linux/fs/ext4/ialloc.c

1107 lines
31 KiB
C
Raw Normal View History

/*
* linux/fs/ext4/ialloc.c
*
* Copyright (C) 1992, 1993, 1994, 1995
* Remy Card (card@masi.ibp.fr)
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie (Paris VI)
*
* BSD ufs-inspired inode and directory allocation by
* Stephen Tweedie (sct@redhat.com), 1993
* Big-endian to little-endian byte-swapping/bitmaps by
* David S. Miller (davem@caip.rutgers.edu), 1995
*/
#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd2.h>
#include <linux/stat.h>
#include <linux/string.h>
#include <linux/quotaops.h>
#include <linux/buffer_head.h>
#include <linux/random.h>
#include <linux/bitops.h>
#include <linux/blkdev.h>
#include <asm/byteorder.h>
#include "ext4.h"
#include "ext4_jbd2.h"
#include "xattr.h"
#include "acl.h"
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
#include "group.h"
/*
* ialloc.c contains the inodes allocation and deallocation routines
*/
/*
* The free inodes are managed by bitmaps. A file system contains several
* blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap
* block for inodes, N blocks for the inode table and data blocks.
*
* The file system contains group descriptors which are located after the
* super block. Each descriptor contains the number of the bitmap block and
* the free blocks count in the block.
*/
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
/*
* 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.
*/
void mark_bitmap_end(int start_bit, int end_bit, char *bitmap)
{
int i;
if (start_bit >= end_bit)
return;
ext4_debug("mark end bits +%d through +%d used\n", start_bit, end_bit);
for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++)
ext4_set_bit(i, bitmap);
if (i < end_bit)
memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3);
}
/* Initializes an uninitialized inode bitmap */
unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh,
ext4_group_t block_group,
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
struct ext4_group_desc *gdp)
{
struct ext4_sb_info *sbi = EXT4_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 (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
ext4_error(sb, __func__, "Checksum bad for group %u",
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
block_group);
ext4_free_blks_set(sb, gdp, 0);
ext4_free_inodes_set(sb, gdp, 0);
ext4_itable_unused_set(sb, gdp, 0);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
memset(bh->b_data, 0xff, sb->s_blocksize);
return 0;
}
memset(bh->b_data, 0, (EXT4_INODES_PER_GROUP(sb) + 7) / 8);
mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), sb->s_blocksize * 8,
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
bh->b_data);
return EXT4_INODES_PER_GROUP(sb);
}
/*
* Read the inode allocation bitmap for a given block_group, reading
* into the specified slot in the superblock's bitmap cache.
*
* Return buffer_head of bitmap on success or NULL.
*/
static struct buffer_head *
ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
{
struct ext4_group_desc *desc;
struct buffer_head *bh = NULL;
ext4_fsblk_t bitmap_blk;
desc = ext4_get_group_desc(sb, block_group, NULL);
if (!desc)
return NULL;
bitmap_blk = ext4_inode_bitmap(sb, desc);
bh = sb_getblk(sb, bitmap_blk);
if (unlikely(!bh)) {
ext4_error(sb, __func__,
"Cannot read inode bitmap - "
"block_group = %u, inode_bitmap = %llu",
block_group, bitmap_blk);
return NULL;
}
if (bitmap_uptodate(bh))
return bh;
lock_buffer(bh);
if (bitmap_uptodate(bh)) {
unlock_buffer(bh);
return bh;
}
spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
ext4_init_inode_bitmap(sb, bh, block_group, desc);
set_bitmap_uptodate(bh);
set_buffer_uptodate(bh);
spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
unlock_buffer(bh);
return bh;
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
}
spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (buffer_uptodate(bh)) {
/*
* if not uninit if bh is uptodate,
* bitmap is also uptodate
*/
set_bitmap_uptodate(bh);
unlock_buffer(bh);
return bh;
}
/*
* submit the buffer_head for read. We can
* safely mark the bitmap as uptodate now.
* We do it here so the bitmap uptodate bit
* get set with buffer lock held.
