = EXT3_EXT_CACHE_GAP; -+ } else { -+ cbex.ec_block = ex->ee_block; -+ cbex.ec_len = ex->ee_len; -+ cbex.ec_start = ex->ee_start; -+ cbex.ec_type = EXT3_EXT_CACHE_EXTENT; -+ } -+ -+ EXT_ASSERT(cbex.ec_len > 0); -+ EXT_ASSERT(path[depth].p_hdr); -+ err = func(tree, path, &cbex); -+ ext3_ext_drop_refs(path); -+ -+ if (err < 0) -+ break; -+ if (err == EXT_REPEAT) -+ continue; -+ else if (err == EXT_BREAK) { -+ err = 0; -+ break; -+ } -+ -+ if (EXT_DEPTH(tree) != depth) { -+ /* depth was changed. we have to realloc path */ -+ kfree(path); -+ path = NULL; -+ } -+ -+ block = cbex.ec_block + cbex.ec_len; -+ } -+ -+ if (path) { -+ ext3_ext_drop_refs(path); -+ kfree(path); -+ } -+ -+ return err; -+} -+ -+static inline void -+ext3_ext_put_in_cache(struct ext3_extents_tree *tree, __u32 block, -+ __u32 len, __u32 start, int type) -+{ -+ EXT_ASSERT(len > 0); -+ if (tree->cex) { -+ tree->cex->ec_type = type; -+ tree->cex->ec_block = block; -+ tree->cex->ec_len = len; -+ tree->cex->ec_start = start; -+ } -+} -+ -+/* -+ * this routine calculate boundaries of the gap requested block fits into -+ * and cache this gap -+ */ -+static inline void -+ext3_ext_put_gap_in_cache(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ unsigned long block) -+{ -+ int depth = EXT_DEPTH(tree); -+ unsigned long lblock, len; -+ struct ext3_extent *ex; -+ -+ if (!tree->cex) -+ return; -+ -+ ex = path[depth].p_ext; -+ if (ex == NULL) { -+ /* there is no extent yet, so gap is [0;-] */ -+ lblock = 0; -+ len = EXT_MAX_BLOCK; -+ ext_debug(tree, "cache gap(whole file):"); -+ } else if (block < ex->ee_block) { -+ lblock = block; -+ len = ex->ee_block - block; -+ ext_debug(tree, "cache gap(before): %lu [%lu:%lu]", -+ (unsigned long) block, -+ (unsigned long) ex->ee_block, -+ (unsigned long) ex->ee_len); -+ } else if (block >= ex->ee_block + ex->ee_len) { -+ lblock = ex->ee_block + ex->ee_len; -+ len = ext3_ext_next_allocated_block(path); -+ ext_debug(tree, "cache gap(after): [%lu:%lu] %lu", -+ (unsigned long) ex->ee_block, -+ (unsigned long) ex->ee_len, -+ (unsigned long) block); -+ EXT_ASSERT(len > lblock); -+ len = len - lblock; -+ } else { -+ lblock = len = 0; -+ BUG(); -+ } -+ -+ ext_debug(tree, " -> %lu:%lu\n", (unsigned long) lblock, len); -+ ext3_ext_put_in_cache(tree, lblock, len, 0, EXT3_EXT_CACHE_GAP); -+} -+ -+static inline int -+ext3_ext_in_cache(struct ext3_extents_tree *tree, unsigned long block, -+ struct ext3_extent *ex) -+{ -+ struct ext3_ext_cache *cex = tree->cex; -+ -+ /* is there cache storage at all? */ -+ if (!cex) -+ return EXT3_EXT_CACHE_NO; -+ -+ /* has cache valid data? */ -+ if (cex->ec_type == EXT3_EXT_CACHE_NO) -+ return EXT3_EXT_CACHE_NO; -+ -+ EXT_ASSERT(cex->ec_type == EXT3_EXT_CACHE_GAP || -+ cex->ec_type == EXT3_EXT_CACHE_EXTENT); -+ if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) { -+ ex->ee_block = cex->ec_block; -+ ex->ee_start = cex->ec_start; -+ ex->ee_start_hi = 0; -+ ex->ee_len = cex->ec_len; -+ ext_debug(tree, "%lu cached by %lu:%lu:%lu\n", -+ (unsigned long) block, -+ (unsigned long) ex->ee_block, -+ (unsigned long) ex->ee_len, -+ (unsigned long) ex->ee_start); -+ return cex->ec_type; -+ } -+ -+ /* not in cache */ -+ return EXT3_EXT_CACHE_NO; -+} -+ -+/* -+ * routine removes index from the index block -+ * it's used in truncate case only. thus all requests are for -+ * last index in the block only -+ */ -+int ext3_ext_rm_idx(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ struct buffer_head *bh; -+ int err; -+ -+ /* free index block */ -+ path--; -+ EXT_ASSERT(path->p_hdr->eh_entries); -+ if ((err = ext3_ext_get_access(handle, tree, path))) -+ return err; -+ path->p_hdr->eh_entries--; -+ if ((err = ext3_ext_dirty(handle, tree, path))) -+ return err; -+ ext_debug(tree, "index is empty, remove it, free block %d\n", -+ path->p_idx->ei_leaf); -+ bh = sb_find_get_block(tree->inode->i_sb, path->p_idx->ei_leaf); -+ ext3_forget(handle, 1, tree->inode, bh, path->p_idx->ei_leaf); -+ ext3_free_blocks(handle, tree->inode, path->p_idx->ei_leaf, 1); -+ return err; -+} -+ -+int ext3_ext_calc_credits_for_insert(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path) -+{ -+ int depth = EXT_DEPTH(tree); -+ int needed; -+ -+ if (path) { -+ /* probably there is space in leaf? */ -+ if (path[depth].p_hdr->eh_entries < path[depth].p_hdr->eh_max) -+ return 1; -+ } -+ -+ /* -+ * the worste case we're expecting is creation of the -+ * new root (growing in depth) with index splitting -+ * for splitting we have to consider depth + 1 because -+ * previous growing could increase it -+ */ -+ depth = depth + 1; -+ -+ /* -+ * growing in depth: -+ * block allocation + new root + old root -+ */ -+ needed = EXT3_ALLOC_NEEDED + 2; -+ -+ /* index split. we may need: -+ * allocate intermediate indexes and new leaf -+ * change two blocks at each level, but root -+ * modify root block (inode) -+ */ -+ needed += (depth * EXT3_ALLOC_NEEDED) + (2 * depth) + 1; -+ -+ return needed; -+} -+ -+static int -+ext3_ext_split_for_rm(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, unsigned long start, -+ unsigned long end) -+{ -+ struct ext3_extent *ex, tex; -+ struct ext3_ext_path *npath; -+ int depth, creds, err; -+ -+ depth = EXT_DEPTH(tree); -+ ex = path[depth].p_ext; -+ EXT_ASSERT(ex); -+ EXT_ASSERT(end < ex->ee_block + ex->ee_len - 1); -+ EXT_ASSERT(ex->ee_block < start); -+ -+ /* calculate tail extent */ -+ tex.ee_block = end + 1; -+ EXT_ASSERT(tex.ee_block < ex->ee_block + ex->ee_len); -+ tex.ee_len = ex->ee_block + ex->ee_len - tex.ee_block; -+ -+ creds = ext3_ext_calc_credits_for_insert(tree, path); -+ handle = ext3_ext_journal_restart(handle, creds); -+ if (IS_ERR(handle)) -+ return PTR_ERR(handle); -+ -+ /* calculate head extent. use primary extent */ -+ err = ext3_ext_get_access(handle, tree, path + depth); -+ if (err) -+ return err; -+ ex->ee_len = start - ex->ee_block; -+ err = ext3_ext_dirty(handle, tree, path + depth); -+ if (err) -+ return err; -+ -+ /* FIXME: some callback to free underlying resource -+ * and correct ee_start? */ -+ ext_debug(tree, "split extent: head %u:%u, tail %u:%u\n", -+ ex->ee_block, ex->ee_len, tex.ee_block, tex.ee_len); -+ -+ npath = ext3_ext_find_extent(tree, ex->ee_block, NULL); -+ if (IS_ERR(npath)) -+ return PTR_ERR(npath); -+ depth = EXT_DEPTH(tree); -+ EXT_ASSERT(npath[depth].p_ext->ee_block == ex->ee_block); -+ EXT_ASSERT(npath[depth].p_ext->ee_len == ex->ee_len); -+ -+ err = ext3_ext_insert_extent(handle, tree, npath, &tex); -+ ext3_ext_drop_refs(npath); -+ kfree(npath); -+ -+ return err; -+} -+ -+static int -+ext3_ext_rm_leaf(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, unsigned long start, -+ unsigned long end) -+{ -+ struct ext3_extent *ex, *fu = NULL, *lu, *le; -+ int err = 0, correct_index = 0; -+ int depth = EXT_DEPTH(tree), credits; -+ struct ext3_extent_header *eh; -+ unsigned a, b, block, num; -+ -+ ext_debug(tree, "remove [%lu:%lu] in leaf\n", start, end); -+ if (!path[depth].p_hdr) -+ path[depth].p_hdr = EXT_BLOCK_HDR(path[depth].p_bh); -+ eh = path[depth].p_hdr; -+ EXT_ASSERT(eh); -+ EXT_ASSERT(eh->eh_entries <= eh->eh_max); -+ EXT_ASSERT(eh->eh_magic == EXT3_EXT_MAGIC); -+ -+ /* find where to start removing */ -+ le = ex = EXT_LAST_EXTENT(eh); -+ while (ex != EXT_FIRST_EXTENT(eh)) { -+ if (ex->ee_block <= end) -+ break; -+ ex--; -+ } -+ -+ if (start > ex->ee_block && end < ex->ee_block + ex->ee_len - 1) { -+ /* removal of internal part of the extent requested -+ * tail and head must be placed in different extent -+ * so, we have to insert one more extent */ -+ path[depth].p_ext = ex; -+ return ext3_ext_split_for_rm(handle, tree, path, start, end); -+ } -+ -+ lu = ex; -+ while (ex >= EXT_FIRST_EXTENT(eh) && ex->ee_block + ex->ee_len > start) { -+ ext_debug(tree, "remove ext %u:%u\n", ex->ee_block, ex->ee_len); -+ path[depth].p_ext = ex; -+ -+ a = ex->ee_block > start ? ex->ee_block : start; -+ b = ex->ee_block + ex->ee_len - 1 < end ? -+ ex->ee_block + ex->ee_len - 1 : end; -+ -+ ext_debug(tree, " border %u:%u\n", a, b); -+ -+ if (a != ex->ee_block && b != ex->ee_block + ex->ee_len - 1) { -+ block = 0; -+ num = 0; -+ BUG(); -+ } else if (a != ex->ee_block) { -+ /* remove tail of the extent */ -+ block = ex->ee_block; -+ num = a - block; -+ } else if (b != ex->ee_block + ex->ee_len - 1) { -+ /* remove head of the extent */ -+ block = a; -+ num = b - a; -+ } else { -+ /* remove whole extent: excelent! */ -+ block = ex->ee_block; -+ num = 0; -+ EXT_ASSERT(a == ex->ee_block && -+ b == ex->ee_block + ex->ee_len - 1); -+ } -+ -+ if (ex == EXT_FIRST_EXTENT(eh)) -+ correct_index = 1; -+ -+ credits = 1; -+ if (correct_index) -+ credits += (EXT_DEPTH(tree) * EXT3_ALLOC_NEEDED) + 1; -+ if (tree->ops->remove_extent_credits) -+ credits+=tree->ops->remove_extent_credits(tree,ex,a,b); -+ -+ handle = ext3_ext_journal_restart(handle, credits); -+ if (IS_ERR(handle)) { -+ err = PTR_ERR(handle); -+ goto out; -+ } -+ -+ err = ext3_ext_get_access(handle, tree, path + depth); -+ if (err) -+ goto out; -+ -+ if (tree->ops->remove_extent) -+ err = tree->ops->remove_extent(tree, ex, a, b); -+ if (err) -+ goto out; -+ -+ if (num == 0) { -+ /* this extent is removed entirely mark slot unused */ -+ ex->ee_start = ex->ee_start_hi = 0; -+ eh->eh_entries--; -+ fu = ex; -+ } -+ -+ ex->ee_block = block; -+ ex->ee_len = num; -+ -+ err = ext3_ext_dirty(handle, tree, path + depth); -+ if (err) -+ goto out; -+ -+ ext_debug(tree, "new extent: %u:%u:%u\n", -+ ex->ee_block, ex->ee_len, ex->ee_start); -+ ex--; -+ } -+ -+ if (fu) { -+ /* reuse unused slots */ -+ while (lu < le) { -+ if (lu->ee_start) { -+ *fu = *lu; -+ lu->ee_start = lu->ee_start_hi = 0; -+ fu++; -+ } -+ lu++; -+ } -+ } -+ -+ if (correct_index && eh->eh_entries) -+ err = ext3_ext_correct_indexes(handle, tree, path); -+ -+ /* if this leaf is free, then we should -+ * remove it from index block above */ -+ if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL) -+ err = ext3_ext_rm_idx(handle, tree, path + depth); -+ -+out: -+ return err; -+} -+ -+ -+static struct ext3_extent_idx * -+ext3_ext_last_covered(struct ext3_extent_header *hdr, unsigned long block) -+{ -+ struct ext3_extent_idx *ix; -+ -+ ix = EXT_LAST_INDEX(hdr); -+ while (ix != EXT_FIRST_INDEX(hdr)) { -+ if (ix->ei_block <= block) -+ break; -+ ix--; -+ } -+ return ix; -+} -+ -+/* -+ * returns 1 if current index have to be freed (even partial) -+ */ -+static int inline -+ext3_ext_more_to_rm(struct ext3_ext_path *path) -+{ -+ EXT_ASSERT(path->p_idx); -+ -+ if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr)) -+ return 0; -+ -+ /* -+ * if truncate on deeper level happened it it wasn't partial -+ * so we have to consider current index for truncation -+ */ -+ if (path->p_hdr->eh_entries == path->p_block) -+ return 0; -+ return 1; -+} -+ -+int ext3_ext_remove_space(struct ext3_extents_tree *tree, -+ unsigned long start, unsigned long end) -+{ -+ struct inode *inode = tree->inode; -+ struct super_block *sb = inode->i_sb; -+ int depth = EXT_DEPTH(tree); -+ struct ext3_ext_path *path; -+ handle_t *handle; -+ int i = 0, err = 0; -+ -+ ext_debug(tree, "space to be removed: %lu:%lu\n", start, end); -+ -+ /* probably first extent we're gonna free will be last in block */ -+ handle = ext3_journal_start(inode, depth + 1); -+ if (IS_ERR(handle)) -+ return PTR_ERR(handle); -+ -+ ext3_ext_invalidate_cache(tree); -+ -+ /* -+ * we start scanning from right side freeing all the blocks -+ * after i_size and walking into the deep -+ */ -+ path = kmalloc(sizeof(struct ext3_ext_path) * (depth + 1), GFP_KERNEL); -+ if (IS_ERR(path)) { -+ ext3_error(sb, __FUNCTION__, "Can't allocate path array"); -+ ext3_journal_stop(handle); -+ return -ENOMEM; -+ } -+ memset(path, 0, sizeof(struct ext3_ext_path) * (depth + 1)); -+ path[i].p_hdr = EXT_ROOT_HDR(tree); -+ -+ while (i >= 0 && err == 0) { -+ if (i == depth) { -+ /* this is leaf block */ -+ err = ext3_ext_rm_leaf(handle, tree, path, start, end); -+ /* root level have p_bh == NULL, brelse() eats this */ -+ brelse(path[i].p_bh); -+ i--; -+ continue; -+ } -+ -+ /* this is index block */ -+ if (!path[i].p_hdr) { -+ ext_debug(tree, "initialize header\n"); -+ path[i].p_hdr = EXT_BLOCK_HDR(path[i].p_bh); -+ } -+ -+ EXT_ASSERT(path[i].p_hdr->eh_entries <= path[i].p_hdr->eh_max); -+ EXT_ASSERT(path[i].p_hdr->eh_magic == EXT3_EXT_MAGIC); -+ -+ if (!path[i].p_idx) { -+ /* this level hasn't touched yet */ -+ path[i].p_idx = -+ ext3_ext_last_covered(path[i].p_hdr, end); -+ path[i].p_block = path[i].p_hdr->eh_entries + 1; -+ ext_debug(tree, "init index ptr: hdr 0x%p, num %d\n", -+ path[i].p_hdr, path[i].p_hdr->eh_entries); -+ } else { -+ /* we've already was here, see at next index */ -+ path[i].p_idx--; -+ } -+ -+ ext_debug(tree, "level %d - index, first 0x%p, cur 0x%p\n", -+ i, EXT_FIRST_INDEX(path[i].p_hdr), -+ path[i].p_idx); -+ if (ext3_ext_more_to_rm(path + i)) { -+ /* go to the next level */ -+ ext_debug(tree, "move to level %d (block %d)\n", -+ i + 1, path[i].p_idx->ei_leaf); -+ memset(path + i + 1, 0, sizeof(*path)); -+ path[i+1].p_bh = sb_bread(sb, path[i].p_idx->ei_leaf); -+ if (!path[i+1].p_bh) { -+ /* should we reset i_size? */ -+ err = -EIO; -+ break; -+ } -+ /* put actual number of indexes to know is this -+ * number got changed at the next iteration */ -+ path[i].p_block = path[i].p_hdr->eh_entries; -+ i++; -+ } else { -+ /* we finish processing this index, go up */ -+ if (path[i].p_hdr->eh_entries == 0 && i > 0) { -+ /* index is empty, remove it -+ * handle must be already prepared by the -+ * truncatei_leaf() */ -+ err = ext3_ext_rm_idx(handle, tree, path + i); -+ } -+ /* root level have p_bh == NULL, brelse() eats this */ -+ brelse(path[i].p_bh); -+ i--; -+ ext_debug(tree, "return to level %d\n", i); -+ } -+ } -+ -+ /* TODO: flexible tree reduction should be here */ -+ if (path->p_hdr->eh_entries == 0) { -+ /* -+ * truncate to zero freed all the tree -+ * so, we need to correct eh_depth -+ */ -+ err = ext3_ext_get_access(handle, tree, path); -+ if (err == 0) { -+ EXT_ROOT_HDR(tree)->eh_depth = 0; -+ EXT_ROOT_HDR(tree)->eh_max = ext3_ext_space_root(tree); -+ err = ext3_ext_dirty(handle, tree, path); -+ } -+ } -+ ext3_ext_tree_changed(tree); -+ -+ kfree(path); -+ ext3_journal_stop(handle); -+ -+ return err; -+} -+ -+int ext3_ext_calc_metadata_amount(struct ext3_extents_tree *tree, int blocks) -+{ -+ int lcap, icap, rcap, leafs, idxs, num; -+ -+ rcap = ext3_ext_space_root(tree); -+ if (blocks <= rcap) { -+ /* all extents fit to the root */ -+ return 0; -+ } -+ -+ rcap = ext3_ext_space_root_idx(tree); -+ lcap = ext3_ext_space_block(tree); -+ icap = ext3_ext_space_block_idx(tree); -+ -+ num = leafs = (blocks + lcap - 1) / lcap; -+ if (leafs <= rcap) { -+ /* all pointers to leafs fit to the root */ -+ return leafs; -+ } -+ -+ /* ok. we need separate index block(s) to link all leaf blocks */ -+ idxs = (leafs + icap - 1) / icap; -+ do { -+ num += idxs; -+ idxs = (idxs + icap - 1) / icap; -+ } while (idxs > rcap); -+ -+ return num; -+} -+ -+/* -+ * called at mount time -+ */ -+void ext3_ext_init(struct super_block *sb) -+{ -+ /* -+ * possible initialization would be here -+ */ -+ -+ if (test_opt(sb, EXTENTS)) { -+ printk("EXT3-fs: file extents enabled"); -+#ifdef AGRESSIVE_TEST -+ printk(", agressive tests"); -+#endif -+#ifdef CHECK_BINSEARCH -+ printk(", check binsearch"); -+#endif -+ printk("\n"); -+ } -+} -+ -+/* -+ * called at umount time -+ */ -+void ext3_ext_release(struct super_block *sb) -+{ -+} -+ -+/************************************************************************ -+ * VFS related routines -+ ************************************************************************/ -+ -+static int ext3_get_inode_write_access(handle_t *handle, void *buffer) -+{ -+ /* we use in-core data, not bh */ -+ return 0; -+} -+ -+static int ext3_mark_buffer_dirty(handle_t *handle, void *buffer) -+{ -+ struct inode *inode = buffer; -+ return ext3_mark_inode_dirty(handle, inode); -+} -+ -+static int ext3_ext_mergable(struct ext3_extent *ex1, -+ struct ext3_extent *ex2) -+{ -+ /* FIXME: support for large fs */ -+ if (ex1->ee_start + ex1->ee_len == ex2->ee_start) -+ return 1; -+ return 0; -+} -+ -+static int -+ext3_remove_blocks_credits(struct ext3_extents_tree *tree, -+ struct ext3_extent *ex, -+ unsigned long from, unsigned long to) -+{ -+ int needed; -+ -+ /* at present, extent can't cross block group */; -+ needed = 4; /* bitmap + group desc + sb + inode */ -+ -+#ifdef CONFIG_QUOTA -+ needed += 2 * EXT3_SINGLEDATA_TRANS_BLOCKS; -+#endif -+ return needed; -+} -+ -+static int -+ext3_remove_blocks(struct ext3_extents_tree *tree, -+ struct ext3_extent *ex, -+ unsigned long from, unsigned long to) -+{ -+ int needed = ext3_remove_blocks_credits(tree, ex, from, to); -+ handle_t *handle = ext3_journal_start(tree->inode, needed); -+ struct buffer_head *bh; -+ int i; -+ -+ if (IS_ERR(handle)) -+ return PTR_ERR(handle); -+ if (from >= ex->ee_block && to == ex->ee_block + ex->ee_len - 1) { -+ /* tail removal */ -+ unsigned long num, start; -+ num = ex->ee_block + ex->ee_len - from; -+ start = ex->ee_start + ex->ee_len - num; -+ ext_debug(tree, "free last %lu blocks starting %lu\n", -+ num, start); -+ for (i = 0; i < num; i++) { -+ bh = sb_find_get_block(tree->inode->i_sb, start + i); -+ ext3_forget(handle, 0, tree->inode, bh, start + i); -+ } -+ ext3_free_blocks(handle, tree->inode, start, num); -+ } else if (from == ex->ee_block && to <= ex->ee_block + ex->ee_len - 1) { -+ printk("strange request: removal %lu-%lu from %u:%u\n", -+ from, to, ex->ee_block, ex->ee_len); -+ } else { -+ printk("strange request: removal(2) %lu-%lu from %u:%u\n", -+ from, to, ex->ee_block, ex->ee_len); -+ } -+ ext3_journal_stop(handle); -+ return 0; -+} -+ -+static int ext3_ext_find_goal(struct inode *inode, -+ struct ext3_ext_path *path, unsigned long block) -+{ -+ struct ext3_inode_info *ei = EXT3_I(inode); -+ unsigned long bg_start; -+ unsigned long colour; -+ int depth; -+ -+ if (path) { -+ struct ext3_extent *ex; -+ depth = path->p_depth; -+ -+ /* try to predict block placement */ -+ if ((ex = path[depth].p_ext)) -+ return ex->ee_start + (block - ex->ee_block); -+ -+ /* it looks index is empty -+ * try to find starting from index itself */ -+ if (path[depth].p_bh) -+ return path[depth].p_bh->b_blocknr; -+ } -+ -+ /* OK. use inode's group */ -+ bg_start = (ei->i_block_group * EXT3_BLOCKS_PER_GROUP(inode->i_sb)) + -+ le32_to_cpu(EXT3_SB(inode->i_sb)->s_es->s_first_data_block); -+ colour = (current->pid % 16) * -+ (EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16); -+ return bg_start + colour + block; -+} -+ -+static int ext3_new_block_cb(handle_t *handle, struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_extent *ex, int *err) -+{ -+ struct inode *inode = tree->inode; -+ int newblock, goal; -+ -+ EXT_ASSERT(path); -+ EXT_ASSERT(ex); -+ EXT_ASSERT(ex->ee_start); -+ EXT_ASSERT(ex->ee_len); -+ -+ /* reuse block from the extent to order data/metadata */ -+ newblock = ex->ee_start++; -+ ex->ee_len--; -+ if (ex->ee_len == 0) { -+ ex->ee_len = 1; -+ /* allocate new block for the extent */ -+ goal = ext3_ext_find_goal(inode, path, ex->ee_block); -+ ex->ee_start = ext3_new_block(handle, inode, goal, err); -+ ex->ee_start_hi = 0; -+ if (ex->ee_start == 0) { -+ /* error occured: restore old extent */ -+ ex->ee_start = newblock; -+ return 0; -+ } -+ } -+ return newblock; -+} -+ -+static struct ext3_extents_helpers ext3_blockmap_helpers = { -+ .get_write_access = ext3_get_inode_write_access, -+ .mark_buffer_dirty = ext3_mark_buffer_dirty, -+ .mergable = ext3_ext_mergable, -+ .new_block = ext3_new_block_cb, -+ .remove_extent = ext3_remove_blocks, -+ .remove_extent_credits = ext3_remove_blocks_credits, -+}; -+ -+void ext3_init_tree_desc(struct ext3_extents_tree *tree, -+ struct inode *inode) -+{ -+ tree->inode = inode; -+ tree->root = (void *) EXT3_I(inode)->i_data; -+ tree->buffer = (void *) inode; -+ tree->buffer_len = sizeof(EXT3_I(inode)->i_data); -+ tree->cex = &EXT3_I(inode)->i_cached_extent; -+ tree->ops = &ext3_blockmap_helpers; -+} -+ -+int ext3_ext_get_block(handle_t *handle, struct inode *inode, -+ long iblock, struct buffer_head *bh_result, -+ int create, int extend_disksize) -+{ -+ struct ext3_ext_path *path = NULL; -+ struct ext3_extent newex; -+ struct ext3_extent *ex; -+ int goal, newblock, err = 0, depth; -+ struct ext3_extents_tree tree; -+ -+ __clear_bit(BH_New, &bh_result->b_state); -+ ext3_init_tree_desc(&tree, inode); -+ ext_debug(&tree, "block %d requested for inode %u\n", -+ (int) iblock, (unsigned) inode->i_ino); -+ down(&EXT3_I(inode)->truncate_sem); -+ -+ /* check in cache */ -+ if ((goal = ext3_ext_in_cache(&tree, iblock, &newex))) { -+ if (goal == EXT3_EXT_CACHE_GAP) { -+ if (!create) { -+ /* block isn't allocated yet and -+ * user don't want to allocate it */ -+ goto out2; -+ } -+ /* we should allocate requested block */ -+ } else if (goal == EXT3_EXT_CACHE_EXTENT) { -+ /* block is already allocated */ -+ newblock = iblock - newex.ee_block + newex.ee_start; -+ goto out; -+ } else { -+ EXT_ASSERT(0); -+ } -+ } -+ -+ /* find extent for this block */ -+ path = ext3_ext_find_extent(&tree, iblock, NULL); -+ if (IS_ERR(path)) { -+ err = PTR_ERR(path); -+ path = NULL; -+ goto out2; -+ } -+ -+ depth = EXT_DEPTH(&tree); -+ -+ /* -+ * consistent leaf must not be empty -+ * this situations is possible, though, _during_ tree modification -+ * this is why assert can't be put in ext3_ext_find_extent() -+ */ -+ EXT_ASSERT(path[depth].p_ext != NULL || depth == 0); -+ -+ if ((ex = path[depth].p_ext)) { -+ /* if found exent covers block, simple return it */ -+ if (iblock >= ex->ee_block && iblock < ex->ee_block + ex->ee_len) { -+ newblock = iblock - ex->ee_block + ex->ee_start; -+ ext_debug(&tree, "%d fit into %d:%d -> %d\n", -+ (int) iblock, ex->ee_block, ex->ee_len, -+ newblock); -+ ext3_ext_put_in_cache(&tree, ex->ee_block, -+ ex->ee_len, ex->ee_start, -+ EXT3_EXT_CACHE_EXTENT); -+ goto out; -+ } -+ } -+ -+ /* -+ * requested block isn't allocated yet -+ * we couldn't try to create block if create flag is zero -+ */ -+ if (!create) { -+ /* put just found gap into cache to speedup subsequest reqs */ -+ ext3_ext_put_gap_in_cache(&tree, path, iblock); -+ goto out2; -+ } -+ -+ /* allocate new block */ -+ goal = ext3_ext_find_goal(inode, path, iblock); -+ newblock = ext3_new_block(handle, inode, goal, &err); -+ if (!newblock) -+ goto out2; -+ ext_debug(&tree, "allocate new block: goal %d, found %d\n", -+ goal, newblock); -+ -+ /* try to insert new extent into found leaf and return */ -+ newex.ee_block = iblock; -+ newex.ee_start = newblock; -+ newex.ee_start_hi = 0; -+ newex.ee_len = 1; -+ err = ext3_ext_insert_extent(handle, &tree, path, &newex); -+ if (err) -+ goto out2; -+ -+ if (extend_disksize && inode->i_size > EXT3_I(inode)->i_disksize) -+ EXT3_I(inode)->i_disksize = inode->i_size; -+ -+ /* previous routine could use block we allocated */ -+ newblock = newex.ee_start; -+ __set_bit(BH_New, &bh_result->b_state); -+ -+ ext3_ext_put_in_cache(&tree, newex.ee_block, newex.ee_len, -+ newex.ee_start, EXT3_EXT_CACHE_EXTENT); -+out: -+ ext3_ext_show_leaf(&tree, path); -+ __set_bit(BH_Mapped, &bh_result->b_state); -+ bh_result->b_bdev = inode->i_sb->s_bdev; -+ bh_result->b_blocknr = newblock; -+out2: -+ if (path) { -+ ext3_ext_drop_refs(path); -+ kfree(path); -+ } -+ up(&EXT3_I(inode)->truncate_sem); -+ -+ return err; -+} -+ -+void ext3_ext_truncate(struct inode * inode, struct page *page) -+{ -+ struct address_space *mapping = inode->i_mapping; -+ struct super_block *sb = inode->i_sb; -+ struct ext3_extents_tree tree; -+ unsigned long last_block; -+ handle_t *handle; -+ int err = 0; -+ -+ ext3_init_tree_desc(&tree, inode); -+ -+ /* -+ * probably first extent we're gonna free will be last in block -+ */ -+ err = ext3_writepage_trans_blocks(inode) + 3; -+ handle = ext3_journal_start(inode, err); -+ if (IS_ERR(handle)) { -+ if (page) { -+ clear_highpage(page); -+ flush_dcache_page(page); -+ unlock_page(page); -+ page_cache_release(page); -+ } -+ return; -+ } -+ -+ if (page) -+ ext3_block_truncate_page(handle, page, mapping, inode->i_size); -+ -+ down(&EXT3_I(inode)->truncate_sem); -+ ext3_ext_invalidate_cache(&tree); -+ -+ /* -+ * TODO: optimization is possible here -+ * probably we need not scaning at all, -+ * because page truncation is enough -+ */ -+ if (ext3_orphan_add(handle, inode)) -+ goto out_stop; -+ -+ /* we have to know where to truncate from in crash case */ -+ EXT3_I(inode)->i_disksize = inode->i_size; -+ ext3_mark_inode_dirty(handle, inode); -+ -+ last_block = (inode->i_size + sb->s_blocksize - 1) >> -+ EXT3_BLOCK_SIZE_BITS(sb); -+ err = ext3_ext_remove_space(&tree, last_block, EXT_MAX_BLOCK); -+ -+ /* In a multi-transaction truncate, we only make the final -+ * transaction synchronous */ -+ if (IS_SYNC(inode)) -+ handle->h_sync = 1; -+ -+out_stop: -+ /* -+ * If this was a simple ftruncate(), and the file will remain alive -+ * then we need to clear up the orphan record which we created above. -+ * However, if this was a real unlink then we were called by -+ * ext3_delete_inode(), and we allow that function to clean up the -+ * orphan info for us. -+ */ -+ if (inode->i_nlink) -+ ext3_orphan_del(handle, inode); -+ -+ up(&EXT3_I(inode)->truncate_sem); -+ ext3_journal_stop(handle); -+} -+ -+/* -+ * this routine calculate max number of blocks we could modify -+ * in order to allocate new block for an inode -+ */ -+int ext3_ext_writepage_trans_blocks(struct inode *inode, int num) -+{ -+ struct ext3_extents_tree tree; -+ int needed; -+ -+ ext3_init_tree_desc(&tree, inode); -+ -+ needed = ext3_ext_calc_credits_for_insert(&tree, NULL); -+ -+ /* caller want to allocate num blocks */ -+ needed *= num; -+ -+#ifdef CONFIG_QUOTA -+ /* -+ * FIXME: real calculation should be here -+ * it depends on blockmap format of qouta file -+ */ -+ needed += 2 * EXT3_SINGLEDATA_TRANS_BLOCKS; -+#endif -+ -+ return needed; -+} -+ -+void ext3_extents_initialize_blockmap(handle_t *handle, struct inode *inode) -+{ -+ struct ext3_extents_tree tree; -+ -+ ext3_init_tree_desc(&tree, inode); -+ ext3_extent_tree_init(handle, &tree); -+} -+ -+int ext3_ext_calc_blockmap_metadata(struct inode *inode, int blocks) -+{ -+ struct ext3_extents_tree tree; -+ -+ ext3_init_tree_desc(&tree, inode); -+ return ext3_ext_calc_metadata_amount(&tree, blocks); -+} -+ -+static int -+ext3_ext_store_extent_cb(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_ext_cache *newex) -+{ -+ struct ext3_extent_buf *buf = (struct ext3_extent_buf *) tree->private; -+ -+ if (newex->ec_type != EXT3_EXT_CACHE_EXTENT) -+ return EXT_CONTINUE; -+ -+ if (buf->err < 0) -+ return EXT_BREAK; -+ if (buf->cur - buf->buffer + sizeof(*newex) > buf->buflen) -+ return EXT_BREAK; -+ -+ if (!copy_to_user(buf->cur, newex, sizeof(*newex))) { -+ buf->err++; -+ buf->cur += sizeof(*newex); -+ } else { -+ buf->err = -EFAULT; -+ return EXT_BREAK; -+ } -+ return EXT_CONTINUE; -+} -+ -+static int -+ext3_ext_collect_stats_cb(struct ext3_extents_tree *tree, -+ struct ext3_ext_path *path, -+ struct ext3_ext_cache *ex) -+{ -+ struct ext3_extent_tree_stats *buf = -+ (struct ext3_extent_tree_stats *) tree->private; -+ int depth; -+ -+ if (ex->ec_type != EXT3_EXT_CACHE_EXTENT) -+ return EXT_CONTINUE; -+ -+ depth = EXT_DEPTH(tree); -+ buf->extents_num++; -+ if (path[depth].p_ext == EXT_FIRST_EXTENT(path[depth].p_hdr)) -+ buf->leaf_num++; -+ return EXT_CONTINUE; -+} -+ -+int ext3_ext_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, -+ unsigned long arg) -+{ -+ int err = 0; -+ -+ if (!(EXT3_I(inode)->i_flags & EXT3_EXTENTS_FL)) -+ return -EINVAL; -+ -+ if (cmd == EXT3_IOC_GET_EXTENTS) { -+ struct ext3_extent_buf buf; -+ struct ext3_extents_tree tree; -+ -+ if (copy_from_user(&buf, (void *) arg, sizeof(buf))) -+ return -EFAULT; -+ -+ ext3_init_tree_desc(&tree, inode); -+ buf.cur = buf.buffer; -+ buf.err = 0; -+ tree.private = &buf; -+ down(&EXT3_I(inode)->truncate_sem); -+ err = ext3_ext_walk_space(&tree, buf.start, EXT_MAX_BLOCK, -+ ext3_ext_store_extent_cb); -+ up(&EXT3_I(inode)->truncate_sem); -+ if (err == 0) -+ err = buf.err; -+ } else if (cmd == EXT3_IOC_GET_TREE_STATS) { -+ struct ext3_extent_tree_stats buf; -+ struct ext3_extents_tree tree; -+ -+ ext3_init_tree_desc(&tree, inode); -+ down(&EXT3_I(inode)->truncate_sem); -+ buf.depth = EXT_DEPTH(&tree); -+ buf.extents_num = 0; -+ buf.leaf_num = 0; -+ tree.private = &buf; -+ err = ext3_ext_walk_space(&tree, 0, EXT_MAX_BLOCK, -+ ext3_ext_collect_stats_cb); -+ up(&EXT3_I(inode)->truncate_sem); -+ if (!err) -+ err = copy_to_user((void *) arg, &buf, sizeof(buf)); -+ } else if (cmd == EXT3_IOC_GET_TREE_DEPTH) { -+ struct ext3_extents_tree tree; -+ ext3_init_tree_desc(&tree, inode); -+ down(&EXT3_I(inode)->truncate_sem); -+ err = EXT_DEPTH(&tree); -+ up(&EXT3_I(inode)->truncate_sem); -+ } -+ -+ return err; -+} -+ -+EXPORT_SYMBOL(ext3_init_tree_desc); -+EXPORT_SYMBOL(ext3_mark_inode_dirty); -+EXPORT_SYMBOL(ext3_ext_invalidate_cache); -+EXPORT_SYMBOL(ext3_ext_insert_extent); -+EXPORT_SYMBOL(ext3_ext_walk_space); -+EXPORT_SYMBOL(ext3_ext_find_goal); -+EXPORT_SYMBOL(ext3_ext_calc_credits_for_insert); -Index: linux-2.6.5-sles9/fs/ext3/ialloc.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/ialloc.c 2005-02-23 01:01:52.366281264 +0300 -+++ linux-2.6.5-sles9/fs/ext3/ialloc.c 2005-02-23 01:02:37.398435336 +0300 -@@ -566,7 +566,7 @@ repeat: - ei->i_dir_start_lookup = 0; - ei->i_disksize = 0; - -- ei->i_flags = EXT3_I(dir)->i_flags & ~EXT3_INDEX_FL; -+ ei->i_flags = EXT3_I(dir)->i_flags & ~(EXT3_INDEX_FL|EXT3_EXTENTS_FL); - if (S_ISLNK(mode)) - ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL); - /* dirsync only applies to directories */ -@@ -647,6 +647,18 @@ - DQUOT_FREE_INODE(inode); - goto fail2; - } -+ if (test_opt(sb, EXTENTS) && S_ISREG(inode->i_mode)) { -+ EXT3_I(inode)->i_flags |= EXT3_EXTENTS_FL; -+ ext3_extents_initialize_blockmap(handle, inode); -+ if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_EXTENTS)) { -+ err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh); -+ if (err) goto fail; -+ EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_EXTENTS); -+ BUFFER_TRACE(EXT3_SB(sb)->s_sbh, "call ext3_journal_dirty_metadata"); -+ err = ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); -+ } -+ } -+ - err = ext3_mark_inode_dirty(handle, inode); - if (err) { - ext3_std_error(sb, err); -Index: linux-2.6.5-sles9/fs/ext3/inode.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/inode.c 2005-02-23 01:01:52.373280200 +0300 -+++ linux-2.6.5-sles9/fs/ext3/inode.c 2005-02-23 01:02:37.404434424 +0300 -@@ -796,6 +796,17 @@ - goto reread; - } - -+static inline int -+ext3_get_block_wrap(handle_t *handle, struct inode *inode, long block, -+ struct buffer_head *bh, int create, int extend_disksize) -+{ -+ if (EXT3_I(inode)->i_flags & EXT3_EXTENTS_FL) -+ return ext3_ext_get_block(handle, inode, block, bh, create, -+ extend_disksize); -+ return ext3_get_block_handle(handle, inode, block, bh, create, -+ extend_disksize); -+} -+ - static int ext3_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) - { -@@ -806,8 +817,8 @@ - handle = ext3_journal_current_handle(); - J_ASSERT(handle != 0); - } -- ret = ext3_get_block_handle(handle, inode, iblock, -- bh_result, create, 1); -+ ret = ext3_get_block_wrap(handle, inode, iblock, -+ bh_result, create, 1); - return ret; - } - -@@ -833,8 +844,8 @@ - } - } - if (ret == 0) -- ret = ext3_get_block_handle(handle, inode, iblock, -- bh_result, create, 0); -+ ret = ext3_get_block_wrap(handle, inode, iblock, -+ bh_result, create, 0); - if (ret == 0) - bh_result->b_size = (1 << inode->i_blkbits); - return ret; -@@ -855,7 +866,7 @@ - dummy.b_state = 0; - dummy.b_blocknr = -1000; - buffer_trace_init(&dummy.b_history); -- *errp = ext3_get_block_handle(handle, inode, block, &dummy, create, 1); -+ *errp = ext3_get_block_wrap(handle, inode, block, &dummy, create, 1); - if (!*errp && buffer_mapped(&dummy)) { - struct buffer_head *bh; - bh = sb_getblk(inode->i_sb, dummy.b_blocknr); -@@ -1587,7 +1598,7 @@ - * This required during truncate. We need to physically zero the tail end - * of that block so it doesn't yield old data if the file is later grown. - */ --static int ext3_block_truncate_page(handle_t *handle, struct page *page, -+int ext3_block_truncate_page(handle_t *handle, struct page *page, - struct address_space *mapping, loff_t from) - { - unsigned long index = from >> PAGE_CACHE_SHIFT; -@@ -2083,6 +2094,9 @@ - return; - } - -+ if (EXT3_I(inode)->i_flags & EXT3_EXTENTS_FL) -+ return ext3_ext_truncate(inode, page); -+ - handle = start_transaction(inode); - if (IS_ERR(handle)) { - if (page) { -@@ -2789,6 +2803,9 @@ - int indirects = (EXT3_NDIR_BLOCKS % bpp) ? 5 : 3; - int ret; - -+ if (EXT3_I(inode)->i_flags & EXT3_EXTENTS_FL) -+ return ext3_ext_writepage_trans_blocks(inode, bpp); -+ - if (ext3_should_journal_data(inode)) - ret = 3 * (bpp + indirects) + 2; - else -Index: linux-2.6.5-sles9/fs/ext3/Makefile -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/Makefile 2005-02-23 01:01:46.501172896 +0300 -+++ linux-2.6.5-sles9/fs/ext3/Makefile 2005-02-23 01:02:37.405434272 +0300 -@@ -5,7 +5,8 @@ - obj-$(CONFIG_EXT3_FS) += ext3.o - - ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o iopen.o \ -- ioctl.o namei.o super.o symlink.o hash.o -+ ioctl.o namei.o super.o symlink.o hash.o \ -+ extents.o - - ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o - ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o -Index: linux-2.6.5-sles9/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/super.c 2005-02-23 01:02:34.072940888 +0300 -+++ linux-2.6.5-sles9/fs/ext3/super.c 2005-02-23 01:47:15.291333736 +0300 -@@ -389,6 +389,7 @@ - struct ext3_super_block *es = sbi->s_es; - int i; - -+ ext3_ext_release(sb); - ext3_xattr_put_super(sb); - journal_destroy(sbi->s_journal); - if (!