*/
set_bitmap_uptodate(bh);
if (bh_submit_read(bh) < 0) {
put_bh(bh);
ext4_error(sb, __func__,
"Cannot read inode bitmap - "
"block_group = %u, inode_bitmap = %llu",
block_group, bitmap_blk);
return NULL;
}
return bh;
}
/*
* NOTE! When we get the inode, we're the only people
* that have access to it, and as such there are no
* race conditions we have to worry about. The inode
* is not on the hash-lists, and it cannot be reached
* through the filesystem because the directory entry
* has been deleted earlier.
*
* HOWEVER: we must make sure that we get no aliases,
* which means that we have to call "clear_inode()"
* _before_ we mark the inode not in use in the inode
* bitmaps. Otherwise a newly created file might use
* the same inode number (not actually the same pointer
* though), and then we'd have two inodes sharing the
* same inode number and space on the harddisk.
*/
void ext4_free_inode(handle_t *handle, struct inode *inode)
{
struct super_block *sb = inode->i_sb;
int is_directory;
unsigned long ino;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
ext4_group_t block_group;
unsigned long bit;
struct ext4_group_desc *gdp;
struct ext4_super_block *es;
struct ext4_sb_info *sbi;
int fatal = 0, err, count, cleared;
ext4_group_t flex_group;
if (atomic_read(&inode->i_count) > 1) {
printk(KERN_ERR "ext4_free_inode: inode has count=%d\n",
atomic_read(&inode->i_count));
return;
}
if (inode->i_nlink) {
printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n",
inode->i_nlink);
return;
}
if (!sb) {
printk(KERN_ERR "ext4_free_inode: inode on "
"nonexistent device\n");
return;
}
sbi = EXT4_SB(sb);
ino = inode->i_ino;
ext4_debug("freeing inode %lu\n", ino);
trace_mark(ext4_free_inode,
"dev %s ino %lu mode %d uid %lu gid %lu bocks %llu",
sb->s_id, inode->i_ino, inode->i_mode,
(unsigned long) inode->i_uid, (unsigned long) inode->i_gid,
(unsigned long long) inode->i_blocks);
/*
* Note: we must free any quota before locking the superblock,
* as writing the quota to disk may need the lock as well.
*/
DQUOT_INIT(inode);
ext4_xattr_delete_inode(handle, inode);
DQUOT_FREE_INODE(inode);
DQUOT_DROP(inode);
is_directory = S_ISDIR(inode->i_mode);
/* Do this BEFORE marking the inode not in use or returning an error */
clear_inode(inode);
es = EXT4_SB(sb)->s_es;
if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
ext4_error(sb, "ext4_free_inode",
"reserved or nonexistent inode %lu", ino);
goto error_return;
}
block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
BUFFER_TRACE(bitmap_bh, "get_write_access");
fatal = ext4_journal_get_write_access(handle, bitmap_bh);
if (fatal)
goto error_return;
/* Ok, now we can actually update the inode bitmaps.. */
spin_lock(sb_bgl_lock(sbi, block_group));
cleared = ext4_clear_bit(bit, bitmap_bh->b_data);
spin_unlock(sb_bgl_lock(sbi, block_group));
if (!cleared)
ext4_error(sb, "ext4_free_inode",
"bit already cleared for inode %lu", ino);
else {
gdp = ext4_get_group_desc(sb, block_group, &bh2);
BUFFER_TRACE(bh2, "get_write_access");
fatal = ext4_journal_get_write_access(handle, bh2);
if (fatal) goto error_return;
if (gdp) {
spin_lock(sb_bgl_lock(sbi, block_group));
count = ext4_free_inodes_count(sb, gdp) + 1;
ext4_free_inodes_set(sb, gdp, count);
if (is_directory) {
count = ext4_used_dirs_count(sb, gdp) - 1;
ext4_used_dirs_set(sb, gdp, count);
}
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
gdp->bg_checksum = ext4_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)
percpu_counter_dec(&sbi->s_dirs_counter);
if (sbi->s_log_groups_per_flex) {
flex_group = ext4_flex_group(sbi, block_group);
spin_lock(sb_bgl_lock(sbi, flex_group));
sbi->s_flex_groups[flex_group].free_inodes++;
spin_unlock(sb_bgl_lock(sbi, flex_group));
}
}
BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, NULL, bh2);
if (!fatal) fatal = err;
}
BUFFER_TRACE(bitmap_bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
if (!fatal)
fatal = err;
sb->s_dirt = 1;
error_return:
brelse(bitmap_bh);
ext4_std_error(sb, fatal);
}
/*
* There are two policies for allocating an inode. If the new inode is
* a directory, then a forward search is made for a block group with both
* free space and a low directory-to-inode ratio; if that fails, then of
* the groups with above-average free space, that group with the fewest
* directories already is chosen.
*
* For other inodes, search forward from the parent directory\'s block
* group to find a free inode.
*/
static int find_group_dir(struct super_block *sb, struct inode *parent,
ext4_group_t *best_group)
{
ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
unsigned int freei, avefreei;
struct ext4_group_desc *desc, *best_desc = NULL;
ext4_group_t group;
int ret = -1;
freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter);
avefreei = freei / ngroups;
for (group = 0; group < ngroups; group++) {
desc = ext4_get_group_desc(sb, group, NULL);
if (!desc || !ext4_free_inodes_count(sb, desc))
continue;
if (ext4_free_inodes_count(sb, desc) < avefreei)
continue;
if (!best_desc ||
(ext4_free_blks_count(sb, desc) >
ext4_free_blks_count(sb, best_desc))) {
*best_group = group;
best_desc = desc;
ret = 0;
}
}
return ret;
}
#define free_block_ratio 10
static int find_group_flex(struct super_block *sb, struct inode *parent,
ext4_group_t *best_group)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct ext4_group_desc *desc;
struct buffer_head *bh;
struct flex_groups *flex_group = sbi->s_flex_groups;
ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
ext4_group_t parent_fbg_group = ext4_flex_group(sbi, parent_group);
ext4_group_t ngroups = sbi->s_groups_count;
int flex_size = ext4_flex_bg_size(sbi);
ext4_group_t best_flex = parent_fbg_group;
int blocks_per_flex = sbi->s_blocks_per_group * flex_size;
int flexbg_free_blocks;
int flex_freeb_ratio;
ext4_group_t n_fbg_groups;
ext4_group_t i;
n_fbg_groups = (sbi->s_groups_count + flex_size - 1) >>
sbi->s_log_groups_per_flex;
find_close_to_parent:
flexbg_free_blocks = flex_group[best_flex].free_blocks;
flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;
if (flex_group[best_flex].free_inodes &&
flex_freeb_ratio > free_block_ratio)
goto found_flexbg;
if (best_flex && best_flex == parent_fbg_group) {
best_flex--;
goto find_close_to_parent;
}
for (i = 0; i < n_fbg_groups; i++) {
if (i == parent_fbg_group || i == parent_fbg_group - 1)
continue;
flexbg_free_blocks = flex_group[i].free_blocks;
flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;
if (flex_freeb_ratio > free_block_ratio &&
flex_group[i].free_inodes) {
best_flex = i;
goto found_flexbg;
}
if (flex_group[best_flex].free_inodes == 0 ||
(flex_group[i].free_blocks >
flex_group[best_flex].free_blocks &&
flex_group[i].free_inodes))
best_flex = i;
}
if (!flex_group[best_flex].free_inodes ||
!flex_group[best_flex].free_blocks)
return -1;
found_flexbg:
for (i = best_flex * flex_size; i < ngroups &&
i < (best_flex + 1) * flex_size; i++) {
desc = ext4_get_group_desc(sb, i, &bh);
if (ext4_free_inodes_count(sb, desc)) {
*best_group = i;
goto out;
}
}
return -1;
out:
return 0;
}
/*
* Orlov's allocator for directories.
*
* We always try to spread first-level directories.
*
* If there are blockgroups with both free inodes and free blocks counts
* not worse than average we return one with smallest directory count.
* Otherwise we simply return a random group.
*
* For the rest rules look so:
*
* It's OK to put directory into a group unless
* it has too many directories already (max_dirs) or
* it has too few free inodes left (min_inodes) or
* it has too few free blocks left (min_blocks) or
* it's already running too large debt (max_debt).