(sb->s_flags & MS_RDONLY)) { -@@ -447,6 +448,8 @@ - #endif - ei->i_rsv_window.rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - ei->vfs_inode.i_version = 1; -+ -+ memset(&ei->i_cached_extent, 0, sizeof(ei->i_cached_extent)); - return &ei->vfs_inode; - } - -@@ -537,6 +540,7 @@ - Opt_ignore, Opt_barrier, - Opt_err, - Opt_iopen, Opt_noiopen, Opt_iopen_nopriv, -+ Opt_extents, Opt_noextents, Opt_extdebug, - }; - - static match_table_t tokens = { -@@ -582,6 +585,9 @@ - {Opt_iopen, "iopen"}, - {Opt_noiopen, "noiopen"}, - {Opt_iopen_nopriv, "iopen_nopriv"}, -+ {Opt_extents, "extents"}, -+ {Opt_noextents, "noextents"}, -+ {Opt_extdebug, "extdebug"}, - {Opt_barrier, "barrier=%u"}, - {Opt_err, NULL} - }; -@@ -797,6 +802,15 @@ - break; - case Opt_ignore: - break; -+ case Opt_extents: -+ set_opt (sbi->s_mount_opt, EXTENTS); -+ break; -+ case Opt_noextents: -+ clear_opt (sbi->s_mount_opt, EXTENTS); -+ break; -+ case Opt_extdebug: -+ set_opt (sbi->s_mount_opt, EXTDEBUG); -+ break; - default: - printk (KERN_ERR - "EXT3-fs: Unrecognized mount option \"%s\" " -@@ -1449,6 +1460,8 @@ - percpu_counter_mod(&sbi->s_dirs_counter, - ext3_count_dirs(sb)); - -+ ext3_ext_init(sb); -+ - return 0; - - failed_mount3: -Index: linux-2.6.5-sles9/fs/ext3/ioctl.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/ioctl.c 2005-02-23 01:01:42.887722224 +0300 -+++ linux-2.6.5-sles9/fs/ext3/ioctl.c 2005-02-23 01:02:37.412433208 +0300 -@@ -124,6 +124,10 @@ - err = ext3_change_inode_journal_flag(inode, jflag); - return err; - } -+ case EXT3_IOC_GET_EXTENTS: -+ case EXT3_IOC_GET_TREE_STATS: -+ case EXT3_IOC_GET_TREE_DEPTH: -+ return ext3_ext_ioctl(inode, filp, cmd, arg); - case EXT3_IOC_GETVERSION: - case EXT3_IOC_GETVERSION_OLD: - return put_user(inode->i_generation, (int *) arg); -Index: linux-2.6.5-sles9/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_fs.h 2005-02-23 01:02:35.823674736 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_fs.h 2005-02-23 01:02:37.414432904 +0300 -@@ -186,8 +186,9 @@ - #define EXT3_NOTAIL_FL 0x00008000 /* don't merge file tail */ - #define EXT3_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ - #define EXT3_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ -+#define EXT3_EXTENTS_FL 0x00080000 /* Inode uses extents */ - #define EXT3_RESERVED_FL 0x80000000 /* reserved for ext3 lib */ - --#define EXT3_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */ -+#define EXT3_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ - #define EXT3_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ - -@@ -211,6 +212,9 @@ - #endif - #define EXT3_IOC_GETRSVSZ _IOR('f', 5, long) - #define EXT3_IOC_SETRSVSZ _IOW('f', 6, long) -+#define EXT3_IOC_GET_EXTENTS _IOR('f', 7, long) -+#define EXT3_IOC_GET_TREE_DEPTH _IOR('f', 8, long) -+#define EXT3_IOC_GET_TREE_STATS _IOR('f', 9, long) - - /* - * Structure of an inode on the disk -@@ -333,6 +337,8 @@ - #define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ - #define EXT3_MOUNT_IOPEN 0x80000 /* Allow access via iopen */ - #define EXT3_MOUNT_IOPEN_NOPRIV 0x100000/* Make iopen world-readable */ -+#define EXT3_MOUNT_EXTENTS 0x200000/* Extents support */ -+#define EXT3_MOUNT_EXTDEBUG 0x400000/* Extents debug */ - - /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ - #ifndef clear_opt -@@ -503,11 +509,13 @@ - #define EXT3_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ - #define EXT3_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ - #define EXT3_FEATURE_INCOMPAT_META_BG 0x0010 -+#define EXT3_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */ - - #define EXT3_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR - #define EXT3_FEATURE_INCOMPAT_SUPP (EXT3_FEATURE_INCOMPAT_FILETYPE| \ - EXT3_FEATURE_INCOMPAT_RECOVER| \ -- EXT3_FEATURE_INCOMPAT_META_BG) -+ EXT3_FEATURE_INCOMPAT_META_BG| \ -+ EXT3_FEATURE_INCOMPAT_EXTENTS) - #define EXT3_FEATURE_RO_COMPAT_SUPP (EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER| \ - EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \ - EXT3_FEATURE_RO_COMPAT_BTREE_DIR) -@@ -729,6 +735,9 @@ - - - /* inode.c */ -+extern int ext3_block_truncate_page(handle_t *, struct page *, -+ struct address_space *, loff_t); -+extern int ext3_writepage_trans_blocks(struct inode *inode); - extern int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int); - extern struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *); - extern struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *); -@@ -802,6 +809,16 @@ - extern struct inode_operations ext3_symlink_inode_operations; - extern struct inode_operations ext3_fast_symlink_inode_operations; - -+/* extents.c */ -+extern int ext3_ext_writepage_trans_blocks(struct inode *, int); -+extern int ext3_ext_get_block(handle_t *, struct inode *, long, -+ struct buffer_head *, int, int); -+extern void ext3_ext_truncate(struct inode *, struct page *); -+extern void ext3_ext_init(struct super_block *); -+extern void ext3_ext_release(struct super_block *); -+extern void ext3_extents_initialize_blockmap(handle_t *, struct inode *); -+extern int ext3_ext_ioctl(struct inode *inode, struct file *filp, -+ unsigned int cmd, unsigned long arg); - - #endif /* __KERNEL__ */ - -Index: linux-2.6.5-sles9/include/linux/ext3_extents.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_extents.h 2005-02-17 22:07:57.023609040 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_extents.h 2005-02-23 01:02:37.416432600 +0300 -@@ -0,0 +1,262 @@ -+/* -+ * Copyright (c) 2003, Cluster File Systems, Inc, info@clusterfs.com -+ * Written by Alex Tomas -+ * -+ * This program is free software; you can redistribute it and/or modify -+ * it under the terms of the GNU General Public License version 2 as -+ * published by the Free Software Foundation. -+ * -+ * This program is distributed in the hope that it will be useful, -+ * but WITHOUT ANY WARRANTY; without even the implied warranty of -+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ * GNU General Public License for more details. -+ * -+ * You should have received a copy of the GNU General Public Licens -+ * along with this program; if not, write to the Free Software -+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- -+ */ -+ -+#ifndef _LINUX_EXT3_EXTENTS -+#define _LINUX_EXT3_EXTENTS -+ -+/* -+ * with AGRESSIVE_TEST defined capacity of index/leaf blocks -+ * become very little, so index split, in-depth growing and -+ * other hard changes happens much more often -+ * this is for debug purposes only -+ */ -+#define AGRESSIVE_TEST_ -+ -+/* -+ * if CHECK_BINSEARCH defined, then results of binary search -+ * will be checked by linear search -+ */ -+#define CHECK_BINSEARCH_ -+ -+/* -+ * if EXT_DEBUG is defined you can use 'extdebug' mount option -+ * to get lots of info what's going on -+ */ -+#define EXT_DEBUG_ -+#ifdef EXT_DEBUG -+#define ext_debug(tree,fmt,a...) \ -+do { \ -+ if (test_opt((tree)->inode->i_sb, EXTDEBUG)) \ -+ printk(fmt, ##a); \ -+} while (0); -+#else -+#define ext_debug(tree,fmt,a...) -+#endif -+ -+/* -+ * if EXT_STATS is defined then stats numbers are collected -+ * these number will be displayed at umount time -+ */ -+#define EXT_STATS_ -+ -+ -+#define EXT3_ALLOC_NEEDED 3 /* block bitmap + group desc. + sb */ -+ -+/* -+ * ext3_inode has i_block array (total 60 bytes) -+ * first 4 bytes are used to store: -+ * - tree depth (0 mean there is no tree yet. all extents in the inode) -+ * - number of alive extents in the inode -+ */ -+ -+/* -+ * this is extent on-disk structure -+ * it's used at the bottom of the tree -+ */ -+struct ext3_extent { -+ __u32 ee_block; /* first logical block extent covers */ -+ __u16 ee_len; /* number of blocks covered by extent */ -+ __u16 ee_start_hi; /* high 16 bits of physical block */ -+ __u32 ee_start; /* low 32 bigs of physical block */ -+}; -+ -+/* -+ * this is index on-disk structure -+ * it's used at all the levels, but the bottom -+ */ -+struct ext3_extent_idx { -+ __u32 ei_block; /* index covers logical blocks from 'block' */ -+ __u32 ei_leaf; /* pointer to the physical block of the next * -+ * level. leaf or next index could bet here */ -+ __u16 ei_leaf_hi; /* high 16 bits of physical block */ -+ __u16 ei_unused; -+}; -+ -+/* -+ * each block (leaves and indexes), even inode-stored has header -+ */ -+struct ext3_extent_header { -+ __u16 eh_magic; /* probably will support different formats */ -+ __u16 eh_entries; /* number of valid entries */ -+ __u16 eh_max; /* capacity of store in entries */ -+ __u16 eh_depth; /* has tree real underlaying blocks? */ -+ __u32 eh_generation; /* flags(8 bits) | generation of the tree */ -+}; -+ -+#define EXT3_EXT_MAGIC 0xf30a -+ -+/* -+ * array of ext3_ext_path contains path to some extent -+ * creation/lookup routines use it for traversal/splitting/etc -+ * truncate uses it to simulate recursive walking -+ */ -+struct ext3_ext_path { -+ __u32 p_block; -+ __u16 p_depth; -+ struct ext3_extent *p_ext; -+ struct ext3_extent_idx *p_idx; -+ struct ext3_extent_header *p_hdr; -+ struct buffer_head *p_bh; -+}; -+ -+/* -+ * structure for external API -+ */ -+ -+/* -+ * storage for cached extent -+ */ -+struct ext3_ext_cache { -+ __u32 ec_start; -+ __u32 ec_block; -+ __u32 ec_len; -+ __u32 ec_type; -+}; -+ -+#define EXT3_EXT_CACHE_NO 0 -+#define EXT3_EXT_CACHE_GAP 1 -+#define EXT3_EXT_CACHE_EXTENT 2 -+ -+/* -+ * ext3_extents_tree is used to pass initial information -+ * to top-level extents API -+ */ -+struct ext3_extents_helpers; -+struct ext3_extents_tree { -+ struct inode *inode; /* inode which tree belongs to */ -+ void *root; /* ptr to data top of tree resides at */ -+ void *buffer; /* will be passed as arg to ^^ routines */ -+ int buffer_len; -+ void *private; -+ struct ext3_ext_cache *cex;/* last found extent */ -+ struct ext3_extents_helpers *ops; -+}; -+ -+struct ext3_extents_helpers { -+ int (*get_write_access)(handle_t *h, void *buffer); -+ int (*mark_buffer_dirty)(handle_t *h, void *buffer); -+ int (*mergable)(struct ext3_extent *ex1, struct ext3_extent *ex2); -+ int (*remove_extent_credits)(struct ext3_extents_tree *, -+ struct ext3_extent *, unsigned long, -+ unsigned long); -+ int (*remove_extent)(struct ext3_extents_tree *, -+ struct ext3_extent *, unsigned long, -+ unsigned long); -+ int (*new_block)(handle_t *, struct ext3_extents_tree *, -+ struct ext3_ext_path *, struct ext3_extent *, -+ int *); -+}; -+ -+/* -+ * to be called by ext3_ext_walk_space() -+ * negative retcode - error -+ * positive retcode - signal for ext3_ext_walk_space(), see below -+ * callback must return valid extent (passed or newly created) -+ */ -+typedef int (*ext_prepare_callback)(struct ext3_extents_tree *, -+ struct ext3_ext_path *, -+ struct ext3_ext_cache *); -+ -+#define EXT_CONTINUE 0 -+#define EXT_BREAK 1 -+#define EXT_REPEAT 2 -+ -+ -+#define EXT_MAX_BLOCK 0xffffffff -+ -+ -+#define EXT_FIRST_EXTENT(__hdr__) \ -+ ((struct ext3_extent *) (((char *) (__hdr__)) + \ -+ sizeof(struct ext3_extent_header))) -+#define EXT_FIRST_INDEX(__hdr__) \ -+ ((struct ext3_extent_idx *) (((char *) (__hdr__)) + \ -+ sizeof(struct ext3_extent_header))) -+#define EXT_HAS_FREE_INDEX(__path__) \ -+ ((__path__)->p_hdr->eh_entries < (__path__)->p_hdr->eh_max) -+#define EXT_LAST_EXTENT(__hdr__) \ -+ (EXT_FIRST_EXTENT((__hdr__)) + (__hdr__)->eh_entries - 1) -+#define EXT_LAST_INDEX(__hdr__) \ -+ (EXT_FIRST_INDEX((__hdr__)) + (__hdr__)->eh_entries - 1) -+#define EXT_MAX_EXTENT(__hdr__) \ -+ (EXT_FIRST_EXTENT((__hdr__)) + (__hdr__)->eh_max - 1) -+#define EXT_MAX_INDEX(__hdr__) \ -+ (EXT_FIRST_INDEX((__hdr__)) + (__hdr__)->eh_max - 1) -+#define EXT_HDR_GEN(__hdr__) ((__hdr__)->eh_generation & 0x00ffffff) -+#define EXT_FLAGS(__hdr__) ((__hdr__)->eh_generation >> 24) -+#define EXT_FLAGS_CLR_UNKNOWN 0x7 /* Flags cleared on modification */ -+ -+#define EXT_BLOCK_HDR(__bh__) ((struct ext3_extent_header *)(__bh__)->b_data) -+#define EXT_ROOT_HDR(__tree__) ((struct ext3_extent_header *)(__tree__)->root) -+#define EXT_DEPTH(__tree__) (EXT_ROOT_HDR(__tree__)->eh_depth) -+#define EXT_GENERATION(__tree__) EXT_HDR_GEN(EXT_ROOT_HDR(__tree__)) -+ -+#define EXT_ASSERT(__x__) if (!(__x__)) BUG(); -+ -+#define EXT_CHECK_PATH(tree,path) \ -+{ \ -+ int depth = EXT_DEPTH(tree); \ -+ BUG_ON((unsigned long) (path) < __PAGE_OFFSET); \ -+ BUG_ON((unsigned long) (path)[depth].p_idx < \ -+ __PAGE_OFFSET && (path)[depth].p_idx != NULL); \ -+ BUG_ON((unsigned long) (path)[depth].p_ext < \ -+ __PAGE_OFFSET && (path)[depth].p_ext != NULL); \ -+ BUG_ON((unsigned long) (path)[depth].p_hdr < __PAGE_OFFSET); \ -+ BUG_ON((unsigned long) (path)[depth].p_bh < __PAGE_OFFSET \ -+ && depth != 0); \ -+ BUG_ON((path)[0].p_depth != depth); \ -+} -+ -+ -+/* -+ * this structure is used to gather extents from the tree via ioctl -+ */ -+struct ext3_extent_buf { -+ unsigned long start; -+ int buflen; -+ void *buffer; -+ void *cur; -+ int err; -+}; -+ -+/* -+ * this structure is used to collect stats info about the tree -+ */ -+struct ext3_extent_tree_stats { -+ int depth; -+ int extents_num; -+ int leaf_num; -+}; -+ -+extern void ext3_init_tree_desc(struct ext3_extents_tree *, struct inode *); -+extern int ext3_extent_tree_init(handle_t *, struct ext3_extents_tree *); -+extern int ext3_ext_calc_credits_for_insert(struct ext3_extents_tree *, struct ext3_ext_path *); -+extern int ext3_ext_insert_extent(handle_t *, struct ext3_extents_tree *, struct ext3_ext_path *, struct ext3_extent *); -+extern int ext3_ext_walk_space(struct ext3_extents_tree *, unsigned long, unsigned long, ext_prepare_callback); -+extern int ext3_ext_remove_space(struct ext3_extents_tree *, unsigned long, unsigned long); -+extern struct ext3_ext_path * ext3_ext_find_extent(struct ext3_extents_tree *, int, struct ext3_ext_path *); -+extern int ext3_ext_calc_blockmap_metadata(struct inode *, int); -+ -+static inline void -+ext3_ext_invalidate_cache(struct ext3_extents_tree *tree) -+{ -+ if (tree->cex) -+ tree->cex->ec_type = EXT3_EXT_CACHE_NO; -+} -+ -+ -+#endif /* _LINUX_EXT3_EXTENTS */ -Index: linux-2.6.5-sles9/include/linux/ext3_fs_i.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_fs_i.h 2005-02-23 01:01:52.425272296 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_fs_i.h 2005-02-23 01:45:55.611446920 +0300 -@@ -19,6 +19,7 @@ - #include - #include - #include -+#include - - struct reserve_window { - __u32 _rsv_start; /* First byte reserved */ -@@ -128,6 +129,8 @@ - */ - struct semaphore truncate_sem; - struct inode vfs_inode; -+ -+ struct ext3_ext_cache i_cached_extent; - }; - - #endif /* _LINUX_EXT3_FS_I */ - -%diffstat - fs/ext3/Makefile | 2 - fs/ext3/extents.c | 2356 +++++++++++++++++++++++++++++++++++++++++++ - fs/ext3/ialloc.c | 4 - fs/ext3/inode.c | 29 - fs/ext3/ioctl.c | 4 - fs/ext3/super.c | 15 - include/linux/ext3_extents.h | 265 ++++ - include/linux/ext3_fs.h | 17 - include/linux/ext3_fs_i.h | 3 - 9 files changed, 2687 insertions(+), 8 deletions(-) - diff --git a/ldiskfs/kernel_patches/patches/ext3-extents-multiblock-directio-2.6.5-suse.patch b/ldiskfs/kernel_patches/patches/ext3-extents-multiblock-directio-2.6.5-suse.patch deleted file mode 100644 index 744cc45..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-extents-multiblock-directio-2.6.5-suse.patch +++ /dev/null @@ -1,157 +0,0 @@ -Index: linux-2.6.5-7.283-full/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/include/linux/ext3_fs.h 2007-03-28 02:13:37.000000000 +0400 -+++ linux-2.6.5-7.283-full/include/linux/ext3_fs.h 2007-03-28 02:21:37.000000000 +0400 -@@ -815,7 +815,7 @@ extern struct inode_operations ext3_fast - - /* extents.c */ - extern int ext3_ext_writepage_trans_blocks(struct inode *, int); --extern int ext3_ext_get_block(handle_t *, struct inode *, long, -+extern int ext3_ext_get_block(handle_t *, struct inode *, long, int, - struct buffer_head *, int, int); - extern void ext3_ext_truncate(struct inode *, struct page *); - extern void ext3_ext_init(struct super_block *); -Index: linux-2.6.5-7.283-full/fs/ext3/extents.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/extents.c 2007-03-28 02:14:25.