* Parent's group is preferred, if it doesn't satisfy these
* conditions we search cyclically through the rest. If none
* of the groups look good we just look for a group with more
* free inodes than average (starting at parent's group).
*
* Debt is incremented each time we allocate a directory and decremented
* when we allocate an inode, within 0--255.
*/
#define INODE_COST 64
#define BLOCK_COST 256
static int find_group_orlov(struct super_block *sb, struct inode *parent,
ext4_group_t *group)
{
ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct ext4_super_block *es = sbi->s_es;
ext4_group_t ngroups = sbi->s_groups_count;
int inodes_per_group = EXT4_INODES_PER_GROUP(sb);
unsigned int freei, avefreei;
ext4_fsblk_t freeb, avefreeb;
ext4_fsblk_t blocks_per_dir;
unsigned int ndirs;
int max_debt, max_dirs, min_inodes;
ext4_grpblk_t min_blocks;
ext4_group_t i;
struct ext4_group_desc *desc;
freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
avefreei = freei / ngroups;
freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
avefreeb = freeb;
do_div(avefreeb, ngroups);
ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);
if ((parent == sb->s_root->d_inode) ||
(EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL)) {
int best_ndir = inodes_per_group;
ext4_group_t grp;
int ret = -1;
get_random_bytes(&grp, sizeof(grp));
parent_group = (unsigned)grp % ngroups;
for (i = 0; i < ngroups; i++) {
grp = (parent_group + i) % ngroups;
desc = ext4_get_group_desc(sb, grp, NULL);
if (!desc || !ext4_free_inodes_count(sb, desc))
continue;
if (ext4_used_dirs_count(sb, desc) >= best_ndir)
continue;
if (ext4_free_inodes_count(sb, desc) < avefreei)
continue;
if (ext4_free_blks_count(sb, desc) < avefreeb)
continue;
*group = grp;
ret = 0;
best_ndir = ext4_used_dirs_count(sb, desc);
}
if (ret == 0)
return ret;
goto fallback;
}
blocks_per_dir = ext4_blocks_count(es) - freeb;
do_div(blocks_per_dir, ndirs);
max_dirs = ndirs / ngroups + inodes_per_group / 16;
min_inodes = avefreei - inodes_per_group / 4;
min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb) / 4;
max_debt = EXT4_BLOCKS_PER_GROUP(sb);
max_debt /= max_t(int, blocks_per_dir, BLOCK_COST);
if (max_debt * INODE_COST > inodes_per_group)
max_debt = inodes_per_group / INODE_COST;
if (max_debt > 255)
max_debt = 255;
if (max_debt == 0)
max_debt = 1;
for (i = 0; i < ngroups; i++) {
*group = (parent_group + i) % ngroups;
desc = ext4_get_group_desc(sb, *group, NULL);
if (!desc || !ext4_free_inodes_count(sb, desc))
continue;
if (ext4_used_dirs_count(sb, desc) >= max_dirs)
continue;
if (ext4_free_inodes_count(sb, desc) < min_inodes)
continue;
if (ext4_free_blks_count(sb, desc) < min_blocks)
continue;
return 0;
}
fallback:
for (i = 0; i < ngroups; i++) {
*group = (parent_group + i) % ngroups;
desc = ext4_get_group_desc(sb, *group, NULL);
if (desc && ext4_free_inodes_count(sb, desc) &&
ext4_free_inodes_count(sb, desc) >= avefreei)
return 0;
}
if (avefreei) {
/*
* The free-inodes counter is approximate, and for really small
* filesystems the above test can fail to find any blockgroups
*/
avefreei = 0;
goto fallback;
}
return -1;
}
static int find_group_other(struct super_block *sb, struct inode *parent,
ext4_group_t *group)
{
ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
struct ext4_group_desc *desc;
ext4_group_t i;
/*
* Try to place the inode in its parent directory
*/
*group = parent_group;
desc = ext4_get_group_desc(sb, *group, NULL);
if (desc && ext4_free_inodes_count(sb, desc) &&
ext4_free_blks_count(sb, desc))
return 0;
/*
* We're going to place this inode in a different blockgroup from its
* parent. We want to cause files in a common directory to all land in
* the same blockgroup. But we want files which are in a different
* directory which shares a blockgroup with our parent to land in a
* different blockgroup.