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/extents.c 2007-03-28 02:21:37.000000000 +0400 -@@ -2024,7 +2024,8 @@ void ext3_init_tree_desc(struct ext3_ext - } - - int ext3_ext_get_block(handle_t *handle, struct inode *inode, -- long iblock, struct buffer_head *bh_result, -+ long iblock, int max_blocks, -+ struct buffer_head *bh_result, - int create, int extend_disksize) - { - struct ext3_ext_path *path = NULL; -@@ -2032,6 +2033,11 @@ int ext3_ext_get_block(handle_t *handle, - struct ext3_extent *ex; - int goal, newblock, err = 0, depth; - struct ext3_extents_tree tree; -+ unsigned long next; -+ int allocated = 0; -+ -+ /* until we have multiblock allocation */ -+ max_blocks = 1; - - __clear_bit(BH_New, &bh_result->b_state); - ext3_init_tree_desc(&tree, inode); -@@ -2051,6 +2057,9 @@ int ext3_ext_get_block(handle_t *handle, - } else if (goal == EXT3_EXT_CACHE_EXTENT) { - /* block is already allocated */ - newblock = iblock - newex.ee_block + newex.ee_start; -+ /* number of remaining blocks in the extent */ -+ EXT_ASSERT(iblock >= newex.ee_block); -+ allocated = newex.ee_len - (iblock - newex.ee_block); - goto out; - } else { - EXT_ASSERT(0); -@@ -2078,6 +2087,8 @@ int ext3_ext_get_block(handle_t *handle, - /* if found exent covers block, simple return it */ - if (iblock >= ex->ee_block && iblock < ex->ee_block + ex->ee_len) { - newblock = iblock - ex->ee_block + ex->ee_start; -+ /* number of remaining blocks in the extent */ -+ allocated = ex->ee_len - (iblock - ex->ee_block); - ext_debug(&tree, "%d fit into %d:%d -> %d\n", - (int) iblock, ex->ee_block, ex->ee_len, - newblock); -@@ -2098,6 +2109,15 @@ int ext3_ext_get_block(handle_t *handle, - goto out2; - } - -+ /* find next allocated block so that we know how many -+ * blocks we can allocate without ovelapping next extent */ -+ EXT_ASSERT(iblock >= ex->ee_block + ex->ee_len); -+ next = ext3_ext_next_allocated_block(path); -+ EXT_ASSERT(next > iblock); -+ allocated = next - iblock; -+ if (allocated > max_blocks) -+ allocated = max_blocks; -+ - /* allocate new block */ - goal = ext3_ext_find_goal(inode, path, iblock); - newblock = ext3_new_block(handle, inode, goal, &err); -@@ -2112,8 +2132,11 @@ int ext3_ext_get_block(handle_t *handle, - newex.ee_start_hi = 0; - newex.ee_len = 1; - err = ext3_ext_insert_extent(handle, &tree, path, &newex); -- if (err) -+ if (err) { -+ /* free data blocks we just allocated */ -+ ext3_free_blocks(handle, inode, newex.ee_start, newex.ee_len); - goto out2; -+ } - - if (extend_disksize && inode->i_size > EXT3_I(inode)->i_disksize) - EXT3_I(inode)->i_disksize = inode->i_size; -@@ -2125,10 +2148,13 @@ int ext3_ext_get_block(handle_t *handle, - ext3_ext_put_in_cache(&tree, newex.ee_block, newex.ee_len, - newex.ee_start, EXT3_EXT_CACHE_EXTENT); - out: -+ if (allocated > max_blocks) -+ allocated = max_blocks; - ext3_ext_show_leaf(&tree, path); - __set_bit(BH_Mapped, &bh_result->b_state); - bh_result->b_bdev = inode->i_sb->s_bdev; - bh_result->b_blocknr = newblock; -+ bh_result->b_size = (allocated << inode->i_blkbits); - out2: - if (path) { - ext3_ext_drop_refs(path); -Index: linux-2.6.5-7.283-full/fs/ext3/inode.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/inode.c 2007-03-28 02:13:37.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/inode.c 2007-03-28 02:50:19.000000000 +0400 -@@ -800,13 +800,17 @@ changed: - - static inline int - ext3_get_block_wrap(handle_t *handle, struct inode *inode, long block, -- struct buffer_head *bh, int create, int extend_disksize) -+ int max_blocks, struct buffer_head *bh, int create, -+ int extend_disksize) - { -+ int ret; - if (EXT3_I(inode)->i_flags & EXT3_EXTENTS_FL) -- return ext3_ext_get_block(handle, inode, block, bh, create, -- extend_disksize); -- return ext3_get_block_handle(handle, inode, block, bh, create, -+ return ext3_ext_get_block(handle, inode, block, max_blocks, -+ bh, create, extend_disksize); -+ ret = ext3_get_block_handle(handle, inode, block, bh, create, - extend_disksize); -+ bh->b_size = (1 << inode->i_blkbits); -+ return ret; - } - - static int ext3_get_block(struct inode *inode, sector_t iblock, -@@ -819,7 +823,7 @@ static int ext3_get_block(struct inode * - handle = ext3_journal_current_handle(); - J_ASSERT(handle != 0); - } -- ret = ext3_get_block_wrap(handle, inode, iblock, -+ ret = ext3_get_block_wrap(handle, inode, iblock, 1, - bh_result, create, 1); - return ret; - } -@@ -847,10 +851,8 @@ ext3_direct_io_get_blocks(struct inode * - } - } - if (ret == 0) -- ret = ext3_get_block_wrap(handle, inode, iblock, -+ ret = ext3_get_block_wrap(handle, inode, iblock, max_blocks, - bh_result, create, 0); -- if (ret == 0) -- bh_result->b_size = (1 << inode->i_blkbits); - return ret; - } - -@@ -869,7 +871,7 @@ struct buffer_head *ext3_getblk(handle_t - dummy.b_state = 0; - dummy.b_blocknr = -1000; - buffer_trace_init(&dummy.b_history); -- *errp = ext3_get_block_wrap(handle, inode, block, &dummy, create, 1); -+ *errp = ext3_get_block_wrap(handle, inode, block, 1, &dummy, create, 1); - if (!*errp && buffer_mapped(&dummy)) { - struct buffer_head *bh; - bh = sb_getblk(inode->i_sb, dummy.b_blocknr); diff --git a/ldiskfs/kernel_patches/patches/ext3-include-fixes-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-include-fixes-2.6-suse.patch deleted file mode 100644 index 1ac944b..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-include-fixes-2.6-suse.patch +++ /dev/null @@ -1,20 +0,0 @@ -Index: linux-stage/include/linux/ext3_fs.h -=================================================================== ---- linux-stage.orig/include/linux/ext3_fs.h 2004-04-02 16:43:37.000000000 -0500 -+++ linux-stage/include/linux/ext3_fs.h 2004-04-02 16:43:37.000000000 -0500 -@@ -331,12 +331,13 @@ - #define EXT3_MOUNT_IOPEN_NOPRIV 0x100000/* Make iopen world-readable */ - - /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ --#ifndef _LINUX_EXT2_FS_H -+#ifndef clear_opt - #define clear_opt(o, opt) o &= ~EXT3_MOUNT_##opt - #define set_opt(o, opt) o |= EXT3_MOUNT_##opt - #define test_opt(sb, opt) (EXT3_SB(sb)->s_mount_opt & \ - EXT3_MOUNT_##opt) --#else -+#endif -+#ifndef EXT2_MOUNT_NOLOAD - #define EXT2_MOUNT_NOLOAD EXT3_MOUNT_NOLOAD - #define EXT2_MOUNT_ABORT EXT3_MOUNT_ABORT - #define EXT2_MOUNT_DATA_FLAGS EXT3_MOUNT_DATA_FLAGS diff --git a/ldiskfs/kernel_patches/patches/ext3-mballoc3-suse.patch b/ldiskfs/kernel_patches/patches/ext3-mballoc3-suse.patch deleted file mode 100644 index dd07148..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-mballoc3-suse.patch +++ /dev/null @@ -1,397 +0,0 @@ -Index: linux-2.6.5-7.283-full/include/linux/ext3_fs_i.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/include/linux/ext3_fs_i.h 2007-03-28 02:13:37.000000000 +0400 -+++ linux-2.6.5-7.283-full/include/linux/ext3_fs_i.h 2007-03-28 15:46:02.000000000 +0400 -@@ -131,6 +131,10 @@ struct ext3_inode_info { - struct inode vfs_inode; - - struct ext3_ext_cache i_cached_extent; -+ -+ /* mballoc */ -+ struct list_head i_prealloc_list; -+ spinlock_t i_prealloc_lock; - }; - - #endif /* _LINUX_EXT3_FS_I */ -Index: linux-2.6.5-7.283-full/include/linux/ext3_fs_sb.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/include/linux/ext3_fs_sb.h 2007-03-28 15:46:00.000000000 +0400 -+++ linux-2.6.5-7.283-full/include/linux/ext3_fs_sb.h 2007-03-28 15:46:02.000000000 +0400 -@@ -23,9 +23,16 @@ - #define EXT_INCLUDE - #include - #include -+#include - #endif - #endif - #include -+#include -+ -+struct ext3_buddy_group_blocks; -+struct ext3_locality_group; -+struct ext3_mb_history; -+#define EXT3_BB_MAX_BLOCKS - - /* - * third extended-fs super-block data in memory -Index: linux-2.6.5-7.283-full/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/include/linux/ext3_fs.h 2007-03-28 15:46:00.000000000 +0400 -+++ linux-2.6.5-7.283-full/include/linux/ext3_fs.h 2007-03-28 15:46:02.000000000 +0400 -@@ -363,6 +363,7 @@ struct ext3_inode { - #define EXT3_MOUNT_IOPEN_NOPRIV 0x100000/* Make iopen world-readable */ - #define EXT3_MOUNT_EXTENTS 0x200000/* Extents support */ - #define EXT3_MOUNT_EXTDEBUG 0x400000/* Extents debug */ -+#define EXT3_MOUNT_MBALLOC 0x800000/* Buddy allocation support */ - - /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ - #ifndef clear_opt -@@ -723,8 +724,9 @@ struct dir_private_info { - extern int ext3_bg_has_super(struct super_block *sb, int group); - extern unsigned long ext3_bg_num_gdb(struct super_block *sb, int group); - extern int ext3_new_block (handle_t *, struct inode *, unsigned long, int *); -+extern int ext3_new_block_old (handle_t *, struct inode *, unsigned long, int *); - extern void ext3_free_blocks (handle_t *, struct inode *, unsigned long, -- unsigned long); -+ unsigned long, int); - extern unsigned long ext3_count_free_blocks (struct super_block *); - extern void ext3_check_blocks_bitmap (struct super_block *); - extern struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, -Index: linux-2.6.5-7.283-full/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/super.c 2007-03-28 15:46:00.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/super.c 2007-03-28 15:46:02.000000000 +0400 -@@ -622,6 +622,7 @@ enum { - Opt_err, - Opt_iopen, Opt_noiopen, Opt_iopen_nopriv, - Opt_extents, Opt_noextents, Opt_extdebug, -+ Opt_mballoc, Opt_nomballoc, Opt_stripe, - }; - - static match_table_t tokens = { -@@ -669,6 +670,9 @@ static match_table_t tokens = { - {Opt_noextents, "noextents"}, - {Opt_extdebug, "extdebug"}, - {Opt_barrier, "barrier=%u"}, -+ {Opt_mballoc, "mballoc"}, -+ {Opt_nomballoc, "nomballoc"}, -+ {Opt_stripe, "stripe=%u"}, - {Opt_err, NULL} - }; - -@@ -893,6 +897,19 @@ static int parse_options (char * options - case Opt_extdebug: - set_opt (sbi->s_mount_opt, EXTDEBUG); - break; -+ case Opt_mballoc: -+ set_opt(sbi->s_mount_opt, MBALLOC); -+ break; -+ case Opt_nomballoc: -+ clear_opt(sbi->s_mount_opt, MBALLOC); -+ break; -+ case Opt_stripe: -+ if (match_int(&args[0], &option)) -+ return 0; -+ if (option < 0) -+ return 0; -+ sbi->s_stripe = option; -+ break; - default: - printk (KERN_ERR - "EXT3-fs: Unrecognized mount option \"%s\" " -@@ -1548,6 +1565,7 @@ static int ext3_fill_super (struct super - ext3_count_dirs(sb)); - - ext3_ext_init(sb); -+ ext3_mb_init(sb, needs_recovery); - - return 0; - -Index: linux-2.6.5-7.283-full/fs/ext3/extents.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/extents.c 2007-03-28 03:18:19.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/extents.c 2007-03-28 15:46:02.000000000 +0400 -@@ -779,7 +779,7 @@ cleanup: - for (i = 0; i < depth; i++) { - if (!ablocks[i]) - continue; -- ext3_free_blocks(handle, tree->inode, ablocks[i], 1); -+ ext3_free_blocks(handle, tree->inode, ablocks[i], 1, 1); - } - } - kfree(ablocks); -@@ -1586,7 +1586,7 @@ int ext3_ext_rm_idx(handle_t *handle, st - path->p_idx->ei_leaf); - bh = sb_find_get_block(tree->inode->i_sb, path->p_idx->ei_leaf); - ext3_forget(handle, 1, tree->inode, bh, path->p_idx->ei_leaf); -- ext3_free_blocks(handle, tree->inode, path->p_idx->ei_leaf, 1); -+ ext3_free_blocks(handle, tree->inode, path->p_idx->ei_leaf, 1, 1); - return err; - } - -@@ -2071,10 +2071,12 @@ ext3_remove_blocks(struct ext3_extents_t - int needed = ext3_remove_blocks_credits(tree, ex, from, to); - handle_t *handle = ext3_journal_start(tree->inode, needed); - struct buffer_head *bh; -- int i; -+ int i, metadata = 0; - - if (IS_ERR(handle)) - return PTR_ERR(handle); -+ if (S_ISDIR(tree->inode->i_mode) || S_ISLNK(tree->inode->i_mode)) -+ metadata = 1; - if (from >= ex->ee_block && to == ex->ee_block + ex->ee_len - 1) { - /* tail removal */ - unsigned long num, start; -@@ -2086,7 +2088,7 @@ ext3_remove_blocks(struct ext3_extents_t - bh = sb_find_get_block(tree->inode->i_sb, start + i); - ext3_forget(handle, 0, tree->inode, bh, start + i); - } -- ext3_free_blocks(handle, tree->inode, start, num); -+ ext3_free_blocks(handle, tree->inode, start, num, metadata); - } else if (from == ex->ee_block && to <= ex->ee_block + ex->ee_len - 1) { - printk("strange request: removal %lu-%lu from %u:%u\n", - from, to, ex->ee_block, ex->ee_len); -@@ -2177,11 +2179,8 @@ int ext3_ext_get_block(handle_t *handle, - struct ext3_extent *ex; - int goal, newblock, err = 0, depth; - struct ext3_extents_tree tree; -- unsigned long next; -- int allocated = 0; -- -- /* until we have multiblock allocation */ -- max_blocks = 1; -+ unsigned long allocated = 0; -+ struct ext3_allocation_request ar; - - __clear_bit(BH_New, &bh_result->b_state); - ext3_init_tree_desc(&tree, inode); -@@ -2253,18 +2252,33 @@ int ext3_ext_get_block(handle_t *handle, - goto out2; - } - -+ /* find neighbour allocated blocks */ -+ ar.lleft = iblock; -+ err = ext3_ext_search_left(&tree, path, &ar.lleft, &ar.pleft); -+ if (err) -+ goto out2; -+ ar.lright = iblock; -+ err = ext3_ext_search_right(&tree, path, &ar.lright, &ar.pright); -+ if (err) -+ goto out2; -+ - /* find next allocated block so that we know how many - * blocks we can allocate without ovelapping next extent */ -- EXT_ASSERT(iblock >= ex->ee_block + ex->ee_len); -- next = ext3_ext_next_allocated_block(path); -- EXT_ASSERT(next > iblock); -- allocated = next - iblock; -+ EXT_ASSERT(ar.pright == 0 || ar.lright > iblock); -+ if (ar.pright == 0) -+ allocated = EXT_MAX_BLOCK - iblock; -+ else -+ allocated = ar.lright - iblock; - if (allocated > max_blocks) - allocated = max_blocks; - - /* allocate new block */ -- goal = ext3_ext_find_goal(inode, path, iblock); -- newblock = ext3_new_block(handle, inode, goal, &err); -+ ar.inode = inode; -+ ar.goal = ext3_ext_find_goal(inode, path, iblock); -+ ar.logical = iblock; -+ ar.len = allocated; -+ ar.flags = EXT3_MB_HINT_DATA; -+ newblock = ext3_mb_new_blocks(handle, &ar, &err); - if (!newblock) - goto out2; - ext_debug(&tree, "allocate new block: goal %d, found %d\n", -@@ -2274,11 +2288,14 @@ int ext3_ext_get_block(handle_t *handle, - newex.ee_block = iblock; - newex.ee_start = newblock; - newex.ee_start_hi = 0; -- newex.ee_len = 1; -+ newex.ee_len = ar.len; - err = ext3_ext_insert_extent(handle, &tree, path, &newex); - if (err) { - /* free data blocks we just allocated */ -- ext3_free_blocks(handle, inode, newex.ee_start, newex.ee_len); -+ /* not a good idea to call discard here directly, -+ * but otherwise we'd need to call it every free() */ -+ ext3_mb_discard_inode_preallocations(inode); -+ ext3_free_blocks(handle, inode, newex.ee_start, newex.ee_len, 0); - goto out2; - } - -@@ -2287,6 +2304,7 @@ int ext3_ext_get_block(handle_t *handle, - - /* previous routine could use block we allocated */ - newblock = newex.ee_start; -+ allocated = newex.ee_len; - __set_bit(BH_New, &bh_result->b_state); - - ext3_ext_put_in_cache(&tree, newex.ee_block, newex.ee_len, -@@ -2341,6 +2359,9 @@ void ext3_ext_truncate(struct inode * in - down(&EXT3_I(inode)->truncate_sem); - ext3_ext_invalidate_cache(&tree); - -+ /* it's important to discard preallocations under truncate_sem */ -+ ext3_mb_discard_inode_preallocations(inode); -+ - /* - * TODO: optimization is possible here - * probably we need not scaning at all, -Index: linux-2.6.5-7.283-full/fs/ext3/Makefile -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/Makefile 2007-03-28 15:27:39.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/Makefile 2007-03-28 15:46:02.000000000 +0400 -@@ -6,7 +6,7 @@ obj-$(CONFIG_EXT3_FS) += ext3.o - - ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o iopen.o \ - ioctl.o namei.o super.o symlink.o hash.o \ -- extents.o -+ extents.o mballoc.o - - ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o - ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o -Index: linux-2.6.5-7.283-full/fs/ext3/xattr.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/xattr.c 2007-03-28 02:13:37.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/xattr.c 2007-03-28 15:46:02.000000000 +0400 -@@ -1371,7 +1371,7 @@ ext3_xattr_set_handle2(handle_t *handle, - new_bh = sb_getblk(sb, block); - if (!new_bh) { - getblk_failed: -- ext3_free_blocks(handle, inode, block, 1); -+ ext3_free_blocks(handle, inode, block, 1, 1); - error = -EIO; - goto cleanup; - } -@@ -1411,7 +1411,7 @@ getblk_failed: - if (HDR(old_bh)->h_refcount == cpu_to_le32(1)) { - /* Free the old block. */ - ea_bdebug(old_bh, "freeing"); -- ext3_free_blocks(handle, inode, old_bh->b_blocknr, 1); -+ ext3_free_blocks(handle, inode, old_bh->b_blocknr, 1, 1); - - /* ext3_forget() calls bforget() for us, but we - let our caller release old_bh, so we need to -@@ -1519,7 +1519,7 @@ ext3_xattr_delete_inode(handle_t *handle - mb_cache_entry_free(ce); - ce = NULL; - } -- ext3_free_blocks(handle, inode, EXT3_I(inode)->i_file_acl, 1); -+ ext3_free_blocks(handle, inode, EXT3_I(inode)->i_file_acl, 1, 1); - get_bh(bh); - ext3_forget(handle, 1, inode, bh, EXT3_I(inode)->i_file_acl); - } else { -Index: linux-2.6.5-7.283-full/fs/ext3/balloc.