*
* So add our directory's i_ino into the starting point for the hash.
*/
*group = (*group + parent->i_ino) % ngroups;
/*
* Use a quadratic hash to find a group with a free inode and some free
* blocks.
*/
for (i = 1; i < ngroups; i <<= 1) {
*group += i;
if (*group >= ngroups)
*group -= ngroups;
desc = ext4_get_group_desc(sb, *group, NULL);
if (desc && ext4_free_inodes_count(sb, desc) &&
ext4_free_blks_count(sb, desc))
return 0;
}
/*
* That failed: try linear search for a free inode, even if that group
* has no free blocks.
*/
*group = parent_group;
for (i = 0; i < ngroups; i++) {
if (++*group >= ngroups)
*group = 0;
desc = ext4_get_group_desc(sb, *group, NULL);
if (desc && ext4_free_inodes_count(sb, desc))
return 0;
}
return -1;
}
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
/*
* claim the inode from the inode bitmap. If the group
* is uninit we need to take the groups's sb_bgl_lock
* and clear the uninit flag. The inode bitmap update
* and group desc uninit flag clear should be done
* after holding sb_bgl_lock so that ext4_read_inode_bitmap
* doesn't race with the ext4_claim_inode
*/
static int ext4_claim_inode(struct super_block *sb,
struct buffer_head *inode_bitmap_bh,
unsigned long ino, ext4_group_t group, int mode)
{
int free = 0, retval = 0, count;
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct ext4_group_desc *gdp = ext4_get_group_desc(sb, group, NULL);
spin_lock(sb_bgl_lock(sbi, group));
if (ext4_set_bit(ino, inode_bitmap_bh->b_data)) {
/* not a free inode */
retval = 1;
goto err_ret;
}
ino++;
if ((group == 0 && ino < EXT4_FIRST_INO(sb)) ||
ino > EXT4_INODES_PER_GROUP(sb)) {
spin_unlock(sb_bgl_lock(sbi, group));
ext4_error(sb, __func__,
"reserved inode or inode > inodes count - "
"block_group = %u, inode=%lu", group,
ino + group * EXT4_INODES_PER_GROUP(sb));
return 1;
}
/* If we didn't allocate from within the initialized part of the inode
* table then we need to initialize up to this inode. */
if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) {
if (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT);
/* When marking the block group with
* ~EXT4_BG_INODE_UNINIT we don't want to depend
* on the value of bg_itable_unused even though
* mke2fs could have initialized the same for us.
* Instead we calculated the value below
*/
free = 0;
} else {
free = EXT4_INODES_PER_GROUP(sb) -
ext4_itable_unused_count(sb, gdp);
}
/*
* Check the relative inode number against the last used
* relative inode number in this group. if it is greater
* we need to update the bg_itable_unused count
*
*/
if (ino > free)
ext4_itable_unused_set(sb, gdp,
(EXT4_INODES_PER_GROUP(sb) - ino));
}
count = ext4_free_inodes_count(sb, gdp) - 1;
ext4_free_inodes_set(sb, gdp, count);
if (S_ISDIR(mode)) {
count = ext4_used_dirs_count(sb, gdp) + 1;
ext4_used_dirs_set(sb, gdp, count);
}
gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp);
err_ret:
spin_unlock(sb_bgl_lock(sbi, group));
return retval;
}
/*
* There are two policies for allocating an inode. If the new inode is
* a directory, then a forward search is made for a block group with both
* free space and a low directory-to-inode ratio; if that fails, then of
* the groups with above-average free space, that group with the fewest
* directories already is chosen.
*
* For other inodes, search forward from the parent directory's block
* group to find a free inode.