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/balloc.c 2006-12-01 18:39:48.000000000 +0300 -+++ linux-2.6.5-7.283-full/fs/ext3/balloc.c 2007-03-28 15:46:02.000000000 +0400 -@@ -78,7 +78,7 @@ struct ext3_group_desc * ext3_get_group_ - * - * Return buffer_head on success or NULL in case of failure. - */ --static struct buffer_head * -+struct buffer_head * - read_block_bitmap(struct super_block *sb, unsigned int block_group) - { - struct ext3_group_desc * desc; -@@ -266,6 +266,8 @@ void ext3_discard_reservation(struct ino - struct reserve_window_node *rsv = &ei->i_rsv_window; - spinlock_t *rsv_lock = &EXT3_SB(inode->i_sb)->s_rsv_window_lock; - -+ ext3_mb_discard_inode_preallocations(inode); -+ - if (!rsv_is_empty(&rsv->rsv_window)) { - spin_lock(rsv_lock); - rsv_window_remove(inode->i_sb, rsv); -@@ -274,7 +276,7 @@ void ext3_discard_reservation(struct ino - } - - /* Free given blocks, update quota and i_blocks field */ --void ext3_free_blocks(handle_t *handle, struct inode *inode, -+void ext3_free_blocks_old(handle_t *handle, struct inode *inode, - unsigned long block, unsigned long count) - { - struct buffer_head *bitmap_bh = NULL; -@@ -456,6 +458,29 @@ error_return: - return; - } - -+void ext3_free_blocks(handle_t *handle, struct inode * inode, -+ unsigned long block, unsigned long count, int metadata) -+{ -+ struct super_block *sb; -+ int freed; -+ -+ /* this isn't the right place to decide whether block is metadata -+ * inode.c/extents.c knows better, but for safety ... */ -+ if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode) || -+ ext3_should_journal_data(inode)) -+ metadata = 1; -+ -+ sb = inode->i_sb; -+ if (!test_opt(sb, MBALLOC) || !EXT3_SB(sb)->s_group_info) -+ ext3_free_blocks_old(handle, inode, block, count); -+ else { -+ ext3_mb_free_blocks(handle, inode, block, count, metadata, &freed); -+ if (freed) -+ DQUOT_FREE_BLOCK(inode, freed); -+ } -+ return; -+} -+ - /* - * For ext3 allocations, we must not reuse any blocks which are - * allocated in the bitmap buffer's "last committed data" copy. This -@@ -1142,7 +1167,7 @@ int ext3_should_retry_alloc(struct super - * bitmap, and then for any free bit if that fails. - * This function also updates quota and i_blocks field. - */ --int ext3_new_block(handle_t *handle, struct inode *inode, -+int ext3_new_block_old(handle_t *handle, struct inode *inode, - unsigned long goal, int *errp) - { - struct buffer_head *bitmap_bh = NULL; -Index: linux-2.6.5-7.283-full/fs/ext3/inode.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/inode.c 2007-03-28 02:50:19.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/inode.c 2007-03-28 15:46:02.000000000 +0400 -@@ -574,7 +574,7 @@ static int ext3_alloc_branch(handle_t *h - ext3_journal_forget(handle, branch[i].bh); - } - for (i = 0; i < keys; i++) -- ext3_free_blocks(handle, inode, le32_to_cpu(branch[i].key), 1); -+ ext3_free_blocks(handle, inode, le32_to_cpu(branch[i].key), 1, 0); - return err; - } - -@@ -675,7 +675,7 @@ err_out: - if (err == -EAGAIN) - for (i = 0; i < num; i++) - ext3_free_blocks(handle, inode, -- le32_to_cpu(where[i].key), 1); -+ le32_to_cpu(where[i].key), 1, 0); - return err; - } - -@@ -1839,7 +1839,7 @@ ext3_clear_blocks(handle_t *handle, stru - } - } - -- ext3_free_blocks(handle, inode, block_to_free, count); -+ ext3_free_blocks(handle, inode, block_to_free, count, 0); - } - - /** -@@ -2010,7 +2010,7 @@ static void ext3_free_branches(handle_t - ext3_journal_test_restart(handle, inode); - } - -- ext3_free_blocks(handle, inode, nr, 1); -+ ext3_free_blocks(handle, inode, nr, 1, 1); - - if (parent_bh) { - /* diff --git a/ldiskfs/kernel_patches/patches/ext3-nanosecond-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-nanosecond-2.6-suse.patch deleted file mode 100644 index 7824114..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-nanosecond-2.6-suse.patch +++ /dev/null @@ -1,195 +0,0 @@ -Index: linux-2.6.5-7.283/fs/ext3/ialloc.c -=================================================================== ---- linux-2.6.5-7.283.orig/fs/ext3/ialloc.c -+++ linux-2.6.5-7.283/fs/ext3/ialloc.c -@@ -613,7 +613,8 @@ got: - /* This is the optimal IO size (for stat), not the fs block size */ - inode->i_blksize = PAGE_SIZE; - inode->i_blocks = 0; -- inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; -+ inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime = -+ CURRENT_TIME; - - memset(ei->i_data, 0, sizeof(ei->i_data)); - ei->i_next_alloc_block = 0; -@@ -651,9 +652,8 @@ got: - spin_unlock(&sbi->s_next_gen_lock); - - ei->i_state = EXT3_STATE_NEW; -- ei->i_extra_isize = -- (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) ? -- sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE : 0; -+ -+ ei->i_extra_isize = EXT3_SB(sb)->s_want_extra_isize; - - ret = inode; - if(DQUOT_ALLOC_INODE(inode)) { -Index: linux-2.6.5-7.283/fs/ext3/inode.c -=================================================================== ---- linux-2.6.5-7.283.orig/fs/ext3/inode.c -+++ linux-2.6.5-7.283/fs/ext3/inode.c -@@ -2459,7 +2459,11 @@ void ext3_read_inode(struct inode * inod - inode->i_atime.tv_sec = le32_to_cpu(raw_inode->i_atime); - inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->i_ctime); - inode->i_mtime.tv_sec = le32_to_cpu(raw_inode->i_mtime); -- inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_mtime.tv_nsec = 0; -+ if (EXT3_FITS_IN_INODE(raw_inode, ei, i_crtime)) { -+ ei->i_crtime.tv_sec = le32_to_cpu(raw_inode->i_crtime); -+ } -+ inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_mtime.tv_nsec = -+ ei->i_crtime.tv_nsec = 0; - - ei->i_state = 0; - ei->i_next_alloc_block = 0; -@@ -2603,6 +2607,10 @@ static int ext3_do_update_inode(handle_t - raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec); - raw_inode->i_ctime = cpu_to_le32(inode->i_ctime.tv_sec); - raw_inode->i_mtime = cpu_to_le32(inode->i_mtime.tv_sec); -+ if (EXT3_FITS_IN_INODE(raw_inode, ei, i_crtime)) { -+ raw_inode->i_crtime = cpu_to_le32(ei->i_crtime.tv_sec); -+ } -+ - raw_inode->i_blocks = cpu_to_le32(inode->i_blocks); - raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); - raw_inode->i_flags = cpu_to_le32(ei->i_flags); -Index: linux-2.6.5-7.283/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-7.283.orig/fs/ext3/super.c -+++ linux-2.6.5-7.283/fs/ext3/super.c -@@ -1515,6 +1515,32 @@ static int ext3_fill_super (struct super - } - - ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY); -+ -+ /* determine the minimum size of new large inodes, if present */ -+ if (sbi->s_inode_size > EXT3_GOOD_OLD_INODE_SIZE) { -+ sbi->s_want_extra_isize = sizeof(struct ext3_inode) - -+ EXT3_GOOD_OLD_INODE_SIZE; -+ if (EXT3_HAS_RO_COMPAT_FEATURE(sb, -+ EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE)) { -+ if (sbi->s_want_extra_isize < -+ le16_to_cpu(es->s_want_extra_isize)) -+ sbi->s_want_extra_isize = -+ le16_to_cpu(es->s_want_extra_isize); -+ if (sbi->s_want_extra_isize < -+ le16_to_cpu(es->s_min_extra_isize)) -+ sbi->s_want_extra_isize = -+ le16_to_cpu(es->s_min_extra_isize); -+ } -+ } -+ /* Check if enough inode space is available */ -+ if (EXT3_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize > -+ sbi->s_inode_size) { -+ sbi->s_want_extra_isize = sizeof(struct ext3_inode) - -+ EXT3_GOOD_OLD_INODE_SIZE; -+ printk(KERN_INFO "EXT3-fs: required extra inode space not" -+ "available.\n"); -+ } -+ - /* - * akpm: core read_super() calls in here with the superblock locked. - * That deadlocks, because orphan cleanup needs to lock the superblock -Index: linux-2.6.5-7.283/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-7.283.orig/include/linux/ext3_fs.h -+++ linux-2.6.5-7.283/include/linux/ext3_fs.h -@@ -232,7 +232,7 @@ struct ext3_inode { - __u16 i_uid; /* Low 16 bits of Owner Uid */ - __u32 i_size; /* Size in bytes */ - __u32 i_atime; /* Access time */ -- __u32 i_ctime; /* Creation time */ -+ __u32 i_ctime; /* Inode Change time */ - __u32 i_mtime; /* Modification time */ - __u32 i_dtime; /* Deletion Time */ - __u16 i_gid; /* Low 16 bits of Group Id */ -@@ -281,10 +281,25 @@ struct ext3_inode { - } osd2; /* OS dependent 2 */ - __u16 i_extra_isize; - __u16 i_pad1; -+ __le32 i_ctime_extra; /* extra Change time (nsec << 2 | epoch) */ -+ __le32 i_mtime_extra; /* extra Modification time(nsec << 2 | epoch) */ -+ __le32 i_atime_extra; /* extra Access time (nsec << 2 | epoch) */ -+ __le32 i_crtime; /* File Creation time */ -+ __le32 i_crtime_extra; /* extra File Creation time (nsec << 2 | epoch) */ - }; - - #define i_size_high i_dir_acl - -+#define EXT3_EPOCH_BITS 2 -+#define EXT3_EPOCH_MASK ((1 << EXT3_EPOCH_BITS) - 1) -+#define EXT3_NSEC_MASK (~0UL << EXT3_EPOCH_BITS) -+ -+#define EXT3_FITS_IN_INODE(ext3_inode, einode, field) \ -+ ((offsetof(typeof(*ext3_inode), field) + \ -+ sizeof((ext3_inode)->field)) \ -+ <= (EXT3_GOOD_OLD_INODE_SIZE + \ -+ (einode)->i_extra_isize)) \ -+ - #if defined(__KERNEL__) || defined(__linux__) - #define i_reserved1 osd1.linux1.l_i_reserved1 - #define i_frag osd2.linux2.l_i_frag -@@ -460,11 +475,19 @@ struct ext3_super_block { - __u32 s_last_orphan; /* start of list of inodes to delete */ - __u32 s_hash_seed[4]; /* HTREE hash seed */ - __u8 s_def_hash_version; /* Default hash version to use */ -- __u8 s_reserved_char_pad; -- __u16 s_reserved_word_pad; -+ __u8 s_jnl_backup_type; /* Default type of journal backup */ -+ __u16 s_desc_size; /* Group desc. size: INCOMPAT_64BIT */ - __u32 s_default_mount_opts; -- __u32 s_first_meta_bg; /* First metablock block group */ -- __u32 s_reserved[190]; /* Padding to the end of the block */ -+ __u32 s_first_meta_bg; /* First metablock block group */ -+ __u32 s_mkfs_time; /* When the filesystem was created */ -+ __u32 s_jnl_blocks[17]; /* Backup of the journal inode */ -+ __u32 s_blocks_count_hi; /* Blocks count high 32 bits */ -+ __u32 s_r_blocks_count_hi; /* Reserved blocks count high 32 bits*/ -+ __u32 s_free_blocks_hi; /* Free blocks count high 32 bits */ -+ __u16 s_min_extra_isize; /* All inodes have at least # bytes */ -+ __u16 s_want_extra_isize; /* New inodes should reserve # bytes */ -+ __u32 s_flags; /* Miscellaneous flags */ -+ __u32 s_reserved[167]; /* Padding to the end of the block */ - }; - - #ifdef __KERNEL__ -@@ -539,6 +556,7 @@ static inline struct ext3_inode_info *EX - #define EXT3_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 - #define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010 - #define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020 -+#define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040 - - #define EXT3_FEATURE_INCOMPAT_COMPRESSION 0x0001 - #define EXT3_FEATURE_INCOMPAT_FILETYPE 0x0002 -@@ -555,6 +573,7 @@ static inline struct ext3_inode_info *EX - EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \ - EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \ - EXT4_FEATURE_RO_COMPAT_DIR_NLINK| \ -+ EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE| \ - EXT3_FEATURE_RO_COMPAT_BTREE_DIR) - - /* -Index: linux-2.6.5-7.283/include/linux/ext3_fs_i.h -=================================================================== ---- linux-2.6.5-7.283.orig/include/linux/ext3_fs_i.h -+++ linux-2.6.5-7.283/include/linux/ext3_fs_i.h -@@ -130,6 +130,7 @@ struct ext3_inode_info { - - /* on-disk additional length */ - __u16 i_extra_isize; -+ struct timespec i_crtime; - - /* - * truncate_sem is for serialising ext3_truncate() against -Index: linux-2.6.5-7.283/include/linux/ext3_fs_sb.h -=================================================================== ---- linux-2.6.5-7.283.orig/include/linux/ext3_fs_sb.h -+++ linux-2.6.5-7.283/include/linux/ext3_fs_sb.h -@@ -71,6 +71,8 @@ struct ext3_sb_info { - /* Last group used to allocate inode */ - int s_last_alloc_group; - -+ unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */ -+ - /* root of the per fs reservation window tree */ - spinlock_t s_rsv_window_lock; - struct rb_root s_rsv_window_root; diff --git a/ldiskfs/kernel_patches/patches/ext3-rename-reserve-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-rename-reserve-2.6-suse.patch deleted file mode 100644 index f323584..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-rename-reserve-2.6-suse.patch +++ /dev/null @@ -1,263 +0,0 @@ -Index: linux-2.6.5-sles9/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_fs.h 2004-11-09 02:29:14.878513832 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_fs.h 2004-11-09 02:32:14.151260232 +0300 -@@ -709,7 +709,7 @@ - unsigned int block_group, - struct buffer_head ** bh); - extern int ext3_should_retry_alloc(struct super_block *sb, int *retries); --extern void rsv_window_add(struct super_block *sb, struct reserve_window_node *rsv); -+extern void rsv_window_add(struct super_block *sb, struct ext3_reserve_window_node *rsv); - - /* dir.c */ - extern int ext3_check_dir_entry(const char *, struct inode *, -Index: linux-2.6.5-sles9/include/linux/ext3_fs_sb.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_fs_sb.h 2004-11-09 02:28:18.753046200 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_fs_sb.h 2004-11-09 02:32:27.996155488 +0300 -@@ -86,7 +86,7 @@ - /* root of the per fs reservation window tree */ - spinlock_t s_rsv_window_lock; - struct rb_root s_rsv_window_root; -- struct reserve_window_node s_rsv_window_head; -+ struct ext3_reserve_window_node s_rsv_window_head; - - /* Journaling */ - struct inode * s_journal_inode; -Index: linux-2.6.5-sles9/include/linux/ext3_fs_i.h -=================================================================== ---- linux-2.6.5-sles9.orig/include/linux/ext3_fs_i.h 2004-11-09 02:23:21.606219384 +0300 -+++ linux-2.6.5-sles9/include/linux/ext3_fs_i.h 2004-11-09 02:32:08.752081032 +0300 -@@ -20,17 +20,17 @@ - #include - #include - --struct reserve_window { -+struct ext3_reserve_window { - __u32 _rsv_start; /* First byte reserved */ - __u32 _rsv_end; /* Last byte reserved or 0 */ - }; - --struct reserve_window_node { -+struct ext3_reserve_window_node { - struct rb_node rsv_node; - atomic_t rsv_goal_size; - atomic_t rsv_alloc_hit; - seqlock_t rsv_seqlock; -- struct reserve_window rsv_window; -+ struct ext3_reserve_window rsv_window; - }; - - #define rsv_start rsv_window._rsv_start -@@ -76,7 +76,7 @@ - */ - __u32 i_next_alloc_goal; - /* block reservation window */ -- struct reserve_window_node i_rsv_window; -+ struct ext3_reserve_window_node i_rsv_window; - - __u32 i_dir_start_lookup; - #ifdef CONFIG_EXT3_FS_XATTR -Index: linux-2.6.5-sles9/fs/ext3/balloc.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/balloc.c 2004-11-09 02:26:53.078070776 +0300 -+++ linux-2.6.5-sles9/fs/ext3/balloc.c 2004-11-09 02:32:43.108858008 +0300 -@@ -115,7 +115,7 @@ - const char *fn) - { - struct rb_node *n; -- struct reserve_window_node *rsv, *prev; -+ struct ext3_reserve_window_node *rsv, *prev; - int bad; - - restart: -@@ -125,7 +125,7 @@ - - printk("Block Allocation Reservation Windows Map (%s):\n", fn); - while (n) { -- rsv = list_entry(n, struct reserve_window_node, rsv_node); -+ rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node); - if (verbose) - printk("reservation window 0x%p " - "start: %d, end: %d\n", -@@ -161,7 +161,7 @@ - #endif - - static int --goal_in_my_reservation(struct reserve_window *rsv, int goal, -+goal_in_my_reservation(struct ext3_reserve_window *rsv, int goal, - unsigned int group, struct super_block * sb) - { - unsigned long group_first_block, group_last_block; -@@ -184,18 +184,18 @@ - * if the goal is not in any window. - * Returns NULL if there are no windows or if all windows start after the goal. - */ --static struct reserve_window_node *search_reserve_window(struct rb_root *root, -+static struct ext3_reserve_window_node *search_ext3_reserve_window(struct rb_root *root, - unsigned long goal) - { - struct rb_node *n = root->rb_node; -- struct reserve_window_node *rsv; -+ struct ext3_reserve_window_node *rsv; - - if (!n) - return NULL; - - while (n) - { -- rsv = rb_entry(n, struct reserve_window_node, rsv_node); -+ rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); - - if (goal < rsv->rsv_start) - n = n->rb_left; -@@ -212,13 +212,13 @@ - */ - if (rsv->rsv_start > goal) { - n = rb_prev(&rsv->rsv_node); -- rsv = rb_entry(n, struct reserve_window_node, rsv_node); -+ rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); - } - return rsv; - } - - void rsv_window_add(struct super_block *sb, -- struct reserve_window_node *rsv) -+ struct ext3_reserve_window_node *rsv) - { - struct rb_root *root = &EXT3_SB(sb)->s_rsv_window_root; - struct rb_node *node = &rsv->rsv_node; -@@ -226,12 +226,12 @@ - - struct rb_node ** p = &root->rb_node; - struct rb_node * parent = NULL; -- struct reserve_window_node *this; -+ struct ext3_reserve_window_node *this; - - while (*p) - { - parent = *p; -- this = rb_entry(parent, struct reserve_window_node, rsv_node); -+ this = rb_entry(parent, struct ext3_reserve_window_node, rsv_node); - - if (start < this->rsv_start) - p = &(*p)->rb_left; -@@ -246,7 +246,7 @@ - } - - static void rsv_window_remove(struct super_block *sb, -- struct reserve_window_node *rsv) -+ struct ext3_reserve_window_node *rsv) - { - rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; -@@ -254,7 +254,7 @@ - rb_erase(&rsv->rsv_node, &EXT3_SB(sb)->s_rsv_window_root); - } - --static inline int rsv_is_empty(struct reserve_window *rsv) -+static inline int rsv_is_empty(struct ext3_reserve_window *rsv) - { - /* a valid reservation end block could not be 0 */ - return (rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED); -@@ -263,7 +263,7 @@ - void ext3_discard_reservation(struct inode *inode) - { - struct ext3_inode_info *ei = EXT3_I(inode); -- struct reserve_window_node *rsv = &ei->i_rsv_window; -+ struct ext3_reserve_window_node *rsv = &ei->i_rsv_window; - spinlock_t *rsv_lock = &EXT3_SB(inode->i_sb)->s_rsv_window_lock; - - if (!rsv_is_empty(&rsv->rsv_window)) { -@@ -600,7 +600,7 @@ - */ - static int - ext3_try_to_allocate(struct super_block *sb, handle_t *handle, int group, -- struct buffer_head *bitmap_bh, int goal, struct reserve_window *my_rsv) -+ struct buffer_head *bitmap_bh, int goal, struct ext3_reserve_window *my_rsv) - { - int group_first_block, start, end; - -@@ -700,13 +700,13 @@ - * on succeed, it returns the reservation window to be appended to. - * failed, return NULL. - */ --static struct reserve_window_node *find_next_reservable_window( -- struct reserve_window_node *search_head, -+static struct ext3_reserve_window_node *find_next_reservable_window( -+ struct ext3_reserve_window_node *search_head, - unsigned long size, int *start_block, - int last_block) - { - struct rb_node *next; -- struct reserve_window_node *rsv, *prev; -+ struct ext3_reserve_window_node *rsv, *prev; - int cur; - - /* TODO: make the start of the reservation window byte-aligned */ -@@ -734,7 +734,7 @@ - - prev = rsv; - next = rb_next(&rsv->rsv_node); -- rsv = list_entry(next, struct reserve_window_node, rsv_node); -+ rsv = list_entry(next, struct ext3_reserve_window_node, rsv_node); - - /* - * Reached the last reservation, we can just append to the -@@ -801,15 +801,15 @@ - * @group: the group we are trying to allocate in - * @bitmap_bh: the block group block bitmap - */ --static int alloc_new_reservation(struct reserve_window_node *my_rsv, -+static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv, - int goal, struct super_block *sb, - unsigned int group, struct buffer_head *bitmap_bh) - { -- struct reserve_window_node *search_head; -+ struct ext3_reserve_window_node *search_head; - int group_first_block, group_end_block, start_block; - int first_free_block; - int reservable_space_start; -- struct reserve_window_node *prev_rsv; -+ struct ext3_reserve_window_node *prev_rsv; - struct rb_root *fs_rsv_root = &EXT3_SB(sb)->s_rsv_window_root; - unsigned long size; - -@@ -859,7 +859,7 @@ - /* - * shift the search start to the window near the goal block - */ -- search_head = search_reserve_window(fs_rsv_root, start_block); -+ search_head = search_ext3_reserve_window(fs_rsv_root, start_block); - - /* - * find_next_reservable_window() simply finds a reservable window -@@ -968,7 +968,7 @@ - static int - ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, - unsigned int group, struct buffer_head *bitmap_bh, -- int goal, struct reserve_window_node * my_rsv, -+ int goal, struct ext3_reserve_window_node * my_rsv, - int *errp) - { - spinlock_t *rsv_lock; -@@ -1027,7 +1027,7 @@ - * then we could go to allocate from the reservation window directly. - */ - while (1) { -- struct reserve_window rsv_copy; -+ struct ext3_reserve_window rsv_copy; - unsigned int seq; - - do { -@@ -1159,8 +1159,8 @@ - struct ext3_group_desc *gdp; - struct ext3_super_block *es; - struct ext3_sb_info *sbi; -- struct reserve_window_node *my_rsv = NULL; -- struct reserve_window_node *rsv = &EXT3_I(inode)->i_rsv_window; -+ struct ext3_reserve_window_node *my_rsv = NULL; -+ struct ext3_reserve_window_node *rsv = &EXT3_I(inode)->i_rsv_window; - unsigned short windowsz = 0; - #ifdef EXT3FS_DEBUG - static int goal_hits, goal_attempts; diff --git a/ldiskfs/kernel_patches/patches/ext3-sector_t-overflow-2.6.5-suse.patch b/ldiskfs/kernel_patches/patches/ext3-sector_t-overflow-2.6.5-suse.patch deleted file mode 100644 index fe655da..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-sector_t-overflow-2.6.5-suse.patch +++ /dev/null @@ -1,44 +0,0 @@ -Subject: Avoid disk sector_t overflow for >2TB ext3 filesystem -From: Mingming Cao - - -If ext3 filesystem is larger than 2TB, and sector_t is a u32 (i.e. -CONFIG_LBD not defined in the kernel), the calculation of the disk sector -will overflow. Add check at ext3_fill_super() and ext3_group_extend() to -prevent mount/remount/resize >2TB ext3 filesystem if sector_t size is 4 -bytes. - -Verified this patch on a 32 bit platform without CONFIG_LBD defined -(sector_t is 32 bits long), mount refuse to mount a 10TB ext3. - -Signed-off-by: Mingming Cao -Acked-by: Andreas Dilger -Signed-off-by: Andrew Morton ---- - - fs/ext3/resize.c | 10 ++++++++++ - fs/ext3/super.c | 10 ++++++++++ - 2 files changed, 20 insertions(+) - -diff -puN fs/ext3/super.c~avoid-disk-sector_t-overflow-for-2tb-ext3-filesystem fs/ext3/super.c ---- devel/fs/ext3/super.c~avoid-disk-sector_t-overflow-for-2tb-ext3-filesystem 2006-05-22 14:09:53.000000000 -0700 -+++ devel-akpm/fs/ext3/super.c 2006-05-22 14:11:10.000000000 -0700 -@@ -1565,6 +1565,17 @@ static int ext3_fill_super (struct super - goto failed_mount; - } - -+ if (le32_to_cpu(es->s_blocks_count) > -+ (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { -+ printk(KERN_ERR "EXT3-fs: filesystem on %s: " -+ "too large to mount safely - %u blocks\n", sb->s_id, -+ le32_to_cpu(es->s_blocks_count)); -+ if (sizeof(sector_t) < 8) -+ printk(KERN_WARNING -+ "EXT3-fs: CONFIG_LBD not enabled\n"); -+ goto failed_mount; -+ } -+ - sbi->s_groups_count = (le32_to_cpu(es->s_blocks_count) - - le32_to_cpu(es->s_first_data_block) + - EXT3_BLOCKS_PER_GROUP(sb) - 1) / -_ diff --git a/ldiskfs/kernel_patches/patches/ext3-uninit-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-uninit-2.6-suse.patch deleted file mode 100644 index 80bf99b8..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-uninit-2.6-suse.patch +++ /dev/null @@ -1,653 +0,0 @@ -Add support for the uninit_groups feature to the kernel. - -Keep a high water mark of used inodes for each group to improve e2fsck time. -Block and inode bitmaps can be uninitialized on disk via a flag in the -group descriptor to avoid reading or scanning them at e2fsck time. -A checksum of each group descriptor is used to ensure that corruption in -the group descriptor's bit flags does not cause incorrect operation. - -Index: linux-2.6.5-7.283-full/include/linux/ext3_fs.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/include/linux/ext3_fs.h 2007-03-28 17:33:05.000000000 +0400 -+++ linux-2.6.5-7.283-full/include/linux/ext3_fs.h 2007-03-28 18:33:35.000000000 +0400 -@@ -153,16 +153,22 @@ struct ext3_allocation_request { - */ - struct ext3_group_desc - { -- __u32 bg_block_bitmap; /* Blocks bitmap block */ -- __u32 bg_inode_bitmap; /* Inodes bitmap block */ -+ __u32 bg_block_bitmap; /* Blocks bitmap block */ -+ __u32 bg_inode_bitmap; /* Inodes bitmap block */ - __u32 bg_inode_table; /* Inodes table block */ - __u16 bg_free_blocks_count; /* Free blocks count */ - __u16 bg_free_inodes_count; /* Free inodes count */ - __u16 bg_used_dirs_count; /* Directories count */ -- __u16 bg_pad; -- __u32 bg_reserved[3]; -+ __u16 bg_flags; /* EXT3_BG_flags (UNINIT, etc) */ -+ __u32 bg_reserved[2]; /* Likely block/inode bitmap checksum */ -+ __u16 bg_itable_unused; /* Unused inodes count */ -+ __u16 bg_checksum; /* crc16(sb_uuid+group+desc) */ - }; - -+#define EXT3_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */ -+#define EXT3_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */ -+#define EXT3_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */ -+ - /* - * Macro-instructions used to manage group descriptors - */ -@@ -458,7 +464,7 @@ struct ext3_super_block { - */ - __u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/ - __u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */ -- __u16 s_padding1; -+ __u16 s_reserved_gdt_blocks; /* Per group desc for online growth */ - /* - * Journaling support valid if EXT3_FEATURE_COMPAT_HAS_JOURNAL set. - */ -@@ -546,6 +552,7 @@ static inline struct ext3_inode_info *EX - #define EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 - #define EXT3_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 - #define EXT3_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 -+#define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010 - #define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020 - - #define EXT3_FEATURE_INCOMPAT_COMPRESSION 0x0001 -@@ -562,6 +569,7 @@ static inline struct ext3_inode_info *EX - EXT3_FEATURE_INCOMPAT_EXTENTS) - #define EXT3_FEATURE_RO_COMPAT_SUPP (EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER| \ - EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \ -+ EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \ - EXT4_FEATURE_RO_COMPAT_DIR_NLINK| \ - EXT3_FEATURE_RO_COMPAT_BTREE_DIR) - -Index: linux-2.6.5-7.283-full/fs/ext3/super.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/super.c 2007-03-28 17:33:05.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/super.c 2007-03-28 18:33:35.000000000 +0400 -@@ -36,6 +36,7 @@ - #include - #include "xattr.h" - #include "acl.h" -+#include "group.h" - - static int ext3_load_journal(struct super_block *, struct ext3_super_block *); - static int ext3_create_journal(struct super_block *, struct ext3_super_block *, -@@ -996,6 +997,90 @@ static int ext3_setup_super(struct super - return res; - } - -+#if !defined(CONFIG_CRC16) && !defined(CONFIG_CRC16_MODULE) -+/** CRC table for the CRC-16. The poly is 0x8005 (x^16 + x^15 + x^2 + 1) */ -+__u16 const crc16_table[256] = { -+ 0x0000, 0xC0C1, 0xC181, 0x0140, 0xC301, 0x03C0, 0x0280, 0xC241, -+ 0xC601, 0x06C0, 0x0780, 0xC741, 0x0500, 0xC5C1, 0xC481, 0x0440, -+ 0xCC01, 0x0CC0, 0x0D80, 0xCD41, 0x0F00, 0xCFC1, 0xCE81, 0x0E40, -+ 0x0A00, 0xCAC1, 0xCB81, 0x0B40, 0xC901, 0x09C0, 0x0880, 0xC841, -+ 0xD801, 0x18C0, 0x1980, 0xD941, 0x1B00, 0xDBC1, 0xDA81, 0x1A40, -+ 0x1E00, 0xDEC1, 0xDF81, 0x1F40, 0xDD01, 0x1DC0, 0x1C80, 0xDC41, -+ 0x1400, 0xD4C1, 0xD581, 0x1540, 0xD701, 0x17C0, 0x1680, 0xD641, -+ 0xD201, 0x12C0, 0x1380, 0xD341, 0x1100, 0xD1C1, 0xD081, 0x1040, -+ 0xF001, 0x30C0, 0x3180, 0xF141, 0x3300, 0xF3C1, 0xF281, 0x3240, -+ 0x3600, 0xF6C1, 0xF781, 0x3740, 0xF501, 0x35C0, 0x3480, 0xF441, -+ 0x3C00, 0xFCC1, 0xFD81, 0x3D40, 0xFF01, 0x3FC0, 0x3E80, 0xFE41, -+ 0xFA01, 0x3AC0, 0x3B80, 0xFB41, 0x3900, 0xF9C1, 0xF881, 0x3840, -+ 0x2800, 0xE8C1, 0xE981, 0x2940, 0xEB01, 0x2BC0, 0x2A80, 0xEA41, -+ 0xEE01, 0x2EC0, 0x2F80, 0xEF41, 0x2D00, 0xEDC1, 0xEC81, 0x2C40, -+ 0xE401, 0x24C0, 0x2580, 0xE541, 0x2700, 0xE7C1, 0xE681, 0x2640, -+ 0x2200, 0xE2C1, 0xE381, 0x2340, 0xE101, 0x21C0, 0x2080, 0xE041, -+ 0xA001, 0x60C0, 0x6180, 0xA141, 0x6300, 0xA3C1, 0xA281, 0x6240, -+ 0x6600, 0xA6C1, 0xA781, 0x6740, 0xA501, 0x65C0, 0x6480, 0xA441, -+ 0x6C00, 0xACC1, 0xAD81, 0x6D40, 0xAF01, 0x6FC0, 0x6E80, 0xAE41, -+ 0xAA01, 0x6AC0, 0x6B80, 0xAB41, 0x6900, 0xA9C1, 0xA881, 0x6840, -+ 0x7800, 0xB8C1, 0xB981, 0x7940, 0xBB01, 0x7BC0, 0x7A80, 0xBA41, -+ 0xBE01, 0x7EC0, 0x7F80, 0xBF41, 0x7D00, 0xBDC1, 0xBC81, 0x7C40, -+ 0xB401, 0x74C0, 0x7580, 0xB541, 0x7700, 0xB7C1, 0xB681, 0x7640, -+ 0x7200, 0xB2C1, 0xB381, 0x7340, 0xB101, 0x71C0, 0x7080, 0xB041, -+ 0x5000, 0x90C1, 0x9181, 0x5140, 0x9301, 0x53C0, 0x5280, 0x9241, -+ 0x9601, 0x56C0, 0x5780, 0x9741, 0x5500, 0x95C1, 0x9481, 0x5440, -+ 0x9C01, 0x5CC0, 0x5D80, 0x9D41, 0x5F00, 0x9FC1, 0x9E81, 0x5E40, -+ 0x5A00, 0x9AC1, 0x9B81, 0x5B40, 0x9901, 0x59C0, 0x5880, 0x9841, -+ 0x8801, 0x48C0, 0x4980, 0x8941, 0x4B00, 0x8BC1, 0x8A81, 0x4A40, -+ 0x4E00, 0x8EC1, 0x8F81, 0x4F40, 0x8D01, 0x4DC0, 0x4C80, 0x8C41, -+ 0x4400, 0x84C1, 0x8581, 0x4540, 0x8701, 0x47C0, 0x4680, 0x8641, -+ 0x8201, 0x42C0, 0x4380, 0x8341, 0x4100, 0x81C1, 0x8081, 0x4040 -+}; -+ -+static inline __u16 crc16_byte(__u16 crc, const __u8 data) -+{ -+ return (crc >> 8) ^ crc16_table[(crc ^ data) & 0xff]; -+} -+ -+__u16 crc16(__u16 crc, __u8 const *buffer, size_t len) -+{ -+ while (len--) -+ crc = crc16_byte(crc, *buffer++); -+ return crc; -+} -+#endif -+ -+__le16 ext3_group_desc_csum(struct ext3_sb_info *sbi, __u32 block_group, -+ struct ext3_group_desc *gdp) -+{ -+ __u16 crc = 0; -+ -+ if (sbi->s_es->s_feature_ro_compat & -+ cpu_to_le32(EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) { -+ int offset = offsetof(struct ext3_group_desc, bg_checksum); -+ __le32 le_group = cpu_to_le32(block_group); -+ -+ crc = crc16(~0, sbi->s_es->s_uuid, sizeof(sbi->s_es->s_uuid)); -+ crc = crc16(crc, (__u8 *)&le_group, sizeof(le_group)); -+ crc = crc16(crc, (__u8 *)gdp, offset); -+ offset += sizeof(gdp->bg_checksum); /* skip checksum */ -+ BUG_ON(offset != sizeof(*gdp)); /* XXX handle s_desc_size */ -+ /* for checksum of struct ext4_group_desc do the rest... -+ if (offset < sbi->s_es->s_desc_size) { -+ crc = crc16(crc, (__u8 *)gdp + offset, -+ sbi->s_es->s_desc_size - offset); -+ */ -+ } -+ -+ return cpu_to_le16(crc); -+} -+ -+int ext3_group_desc_csum_verify(struct ext3_sb_info *sbi, __u32 block_group, -+ struct ext3_group_desc *gdp) -+{ -+ if (gdp->bg_checksum != ext3_group_desc_csum(sbi, block_group, gdp)) -+ return 0; -+ -+ return 1; -+} -+ - static int ext3_check_descriptors (struct super_block * sb) - { - struct ext3_sb_info *sbi = EXT3_SB(sb); -@@ -1044,6 +1129,13 @@ static int ext3_check_descriptors (struc - le32_to_cpu(gdp->bg_inode_table)); - return 0; - } -+ if (!ext3_group_desc_csum_verify(sbi, i, gdp)) { -+ ext3_error(sb, __FUNCTION__, -+ "Checksum for group %d failed (%u!=%u)\n", i, -+ le16_to_cpu(ext3_group_desc_csum(sbi,i,gdp)), -+ le16_to_cpu(gdp->bg_checksum)); -+ return 0; -+ } - block += EXT3_BLOCKS_PER_GROUP(sb); - gdp++; - } -Index: linux-2.6.5-7.283-full/fs/ext3/group.h -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/group.h 2007-02-13 18:39:59.640066087 +0300 -+++ linux-2.6.5-7.283-full/fs/ext3/group.h 2007-03-28 18:33:35.000000000 +0400 -@@ -0,0 +1,29 @@ -+/* -+ * linux/fs/ext3/group.h -+ * -+ * Copyright (C) 2007 Cluster File Systems, Inc -+ * -+ * Author: Andreas Dilger -+ */ -+ -+#ifndef _LINUX_EXT3_GROUP_H -+#define _LINUX_EXT3_GROUP_H -+#if defined(CONFIG_CRC16) || defined(CONFIG_CRC16_MODULE) -+#include -+#endif -+ -+extern __le16 ext3_group_desc_csum(struct ext3_sb_info *sbi, __u32 group, -+ struct ext3_group_desc *gdp); -+extern int ext3_group_desc_csum_verify(struct ext3_sb_info *sbi, __u32 group, -+ struct ext3_group_desc *gdp); -+struct buffer_head *read_block_bitmap(struct super_block *sb, -+ unsigned int block_group); -+extern unsigned ext3_init_block_bitmap(struct super_block *sb, -+ struct buffer_head *bh, int group, -+ struct ext3_group_desc *desc); -+#define ext3_free_blocks_after_init(sb, group, desc) \ -+ ext3_init_block_bitmap(sb, NULL, group, desc) -+extern unsigned ext3_init_inode_bitmap(struct super_block *sb, -+ struct buffer_head *bh, int group, -+ struct ext3_group_desc *desc); -+#endif /* _LINUX_EXT3_GROUP_H */ -Index: linux-2.6.5-7.283-full/fs/ext3/ialloc.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/ialloc.c 2007-03-28 17:33:03.