*/
struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode)
{
struct super_block *sb;
struct buffer_head *inode_bitmap_bh = NULL;
struct buffer_head *group_desc_bh;
ext4_group_t group = 0;
unsigned long ino = 0;
struct inode *inode;
struct ext4_group_desc *gdp = NULL;
struct ext4_super_block *es;
struct ext4_inode_info *ei;
struct ext4_sb_info *sbi;
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
int ret2, err = 0;
struct inode *ret;
ext4_group_t i;
int free = 0;
ext4_group_t flex_group;
/* Cannot create files in a deleted directory */
if (!dir || !dir->i_nlink)
return ERR_PTR(-EPERM);
sb = dir->i_sb;
trace_mark(ext4_request_inode, "dev %s dir %lu mode %d", sb->s_id,
dir->i_ino, mode);
inode = new_inode(sb);
if (!inode)
return ERR_PTR(-ENOMEM);
ei = EXT4_I(inode);
sbi = EXT4_SB(sb);
es = sbi->s_es;
if (sbi->s_log_groups_per_flex) {
ret2 = find_group_flex(sb, dir, &group);
if (ret2 == -1) {
ret2 = find_group_other(sb, dir, &group);
if (ret2 == 0 && printk_ratelimit())
printk(KERN_NOTICE "ext4: find_group_flex "
"failed, fallback succeeded dir %lu\n",
dir->i_ino);
}
goto got_group;
}
if (S_ISDIR(mode)) {
if (test_opt(sb, OLDALLOC))
ret2 = find_group_dir(sb, dir, &group);
else
ret2 = find_group_orlov(sb, dir, &group);
} else
ret2 = find_group_other(sb, dir, &group);
got_group:
err = -ENOSPC;
if (ret2 == -1)
goto out;
for (i = 0; i < sbi->s_groups_count; i++) {
err = -EIO;
gdp = ext4_get_group_desc(sb, group, &group_desc_bh);
if (!gdp)
goto fail;
brelse(inode_bitmap_bh);
inode_bitmap_bh = ext4_read_inode_bitmap(sb, group);
if (!inode_bitmap_bh)
goto fail;
ino = 0;
repeat_in_this_group:
ino = ext4_find_next_zero_bit((unsigned long *)
inode_bitmap_bh->b_data,
EXT4_INODES_PER_GROUP(sb), ino);
if (ino < EXT4_INODES_PER_GROUP(sb)) {
BUFFER_TRACE(inode_bitmap_bh, "get_write_access");
err = ext4_journal_get_write_access(handle,
inode_bitmap_bh);
if (err)
goto fail;
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
BUFFER_TRACE(group_desc_bh, "get_write_access");
err = ext4_journal_get_write_access(handle,
group_desc_bh);
if (err)
goto fail;
if (!ext4_claim_inode(sb, inode_bitmap_bh,
ino, group, mode)) {
/* we won it */
BUFFER_TRACE(inode_bitmap_bh,
"call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle,
inode,
inode_bitmap_bh);
if (err)
goto fail;
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
/* zero bit is inode number 1*/
ino++;
goto got;
}
/* we lost it */
ext4_handle_release_buffer(handle, inode_bitmap_bh);
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
ext4_handle_release_buffer(handle, group_desc_bh);
if (++ino < EXT4_INODES_PER_GROUP(sb))
goto repeat_in_this_group;
}
/*
* This case is possible in concurrent environment. It is very
* rare. We cannot repeat the find_group_xxx() call because
* that will simply return the same blockgroup, because the
* group descriptor metadata has not yet been updated.
* So we just go onto the next blockgroup.