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/ialloc.c 2007-03-28 18:33:35.000000000 +0400 -@@ -28,6 +28,7 @@ - - #include "xattr.h" - #include "acl.h" -+#include "group.h" - - /* - * ialloc.c contains the inodes allocation and deallocation routines -@@ -43,6 +44,52 @@ - * the free blocks count in the block. - */ - -+/* -+ * To avoid calling the atomic setbit hundreds or thousands of times, we only -+ * need to use it within a single byte (to ensure we get endianness right). -+ * We can use memset for the rest of the bitmap as there are no other users. -+ */ -+static void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) -+{ -+ int i; -+ -+ if (start_bit >= end_bit) -+ return; -+ -+ ext3_debug("mark end bits +%d through +%d used\n", start_bit, end_bit); -+ for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++) -+ ext3_set_bit(i, bitmap); -+ if (i < end_bit) -+ memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3); -+} -+ -+/* Initializes an uninitialized inode bitmap */ -+unsigned ext3_init_inode_bitmap(struct super_block *sb, -+ struct buffer_head *bh, int block_group, -+ struct ext3_group_desc *gdp) -+{ -+ struct ext3_sb_info *sbi = EXT3_SB(sb); -+ -+ J_ASSERT_BH(bh, buffer_locked(bh)); -+ -+ /* If checksum is bad mark all blocks and inodes use to prevent -+ * allocation, essentially implementing a per-group read-only flag. */ -+ if (!ext3_group_desc_csum_verify(sbi, block_group, gdp)) { -+ ext3_error(sb, __FUNCTION__, "Checksum bad for group %u\n", -+ block_group); -+ gdp->bg_free_blocks_count = 0; -+ gdp->bg_free_inodes_count = 0; -+ gdp->bg_itable_unused = 0; -+ memset(bh->b_data, 0xff, sb->s_blocksize); -+ return 0; -+ } -+ -+ memset(bh->b_data, 0, (EXT3_INODES_PER_GROUP(sb) + 7) / 8); -+ mark_bitmap_end(EXT3_INODES_PER_GROUP(sb), EXT3_BLOCKS_PER_GROUP(sb), -+ bh->b_data); -+ -+ return EXT3_INODES_PER_GROUP(sb); -+} - - /* - * Read the inode allocation bitmap for a given block_group, reading -@@ -59,8 +106,19 @@ read_inode_bitmap(struct super_block * s - desc = ext3_get_group_desc(sb, block_group, NULL); - if (!desc) - goto error_out; -- -- bh = sb_bread(sb, le32_to_cpu(desc->bg_inode_bitmap)); -+ if (desc->bg_flags & cpu_to_le16(EXT3_BG_INODE_UNINIT)) { -+ bh = sb_getblk(sb, le32_to_cpu(desc->bg_inode_bitmap)); -+ if (!buffer_uptodate(bh)) { -+ lock_buffer(bh); -+ if (!buffer_uptodate(bh)) { -+ ext3_init_inode_bitmap(sb, bh,block_group,desc); -+ set_buffer_uptodate(bh); -+ } -+ unlock_buffer(bh); -+ } -+ } else { -+ bh = sb_bread(sb, le32_to_cpu(desc->bg_inode_bitmap)); -+ } - if (!bh) - ext3_error(sb, "read_inode_bitmap", - "Cannot read inode bitmap - " -@@ -168,6 +226,8 @@ void ext3_free_inode (handle_t *handle, - if (is_directory) - gdp->bg_used_dirs_count = cpu_to_le16( - le16_to_cpu(gdp->bg_used_dirs_count) - 1); -+ gdp->bg_checksum = ext3_group_desc_csum(sbi,block_group, -+ gdp); - spin_unlock(sb_bgl_lock(sbi, block_group)); - percpu_counter_inc(&sbi->s_freeinodes_counter); - if (is_directory) -@@ -454,7 +514,7 @@ struct inode *ext3_new_inode(handle_t *h - struct ext3_sb_info *sbi; - int err = 0; - struct inode *ret; -- int i; -+ int i, free = 0; - - /* Cannot create files in a deleted directory */ - if (!dir || !dir->i_nlink) -@@ -570,11 +630,13 @@ repeat_in_this_group: - goto out; - - got: -- ino += group * EXT3_INODES_PER_GROUP(sb) + 1; -- if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { -- ext3_error (sb, "ext3_new_inode", -- "reserved inode or inode > inodes count - " -- "block_group = %d, inode=%lu", group, ino); -+ ino++; -+ if ((group == 0 && ino < EXT3_FIRST_INO(sb)) || -+ ino > EXT3_INODES_PER_GROUP(sb)) { -+ ext3_error(sb, __FUNCTION__, -+ "reserved inode or inode > inodes count - " -+ "block_group = %d, inode=%lu", group, -+ ino + group * EXT3_INODES_PER_GROUP(sb)); - err = -EIO; - goto fail; - } -@@ -582,13 +644,65 @@ got: - BUFFER_TRACE(bh2, "get_write_access"); - err = ext3_journal_get_write_access(handle, bh2); - if (err) goto fail; -+ -+ /* We may have to initialize the block bitmap if it isn't already */ -+ if (EXT3_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM) && -+ gdp->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) { -+ struct buffer_head *block_bh = read_block_bitmap(sb, group); -+ -+ BUFFER_TRACE(block_bh, "get block bitmap access"); -+ err = ext3_journal_get_write_access(handle, block_bh); -+ if (err) { -+ brelse(block_bh); -+ goto fail; -+ } -+ -+ free = 0; -+ spin_lock(sb_bgl_lock(sbi, group)); -+ /* recheck and clear flag under lock if we still need to */ -+ if (gdp->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) { -+ gdp->bg_flags &= cpu_to_le16(~EXT3_BG_BLOCK_UNINIT); -+ free = ext3_free_blocks_after_init(sb, group, gdp); -+ gdp->bg_free_blocks_count = cpu_to_le16(free); -+ } -+ spin_unlock(sb_bgl_lock(sbi, group)); -+ -+ /* Don't need to dirty bitmap block if we didn't change it */ -+ if (free) { -+ BUFFER_TRACE(block_bh, "dirty block bitmap"); -+ err = ext3_journal_dirty_metadata(handle, block_bh); -+ } -+ -+ brelse(block_bh); -+ if (err) -+ goto fail; -+ } -+ - spin_lock(sb_bgl_lock(sbi, group)); -+ /* If we didn't allocate from within the initialized part of the inode -+ * table then we need to initialize up to this inode. */ -+ if (EXT3_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) { -+ if (gdp->bg_flags & cpu_to_le16(EXT3_BG_INODE_UNINIT)) { -+ gdp->bg_flags &= cpu_to_le16(~EXT3_BG_INODE_UNINIT); -+ free = 0; -+ } else { -+ free = EXT3_INODES_PER_GROUP(sb) - -+ le16_to_cpu(gdp->bg_itable_unused); -+ } -+ -+ if (ino > free) { -+ gdp->bg_itable_unused = -+ cpu_to_le16(EXT3_INODES_PER_GROUP(sb) - ino); -+ } -+ } -+ - gdp->bg_free_inodes_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); - if (S_ISDIR(mode)) { - gdp->bg_used_dirs_count = - cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); - } -+ gdp->bg_checksum = ext3_group_desc_csum(sbi, group, gdp); - spin_unlock(sb_bgl_lock(sbi, group)); - BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh2); -@@ -610,7 +724,7 @@ got: - inode->i_gid = current->fsgid; - inode->i_mode = mode; - -- inode->i_ino = ino; -+ inode->i_ino = ino + group * EXT3_INODES_PER_GROUP(sb); - /* This is the optimal IO size (for stat), not the fs block size */ - inode->i_blksize = PAGE_SIZE; - inode->i_blocks = 0; -Index: linux-2.6.5-7.283-full/fs/ext3/mballoc.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/mballoc.c 2007-03-28 15:46:00.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/mballoc.c 2007-03-28 18:33:35.000000000 +0400 -@@ -36,6 +36,8 @@ - #include - #include - -+#include "group.h" -+ - /* - * MUSTDO: - * - test ext3_ext_search_left() and ext3_ext_search_right() -@@ -323,6 +325,7 @@ struct ext3_group_info { - unsigned long bb_state; - unsigned long bb_tid; - struct ext3_free_metadata *bb_md_cur; -+ struct ext3_group_desc *bb_gdp; - unsigned short bb_first_free; - unsigned short bb_free; - unsigned short bb_fragments; -@@ -928,10 +931,7 @@ static int ext3_mb_init_cache(struct pag - if (first_group + i >= EXT3_SB(sb)->s_groups_count) - break; - -- err = -EIO; -- desc = ext3_get_group_desc(sb, first_group + i, NULL); -- if (desc == NULL) -- goto out; -+ desc = EXT3_GROUP_INFO(sb, first_group + i)->bb_gdp; - - err = -ENOMEM; - bh[i] = sb_getblk(sb, le32_to_cpu(desc->bg_block_bitmap)); -@@ -946,7 +946,12 @@ static int ext3_mb_init_cache(struct pag - unlock_buffer(bh[i]); - continue; - } -- -+ if (desc->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) { -+ ext3_init_block_bitmap(sb, bh[i], first_group + i,desc); -+ set_buffer_uptodate(bh[i]); -+ unlock_buffer(bh[i]); -+ continue; -+ } - get_bh(bh[i]); - bh[i]->b_end_io = end_buffer_read_sync; - submit_bh(READ, bh[i]); -@@ -1703,6 +1708,10 @@ static int ext3_mb_good_group(struct ext - switch (cr) { - case 0: - BUG_ON(ac->ac_2order == 0); -+ /* If this group is uninitialized, skip it initially */ -+ if (grp->bb_gdp->bg_flags & -+ cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) -+ return 0; - bits = ac->ac_sb->s_blocksize_bits + 1; - for (i = ac->ac_2order; i <= bits; i++) - if (grp->bb_counters[i] > 0) -@@ -1796,7 +1805,9 @@ repeat: - } - - ac->ac_groups_scanned++; -- if (cr == 0) -+ if (cr == 0 || (e3b.bd_info->bb_gdp->bg_flags & -+ cpu_to_le16(EXT3_BG_BLOCK_UNINIT) && -+ ac->ac_2order != 0)) - ext3_mb_simple_scan_group(ac, &e3b); - else if (cr == 1 && ac->ac_g_ex.fe_len == sbi->s_stripe) - ext3_mb_scan_aligned(ac, &e3b); -@@ -2267,12 +2278,13 @@ int ext3_mb_init_backend(struct super_bl - i--; - goto err_freebuddy; - } -+ memset(meta_group_info[j], 0, len); - desc = ext3_get_group_desc(sb, i, NULL); -+ meta_group_info[j]->bb_gdp = desc; - if (desc == NULL) { - printk(KERN_ERR"EXT3-fs: can't read descriptor %u\n",i); - goto err_freebuddy; - } -- memset(meta_group_info[j], 0, len); - set_bit(EXT3_GROUP_INFO_NEED_INIT_BIT, - &meta_group_info[j]->bb_state); - -@@ -2936,9 +2948,17 @@ int ext3_mb_mark_diskspace_used(struct e - mb_set_bits(bitmap_bh->b_data, ac->ac_b_ex.fe_start, ac->ac_b_ex.fe_len); - - spin_lock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); -+ if (gdp->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) { -+ gdp->bg_flags &= cpu_to_le16(~EXT3_BG_BLOCK_UNINIT); -+ gdp->bg_free_blocks_count = -+ cpu_to_le16(ext3_free_blocks_after_init(sb, -+ ac->ac_b_ex.fe_group, -+ gdp)); -+ } - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) - - ac->ac_b_ex.fe_len); -+ gdp->bg_checksum = ext3_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); - spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); - percpu_counter_mod(&sbi->s_freeblocks_counter, - ac->ac_b_ex.fe_len); - -@@ -4303,6 +4323,7 @@ do_more: - spin_lock(sb_bgl_lock(sbi, block_group)); - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count); -+ gdp->bg_checksum = ext3_group_desc_csum(sbi, block_group, gdp); - spin_unlock(sb_bgl_lock(sbi, block_group)); - percpu_counter_mod(&sbi->s_freeblocks_counter, count); - -Index: linux-2.6.5-7.283-full/fs/ext3/balloc.c -=================================================================== ---- linux-2.6.5-7.283-full.orig/fs/ext3/balloc.c 2007-03-28 17:33:02.000000000 +0400 -+++ linux-2.6.5-7.283-full/fs/ext3/balloc.c 2007-03-28 18:33:35.000000000 +0400 -@@ -20,6 +20,7 @@ - #include - #include - -+#include "group.h" - /* - * balloc.c contains the blocks allocation and deallocation routines - */ -@@ -72,6 +73,75 @@ struct ext3_group_desc * ext3_get_group_ - return gdp + desc; - } - -+/* Initializes an uninitialized block bitmap if given, and returns the -+ * number of blocks free in the group. */ -+unsigned ext3_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, -+ int block_group, struct ext3_group_desc *gdp) -+{ -+ unsigned long start; -+ int bit, bit_max; -+ unsigned free_blocks; -+ struct ext3_sb_info *sbi = EXT3_SB(sb); -+ -+ if (bh) { -+ J_ASSERT_BH(bh, buffer_locked(bh)); -+ -+ /* If checksum is bad mark all blocks use to prevent allocation, -+ * essentially implementing a per-group read-only flag. */ -+ if (!ext3_group_desc_csum_verify(sbi, block_group, gdp)) { -+ ext3_error(sb, __FUNCTION__, -+ "Checksum bad for group %u\n", block_group); -+ gdp->bg_free_blocks_count = 0; -+ gdp->bg_free_inodes_count = 0; -+ gdp->bg_itable_unused = 0; -+ memset(bh->b_data, 0xff, sb->s_blocksize); -+ return 0; -+ } -+ memset(bh->b_data, 0, sb->s_blocksize); -+ } -+ -+ /* Check for superblock and gdt backups in this group */ -+ bit_max = ext3_bg_has_super(sb, block_group); -+ -+ if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_META_BG) || -+ block_group < le32_to_cpu(sbi->s_es->s_first_meta_bg) * -+ sbi->s_desc_per_block) { -+ if (bit_max) { -+ bit_max += ext3_bg_num_gdb(sb, block_group); -+ bit_max +=le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks); -+ } -+ } else { /* For META_BG_BLOCK_GROUPS */ -+ int group_rel = (block_group - -+ le32_to_cpu(sbi->s_es->s_first_meta_bg)) % -+ EXT3_DESC_PER_BLOCK(sb); -+ if (group_rel == 0 || group_rel == 1 || -+ (group_rel == EXT3_DESC_PER_BLOCK(sb) - 1)) -+ bit_max += 1; -+ } -+ -+ /* Last and first groups are always initialized */ -+ free_blocks = EXT3_BLOCKS_PER_GROUP(sb) - bit_max; -+ -+ if (bh) { -+ for (bit = 0; bit < bit_max; bit++) -+ ext3_set_bit(bit, bh->b_data); -+ -+ start = block_group * EXT3_BLOCKS_PER_GROUP(sb) + -+ le32_to_cpu(sbi->s_es->s_first_data_block); -+ -+ /* Set bits for block and inode bitmaps, and inode table */ -+ ext3_set_bit(le32_to_cpu(gdp->bg_block_bitmap) - start, -+ bh->b_data); -+ ext3_set_bit(le32_to_cpu(gdp->bg_inode_bitmap) - start, -+ bh->b_data); -+ for (bit = le32_to_cpu(gdp->bg_inode_table) - start, -+ bit_max = bit + sbi->s_itb_per_group; bit < bit_max; bit++) -+ ext3_set_bit(bit, bh->b_data); -+ } -+ -+ return free_blocks - sbi->s_itb_per_group - 2; -+} -+ - /* - * Read the bitmap for a given block_group, reading into the specified - * slot in the superblock's bitmap cache. -@@ -87,7 +157,19 @@ read_block_bitmap(struct super_block *sb - desc = ext3_get_group_desc (sb, block_group, NULL); - if (!desc) - goto error_out; -- bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); -+ if (desc->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) { -+ bh = sb_getblk(sb, le32_to_cpu(desc->bg_block_bitmap)); -+ if (!buffer_uptodate(bh)) { -+ lock_buffer(bh); -+ if (!buffer_uptodate(bh)) { -+ ext3_init_block_bitmap(sb, bh,block_group,desc); -+ set_buffer_uptodate(bh); -+ } -+ unlock_buffer(bh); -+ } -+ } else { -+ bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); -+ } - if (!bh) - ext3_error (sb, "read_block_bitmap", - "Cannot read block bitmap - " -@@ -432,6 +514,7 @@ do_more: - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + - dquot_freed_blocks); -+ gdp->bg_checksum = ext3_group_desc_csum(sbi, block_group, gdp); - spin_unlock(sb_bgl_lock(sbi, block_group)); - percpu_counter_mod(&sbi->s_freeblocks_counter, count); - -@@ -1372,8 +1455,11 @@ allocated: - ret_block, goal_hits, goal_attempts); - - spin_lock(sb_bgl_lock(sbi, group_no)); -+ if (gdp->bg_flags & cpu_to_le16(EXT3_BG_BLOCK_UNINIT)) -+ gdp->bg_flags &= cpu_to_le16(~EXT3_BG_BLOCK_UNINIT); - gdp->bg_free_blocks_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) - 1); -+ gdp->bg_checksum = ext3_group_desc_csum(sbi, group_no, gdp); - spin_unlock(sb_bgl_lock(sbi, group_no)); - percpu_counter_mod(&sbi->s_freeblocks_counter, -1); - diff --git a/ldiskfs/kernel_patches/patches/ext3-wantedi-2.6-suse.patch b/ldiskfs/kernel_patches/patches/ext3-wantedi-2.6-suse.patch deleted file mode 100644 index e38bedb..0000000 --- a/ldiskfs/kernel_patches/patches/ext3-wantedi-2.6-suse.patch +++ /dev/null @@ -1,192 +0,0 @@ -diff -urp linux-2.6.5-7.282.orig/fs/ext3/ialloc.c linux-2.6.5-7.282/fs/ext3/ialloc.c ---- linux-2.6.5-7.282.orig/fs/ext3/ialloc.c 2006-08-30 17:12:13.000000000 +0300 -+++ linux-2.6.5-7.282/fs/ext3/ialloc.c 2007-02-16 07:43:08.000000000 +0200 -@@ -420,7 +420,8 @@ static int find_group_other(struct super - * For other inodes, search forward from the parent directory's block - * group to find a free inode. - */ --struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) -+struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode, -+ unsigned long goal) - { - struct super_block *sb; - struct buffer_head *bitmap_bh = NULL; -@@ -448,6 +449,41 @@ struct inode *ext3_new_inode(handle_t *h - - sbi = EXT3_SB(sb); - es = sbi->s_es; -+ if (goal) { -+ group = (goal - 1) / EXT3_INODES_PER_GROUP(sb); -+ ino = (goal - 1) % EXT3_INODES_PER_GROUP(sb); -+ err = -EIO; -+ -+ gdp = ext3_get_group_desc(sb, group, &bh2); -+ if (!gdp) -+ goto fail; -+ -+ bitmap_bh = read_inode_bitmap (sb, group); -+ if (!bitmap_bh) -+ goto fail; -+ -+ BUFFER_TRACE(bh, "get_write_access"); -+ err = ext3_journal_get_write_access(handle, bitmap_bh); -+ if (err) goto fail; -+ -+ if (ext3_set_bit_atomic(sb_bgl_lock(sbi, group), -+ ino, bitmap_bh->b_data)) { -+ printk(KERN_ERR "goal inode %lu unavailable\n", goal); -+ /* Oh well, we tried. */ -+ goto continue_allocation; -+ } -+ -+ BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); -+ err = ext3_journal_dirty_metadata(handle, bitmap_bh); -+ if (err) goto fail; -+ -+ /* We've shortcircuited the allocation system successfully, -+ * now finish filling in the inode. -+ */ -+ goto got; -+ } -+ -+continue_allocation: - if (S_ISDIR(mode)) { - if (test_opt (sb, OLDALLOC)) - group = find_group_dir(sb, dir); -diff -urp linux-2.6.5-7.282.orig/fs/ext3/ioctl.c linux-2.6.5-7.282/fs/ext3/ioctl.c ---- linux-2.6.5-7.282.orig/fs/ext3/ioctl.c 2006-08-30 17:12:13.000000000 +0300 -+++ linux-2.6.5-7.282/fs/ext3/ioctl.c 2007-02-16 07:43:08.000000000 +0200 -@@ -25,6 +25,31 @@ int ext3_ioctl (struct inode * inode, st - ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg); - - switch (cmd) { -+ case EXT3_IOC_CREATE_INUM: { -+ char name[32]; -+ struct dentry *dchild, *dparent; -+ int rc = 0; -+ -+ dparent = list_entry(inode->i_dentry.next, struct dentry, -+ d_alias); -+ snprintf(name, sizeof name, "%lu", arg); -+ dchild = lookup_one_len(name, dparent, strlen(name)); -+ if (dchild->d_inode) { -+ printk(KERN_ERR "%*s/%lu already exists (ino %lu)\n", -+ dparent->d_name.len, dparent->d_name.name, arg, -+ dchild->d_inode->i_ino); -+ rc = -EEXIST; -+ } else { -+ dchild->d_fsdata = (void *)arg; -+ rc = vfs_create(inode, dchild, 0644, NULL); -+ if (rc) -+ printk(KERN_ERR "vfs_create: %d\n", rc); -+ else if (dchild->d_inode->i_ino != arg) -+ rc = -EEXIST; -+ } -+ dput(dchild); -+ return rc; -+ } - case EXT3_IOC_GETFLAGS: - flags = ei->i_flags & EXT3_FL_USER_VISIBLE; - return put_user(flags, (int *) arg); -diff -urp linux-2.6.5-7.282.orig/fs/ext3/namei.c linux-2.6.5-7.282/fs/ext3/namei.c ---- linux-2.6.5-7.282.orig/fs/ext3/namei.c 2006-08-30 17:12:34.000000000 +0300 -+++ linux-2.6.5-7.282/fs/ext3/namei.c 2007-02-16 07:46:13.000000000 +0200 -@@ -144,6 +144,14 @@ struct dx_map_entry - u32 offs; - }; - -+#define LVFS_DENTRY_PARAM_MAGIC 20070216UL -+struct lvfs_dentry_params -+{ -+ unsigned long p_inum; -+ void *p_ptr; -+ u32 magic; -+}; -+ - #ifdef CONFIG_EXT3_INDEX - static inline unsigned dx_get_block (struct dx_entry *entry); - static void dx_set_block (struct dx_entry *entry, unsigned value); -@@ -1625,6 +1633,20 @@ static int ext3_add_nondir(handle_t *han - return err; - } - -+static struct inode * ext3_new_inode_wantedi(handle_t *handle, struct inode *dir, -+ int mode, struct dentry *dentry) -+{ -+ unsigned long inum = 0; -+ -+ if (dentry->d_fsdata != NULL) { -+ struct lvfs_dentry_params *param = dentry->d_fsdata; -+ -+ if (param->magic == LVFS_DENTRY_PARAM_MAGIC) -+ inum = param->p_inum; -+ } -+ return ext3_new_inode(handle, dir, mode, inum); -+} -+ - /* - * By the time this is called, we already have created - * the directory cache entry for the new file, but it -@@ -1649,7 +1671,7 @@ retry: - if (IS_DIRSYNC(dir)) - handle->h_sync = 1; - -- inode = ext3_new_inode (handle, dir, mode); -+ inode = ext3_new_inode_wantedi (handle, dir, mode, dentry); - err = PTR_ERR(inode); - if (!IS_ERR(inode)) { - inode->i_op = &ext3_file_inode_operations; -@@ -1682,7 +1704,7 @@ retry: - if (IS_DIRSYNC(dir)) - handle->h_sync = 1; - -- inode = ext3_new_inode (handle, dir, mode); -+ inode = ext3_new_inode_wantedi (handle, dir, mode, dentry); - err = PTR_ERR(inode); - if (!IS_ERR(inode)) { - init_special_inode(inode, inode->i_mode, rdev); -@@ -1718,7 +1740,7 @@ retry: - if (IS_DIRSYNC(dir)) - handle->h_sync = 1; - -- inode = ext3_new_inode (handle, dir, S_IFDIR | mode); -+ inode = ext3_new_inode_wantedi (handle, dir, S_IFDIR | mode, dentry); - err = PTR_ERR(inode); - if (IS_ERR(inode)) - goto out_stop; -@@ -2113,7 +2135,7 @@ retry: - if (IS_DIRSYNC(dir)) - handle->h_sync = 1; - -- inode = ext3_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); -+ inode = ext3_new_inode_wantedi (handle, dir, S_IFLNK|S_IRWXUGO, dentry); - err = PTR_ERR(inode); - if (IS_ERR(inode)) - goto out_stop; -diff -urp linux-2.6.5-7.282.orig/include/linux/ext3_fs.h linux-2.6.5-7.282/include/linux/ext3_fs.h ---- linux-2.6.5-7.282.orig/include/linux/ext3_fs.h 2006-08-30 17:12:13.000000000 +0300 -+++ linux-2.6.5-7.282/include/linux/ext3_fs.h 2007-02-16 07:43:08.000000000 +0200 -@@ -203,6 +203,7 @@ struct ext3_group_desc - #define EXT3_IOC_SETFLAGS _IOW('f', 2, long) - #define EXT3_IOC_GETVERSION _IOR('f', 3, long) - #define EXT3_IOC_SETVERSION _IOW('f', 4, long) -+/* EXT3_IOC_CREATE_INUM at bottom of file (visible to kernel and user). */ - #define EXT3_IOC_GETVERSION_OLD _IOR('v', 1, long) - #define EXT3_IOC_SETVERSION_OLD _IOW('v', 2, long) - #ifdef CONFIG_JBD_DEBUG -@@ -712,7 +713,8 @@ extern int ext3fs_dirhash(const char *na - dx_hash_info *hinfo); - - /* ialloc.c */ --extern struct inode * ext3_new_inode (handle_t *, struct inode *, int); -+extern struct inode * ext3_new_inode (handle_t *, struct inode *, int, -+ unsigned long); - extern void ext3_free_inode (handle_t *, struct inode *); - extern struct inode * ext3_orphan_get (struct super_block *, unsigned long); - extern unsigned long ext3_count_free_inodes (struct super_block *); -@@ -797,4 +799,5 @@ extern struct inode_operations ext3_fast - - #endif /* __KERNEL__ */ - -+#define EXT3_IOC_CREATE_INUM _IOW('f', 5, long) - #endif /* _LINUX_EXT3_FS_H */ diff --git a/ldiskfs/kernel_patches/patches/iopen-2.6-suse.patch b/ldiskfs/kernel_patches/patches/iopen-2.6-suse.patch deleted file mode 100644 index 1c5e900..0000000 --- a/ldiskfs/kernel_patches/patches/iopen-2.6-suse.patch +++ /dev/null @@ -1,472 +0,0 @@ -Index: linux-stage/fs/ext3/Makefile -=================================================================== ---- linux-stage.orig/fs/ext3/Makefile 2005-02-25 14:31:53.151076368 +0200 -+++ linux-stage/fs/ext3/Makefile 2005-02-25 14:41:51.259150120 +0200 -@@ -4,7 +4,7 @@ - - obj-$(CONFIG_EXT3_FS) += ext3.o - --ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ -+ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o iopen.o \ - ioctl.o namei.o super.o symlink.o hash.o - - ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o -Index: linux-stage/fs/ext3/inode.c -=================================================================== ---- linux-stage.orig/fs/ext3/inode.c 2005-02-25 14:37:30.983718000 +0200 -+++ linux-stage/fs/ext3/inode.c 2005-02-25 14:47:42.069818792 +0200 -@@ -37,6 +37,7 @@ - #include - #include - #include "xattr.h" -+#include "iopen.h" - #include "acl.h" - - /* -@@ -2408,6 +2409,9 @@ - #endif - ei->i_rsv_window.rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - -+ if (ext3_iopen_get_inode(inode)) -+ return; -+ - if (ext3_get_inode_loc(inode, &iloc, 0)) - goto bad_inode; - bh = iloc.bh; -Index: linux-stage/fs/ext3/iopen.c -=================================================================== ---- linux-2.6.5-sles9.orig/fs/ext3/iopen.c 2003-01-30 13:24:37.000000000 +0300 -+++ linux-2.6.5-sles9/fs/ext3/iopen.c 2004-11-09 02:18:27.611913312 +0300 -@@ -0,0 +1,278 @@ -+/* -+ * linux/fs/ext3/iopen.c -+ * -+ * Special support for open by inode number -+ * -+ * Copyright (C) 2001 by Theodore Ts'o (tytso@alum.mit.edu). -+ * -+ * This file may be redistributed under the terms of the GNU General -+ * Public License. -+ * -+ * -+ * Invariants: -+ * - there is only ever a single DCACHE_NFSD_DISCONNECTED dentry alias -+ * for an inode at one time. -+ * - there are never both connected and DCACHE_NFSD_DISCONNECTED dentry -+ * aliases on an inode at the same time. -+ * -+ * If we have any connected dentry aliases for an inode, use one of those -+ * in iopen_lookup(). Otherwise, we instantiate a single NFSD_DISCONNECTED -+ * dentry for this inode, which thereafter will be found by the dcache -+ * when looking up this inode number in __iopen__, so we don't return here -+ * until it is gone. -+ * -+ * If we get an inode via a regular name lookup, then we "rename" the -+ * NFSD_DISCONNECTED dentry to the proper name and parent. This ensures -+ * existing users of the disconnected dentry will continue to use the same -+ * dentry as the connected users, and there will never be both kinds of -+ * dentry aliases at one time. -+ */ -+ -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include -+#include "iopen.h" -+ -+#ifndef assert -+#define assert(test) J_ASSERT(test) -+#endif -+ -+#define IOPEN_NAME_LEN 32 -+ -+/* -+ * This implements looking up an inode by number. -+ */ -+static struct dentry *iopen_lookup(struct inode * dir, struct dentry *dentry, -+ struct nameidata *nd) -+{ -+ struct inode *inode; -+ unsigned long ino; -+ struct list_head *lp; -+ struct dentry *alternate; -+ char buf[IOPEN_NAME_LEN]; -+ -+ if (dentry->d_name.len >= IOPEN_NAME_LEN) -+ return ERR_PTR(-ENAMETOOLONG); -+ -+ memcpy(buf, dentry->d_name.name, dentry->d_name.len); -+ buf[dentry->d_name.len] = 0; -+ -+ if (strcmp(buf, ".") == 0) -+ ino = dir->i_ino; -+ else if (strcmp(buf, "..") == 0) -+ ino = EXT3_ROOT_INO; -+ else -+ ino = simple_strtoul(buf, 0, 0); -+ -+ if ((ino != EXT3_ROOT_INO && -+ //ino != EXT3_ACL_IDX_INO && -+ //ino != EXT3_ACL_DATA_INO && -+ ino < EXT3_FIRST_INO(dir->i_sb)) || -+ ino > le32_to_cpu(EXT3_SB(dir->i_sb)->s_es->s_inodes_count)) -+ return ERR_PTR(-ENOENT); -+ -+ inode = iget(dir->i_sb, ino); -+ if (!inode) -+ return ERR_PTR(-EACCES); -+ if (is_bad_inode(inode)) { -+ iput(inode); -+ return ERR_PTR(-ENOENT); -+ } -+ -+ assert(list_empty(&dentry->d_alias)); /* d_instantiate */ -+ assert(d_unhashed(dentry)); /* d_rehash */ -+ -+ /* preferrably return a connected dentry */ -+ spin_lock(&dcache_lock); -+ list_for_each(lp, &inode->i_dentry) { -+ alternate = list_entry(lp, struct dentry, d_alias); -+ assert(!(alternate->d_flags & DCACHE_DISCONNECTED)); -+ } -+ -+ if (!list_empty(&inode->i_dentry)) { -+ alternate = list_entry(inode->i_dentry.next, -+ struct dentry, d_alias); -+ dget_locked(alternate); -+ spin_lock(&alternate->d_lock); -+ alternate->d_vfs_flags |= DCACHE_REFERENCED; -+ spin_unlock(&alternate->d_lock); -+ iput(inode); -+ spin_unlock(&dcache_lock); -+ return alternate; -+ } -+ dentry->d_flags |= DCACHE_DISCONNECTED; -+ -+ /* d_add(), but don't drop dcache_lock before adding dentry to inode */ -+ list_add(&dentry->d_alias, &inode->i_dentry); /* d_instantiate */ -+ dentry->d_inode = inode; -+ -+ __d_rehash(dentry, 0); /* d_rehash */ -+ spin_unlock(&dcache_lock); -+ -+ return NULL; -+} -+ -+#define do_switch(x,y) do { \ -+ __typeof__ (x) __tmp = x; \ -+ x = y; y = __tmp; } while (0) -+ -+static inline void switch_names(struct dentry *dentry, struct dentry *target) -+{ -+ const unsigned char *old_name, *new_name; -+ -+ memcpy(dentry->d_iname, target->d_iname, DNAME_INLINE_LEN); -+ old_name = target->d_name.name; -+ new_name = dentry->d_name.name; -+ if (old_name == target->d_iname) -+ old_name = dentry->d_iname; -+ if (new_name == dentry->d_iname) -+ new_name = target->d_iname; -+ target->d_name.name = new_name; -+ dentry->d_name.name = old_name; -+} -+ -+/* This function is spliced into ext3_lookup and does the move of a -+ * disconnected dentry (if it exists) to a connected dentry. -+ */ -+struct dentry *iopen_connect_dentry(struct dentry *dentry, struct inode *inode, -+ int rehash) -+{ -+ struct dentry *tmp, *goal = NULL; -+ struct list_head *lp; -+ -+ /* verify this dentry is really new */ -+ assert(dentry->d_inode == NULL); -+ assert(list_empty(&dentry->d_alias)); /* d_instantiate */ -+ if (rehash) -+ assert(d_unhashed(dentry)); /* d_rehash */ -+ assert(list_empty(&dentry->d_subdirs)); -+ -+ spin_lock(&dcache_lock); -+ if (!inode) -+ goto do_rehash; -+ -+ if (!test_opt(inode->i_sb, IOPEN)) -+ goto do_instantiate; -+ -+ /* preferrably return a connected dentry */ -+ list_for_each(lp, &inode->i_dentry) { -+ tmp = list_entry(lp, struct dentry, d_alias); -+ if (tmp->d_flags & DCACHE_DISCONNECTED) { -+ assert(tmp->d_alias.next == &inode->i_dentry); -+ assert(tmp->d_alias.prev == &inode->i_dentry); -+ goal = tmp; -+ dget_locked(goal); -+ break; -+ } -+ } -+ -+ if (!goal) -+ goto do_instantiate; -+ -+ /* Move the goal to the de hash queue */ -+ goal->d_flags &= ~DCACHE_DISCONNECTED; -+ security_d_instantiate(goal, inode); -+ __d_drop(dentry); -+ __d_rehash(dentry, 0); -+ __d_move(goal, dentry); -+ spin_unlock(&dcache_lock); -+ iput(inode); -+ -+ return goal; -+ -+ /* d_add(), but don't drop dcache_lock before adding dentry to inode */ -+do_instantiate: -+ list_add(&dentry->d_alias, &inode->i_dentry); /* d_instantiate */ -+ dentry->d_inode = inode; -+do_rehash: -+ if (rehash) -+ __d_rehash(dentry, 0); /* d_rehash */ -+ spin_unlock(&dcache_lock); -+ -+ return NULL; -+} -+ -+/* -+ * These are the special structures for the iopen pseudo directory. -+ */ -+ -+static struct inode_operations iopen_inode_operations = { -+ lookup: iopen_lookup, /* BKL held */ -+}; -+ -+static struct file_operations iopen_file_operations = { -+ read: generic_read_dir, -+}; -+ -+static int match_dentry(struct dentry *dentry, const char *name) -+{ -+ int len; -+ -+ len = strlen(name); -+ if (dentry->d_name.len != len) -+ return 0; -+ if (strncmp(dentry->d_name.name, name, len)) -+ return 0; -+ return 1; -+} -+ -+/* -+ * This function is spliced into ext3_lookup and returns 1 the file -+ * name is __iopen__ and dentry has been filled in appropriately. -+ */ -+int ext3_check_for_iopen(struct inode *dir, struct dentry *dentry) -+{ -+ struct inode *inode; -+ -+ if (dir->i_ino != EXT3_ROOT_INO || -+ !test_opt(dir->i_sb, IOPEN) || -+ !match_dentry(dentry, "__iopen__")) -+ return 0; -+ -+ inode = iget(dir->i_sb, EXT3_BAD_INO); -+ -+ if (!inode) -+ return 0; -+ d_add(dentry, inode); -+ return 1; -+} -+ -+/* -+ * This function is spliced into read_inode; it returns 1 if inode -+ * number is the one for /__iopen__, in which case the inode is filled -+ * in appropriately. Otherwise, this fuction returns 0. -+ */ -+int ext3_iopen_get_inode(struct inode *inode) -+{ -+ if (inode->i_ino != EXT3_BAD_INO) -+ return 0; -+ -+ inode->i_mode = S_IFDIR | S_IRUSR | S_IXUSR; -+ if (test_opt(inode->i_sb, IOPEN_NOPRIV)) -+ inode->i_mode |= 0777; -+ inode->i_uid = 0; -+ inode->i_gid = 0; -+ inode->i_nlink = 1; -+ inode->i_size = 4096; -+ inode->i_atime = CURRENT_TIME; -+ inode->i_ctime = CURRENT_TIME; -+ inode->i_mtime = CURRENT_TIME; -+ EXT3_I(inode)->i_dtime = 0; -+ inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size -+ * (for stat), not the fs block -+ * size */ -+ inode->i_blocks = 0; -+ inode->i_version = 1; -+ inode->i_generation = 0; -+ -+ inode->i_op = &iopen_inode_operations; -+ inode->i_fop = &iopen_file_operations; -+ inode->i_mapping->a_ops = 0; -+ -+ return 1; -+} -Index: linux-stage/fs/ext3/iopen.h -=================================================================== ---- linux-stage.orig/fs/ext3/iopen.h 2005-02-25 14:41:01.017787968 +0200 -+++ linux-stage/fs/ext3/iopen.h 2005-02-25 14:41:01.045783712 +0200 -@@ -0,0 +1,15 @@ -+/* -+ * iopen.h -+ * -+ * Special support for opening files by inode number. -+ * -+ * Copyright (C) 2001 by Theodore Ts'o (tytso@alum.mit.edu). -+ * -+ * This file may be redistributed under the terms of the GNU General -+ * Public License. -+ */ -+ -+extern int ext3_check_for_iopen(struct inode *dir, struct dentry *dentry); -+extern int ext3_iopen_get_inode(struct inode *inode); -+extern struct dentry *iopen_connect_dentry(struct dentry *dentry, -+ struct inode *inode, int rehash); -Index: linux-stage/fs/ext3/namei.c -=================================================================== ---- linux-stage.orig/fs/ext3/namei.c 2005-02-25 14:37:28.975023368 +0200 -+++ linux-stage/fs/ext3/namei.c 2005-02-25 14:46:43.090784968 +0200 -@@ -37,6 +37,7 @@ - #include - #include - #include "xattr.h" -+#include "iopen.h" - #include "acl.h" - - /* -@@ -980,6 +981,9 @@ - if (dentry->d_name.len > EXT3_NAME_LEN) - return ERR_PTR(-ENAMETOOLONG); - -+ if (ext3_check_for_iopen(dir, dentry)) -+ return NULL; -+ - bh = ext3_find_entry(dentry, &de); - inode = NULL; - if (bh) { -@@ -990,10 +994,8 @@ - if (!inode) - return ERR_PTR(-EACCES); - } -- if (inode) -- return d_splice_alias(inode, dentry); -- d_add(dentry, inode); -- return NULL; -+ -+ return iopen_connect_dentry(dentry, inode, 1); - } - - -@@ -2037,10 +2039,6 @@ - inode->i_nlink); - inode->i_version++; - inode->i_nlink = 0; -- /* There's no need to set i_disksize: the fact that i_nlink is -- * zero will ensure that the right thing happens during any -- * recovery. */ -- inode->i_size = 0; - ext3_orphan_add(handle, inode); - inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME; - ext3_mark_inode_dirty(handle, inode); -@@ -2163,6 +2161,23 @@ - return err; - } - -+/* Like ext3_add_nondir() except for call to iopen_connect_dentry */ -+static int ext3_add_link(handle_t *handle, struct dentry *dentry, -+ struct inode *inode) -+{ -+ int err = ext3_add_entry(handle, dentry, inode); -+ if (!err) { -+ err = ext3_mark_inode_dirty(handle, inode); -+ if (err == 0) { -+ dput(iopen_connect_dentry(dentry, inode, 0)); -+ return 0; -+ } -+ } -+ ext3_dec_count(handle, inode); -+ iput(inode); -+ return err; -+} -+ - static int ext3_link (struct dentry * old_dentry, - struct inode * dir, struct dentry *dentry) - { -@@ -2186,7 +2201,8 @@ - ext3_inc_count(handle, inode); - atomic_inc(&inode->i_count); - -- err = ext3_add_nondir(handle, dentry, inode); -+ err = ext3_add_link(handle, dentry, inode); -+ ext3_orphan_del(handle, inode); - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) - goto retry; -Index: linux-stage/fs/ext3/super.c -=================================================================== ---- linux-stage.orig/fs/ext3/super.c 2005-02-25 14:37:30.987717392 +0200 -+++ linux-stage/fs/ext3/super.c 2005-02-25 14:44:50.495901992 +0200 -@@ -586,6 +586,7 @@ - Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, - Opt_ignore, Opt_barrier, - Opt_err, -+ Opt_iopen, Opt_noiopen, Opt_iopen_nopriv, - }; - - static match_table_t tokens = { -@@ -633,6 +634,9 @@ - {Opt_ignore, "noquota"}, - {Opt_ignore, "quota"}, - {Opt_ignore, "usrquota"}, -+ {Opt_iopen, "iopen"}, -+ {Opt_noiopen, "noiopen"}, -+ {Opt_iopen_nopriv, "iopen_nopriv"}, - {Opt_barrier, "barrier=%u"}, - {Opt_err, NULL} - }; -@@ -914,6 +918,18 @@ - else - clear_opt(sbi->s_mount_opt, BARRIER); - break; -+ case Opt_iopen: -+ set_opt (sbi->s_mount_opt, IOPEN); -+ clear_opt (sbi->s_mount_opt, IOPEN_NOPRIV); -+ break; -+ case Opt_noiopen: -+ clear_opt (sbi->s_mount_opt, IOPEN); -+ clear_opt (sbi->s_mount_opt, IOPEN_NOPRIV); -+ break; -+ case Opt_iopen_nopriv: -+ set_opt (sbi->s_mount_opt, IOPEN); -+ set_opt (sbi->s_mount_opt, IOPEN_NOPRIV); -+ break; - case Opt_ignore: - break; - default: -Index: linux-stage/include/linux/ext3_fs.h -=================================================================== ---- linux-stage.orig/include/linux/ext3_fs.h 2005-02-25 14:37:28.977023064 +0200 -+++ linux-stage/include/linux/ext3_fs.h 2005-02-25 14:49:00.569884968 +0200 -@@ -355,6 +355,8 @@ - #define EXT3_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ - #define EXT3_MOUNT_RESERVATION 0x10000 /* Preallocation */ - #define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ -+#define EXT3_MOUNT_IOPEN 0x80000 /* Allow access via iopen */ -+#define EXT3_MOUNT_IOPEN_NOPRIV 0x100000/* Make iopen world-readable */ - - /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ - #ifndef _LINUX_EXT2_FS_H diff --git a/ldiskfs/kernel_patches/series/ldiskfs-2.6-suse.series b/ldiskfs/kernel_patches/series/ldiskfs-2.6-suse.series deleted file mode 100644 index c543c76..0000000 --- a/ldiskfs/kernel_patches/series/ldiskfs-2.6-suse.series +++ /dev/null @@ -1,22 +0,0 @@ -ext3-wantedi-2.6-suse.patch -iopen-2.6-suse.patch -export_symbols-ext3-2.6-suse.patch -ext3-map_inode_page-2.6-suse.patch -ext3-ea-in-inode-2.6-suse.patch -export-ext3-2.6-suse.patch -ext3-include-fixes-2.6-suse.patch -ext3-extents-2.6.5.patch -ext3-extents-fixes-2.6.9-rhel4.patch -ext3-extents-multiblock-directio-2.6.5-suse.patch -ext3-extents-search-2.6.9-rhel4.patch -ext3-extents-sanity-checks.patch -ext3-mballoc3-core.patch -ext3-mballoc3-suse.patch -ext3-nlinks-2.6.7.patch -ext3-rename-reserve-2.6-suse.patch -ext3-ialloc-2.6.patch -ext3-lookup-dotdot-2.6.9.patch -ext3-sector_t-overflow-2.6.5-suse.patch -ext3-check-jbd-errors-2.6.5.patch -ext3-uninit-2.6-suse.patch -ext3-nanosecond-2.6-suse.patch -- 1.8.3.1