*/
if (++group == sbi->s_groups_count)
group = 0;
}
err = -ENOSPC;
goto out;
got:
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
/* We may have to initialize the block bitmap if it isn't already */
if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM) &&
gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
struct buffer_head *block_bitmap_bh;
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
block_bitmap_bh = ext4_read_block_bitmap(sb, group);
BUFFER_TRACE(block_bitmap_bh, "get block bitmap access");
err = ext4_journal_get_write_access(handle, block_bitmap_bh);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
if (err) {
brelse(block_bitmap_bh);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
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(EXT4_BG_BLOCK_UNINIT)) {
free = ext4_free_blocks_after_init(sb, group, gdp);
gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
ext4_free_blks_set(sb, gdp, free);
gdp->bg_checksum = ext4_group_desc_csum(sbi, group,
gdp);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
}
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_bitmap_bh, "dirty block bitmap");
err = ext4_handle_dirty_metadata(handle,
NULL, block_bitmap_bh);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
}
brelse(block_bitmap_bh);
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
if (err)
goto fail;
}
BUFFER_TRACE(group_desc_bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, NULL, group_desc_bh);
ext4: Fix the race between read_inode_bitmap() and ext4_new_inode() We need to make sure we update the inode bitmap and clear EXT4_BG_INODE_UNINIT flag with sb_bgl_lock held, since ext4_read_inode_bitmap() looks at EXT4_BG_INODE_UNINIT to decide whether to initialize the inode bitmap each time it is called. (introduced by commit c806e68f.) ext4_read_inode_bitmap does: spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { ext4_init_inode_bitmap(sb, bh, block_group, desc); and ext4_new_inode does if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, inode_bitmap_bh->b_data)) ...... ... spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT); i.e., on allocation we update the bitmap then we take the sb_bgl_lock and clear the EXT4_BG_INODE_UNINIT flag. What can happen is a parallel ext4_read_inode_bitmap can zero out the bitmap in between the above ext4_set_bit_atomic and spin_lock(sb_bg_lock..) The race results in below user visible errors EXT4-fs error (device sdb1): ext4_free_inode: bit already cleared for inode 168449 EXT4-fs warning (device sdb1): ext4_unlink: Deleting nonexistent file ... EXT4-fs warning (device sdb1): ext4_rmdir: empty directory has too many links ... # ls -al /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71 ls: /mnt/tmp/f/p369/d3/d6/d39/db2/dee/d10f/d3f/l71: Stale NFS file handle Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2009-01-06 10:38:14 +08:00
if (err)
goto fail;
percpu_counter_dec(&sbi->s_freeinodes_counter);
if (S_ISDIR(mode))
percpu_counter_inc(&sbi->s_dirs_counter);
sb->s_dirt = 1;
if (sbi->s_log_groups_per_flex) {
flex_group = ext4_flex_group(sbi, group);
spin_lock(sb_bgl_lock(sbi, flex_group));
sbi->s_flex_groups[flex_group].free_inodes--;
spin_unlock(sb_bgl_lock(sbi, flex_group));
}
inode->i_uid = current_fsuid();
if (test_opt(sb, GRPID))
inode->i_gid = dir->i_gid;
else if (dir->i_mode & S_ISGID) {
inode->i_gid = dir->i_gid;
if (S_ISDIR(mode))
mode |= S_ISGID;
} else
inode->i_gid = current_fsgid();
inode->i_mode = mode;
Ext4: Uninitialized Block Groups In pass1 of e2fsck, every inode table in the fileystem is scanned and checked, regardless of whether it is in use. This is this the most time consuming part of the filesystem check. The unintialized block group feature can greatly reduce e2fsck time by eliminating checking of uninitialized inodes. With this feature, there is a a high water mark of used inodes for each block group. 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. The feature is enabled through a mkfs option mke2fs /dev/ -O uninit_groups A patch adding support for uninitialized block groups to e2fsprogs tools has been posted to the linux-ext4 mailing list. The patches have been stress tested with fsstress and fsx. In performance tests testing e2fsck time, we have seen that e2fsck time on ext3 grows linearly with the total number of inodes in the filesytem. In ext4 with the uninitialized block groups feature, the e2fsck time is constant, based solely on the number of used inodes rather than the total inode count. Since typical ext4 filesystems only use 1-10% of their inodes, this feature can greatly reduce e2fsck time for users. With performance improvement of 2-20 times, depending on how full the filesystem is. The attached graph shows the major improvements in e2fsck times in filesystems with a large total inode count, but few inodes in use. In each group descriptor if we have EXT4_BG_INODE_UNINIT set in bg_flags: Inode table is not initialized/used in this group. So we can skip the consistency check during fsck. EXT4_BG_BLOCK_UNINIT set in bg_flags: No block in the group is used. So we can skip the block bitmap verification for this group. We also add two new fields to group descriptor as a part of uninitialized group patch. __le16 bg_itable_unused; /* Unused inodes count */ __le16 bg_checksum; /* crc16(sb_uuid+group+desc) */ bg_itable_unused: If we have EXT4_BG_INODE_UNINIT not set in bg_flags then bg_itable_unused will give the offset within the inode table till the inodes are used. This can be used by fsck to skip list of inodes that are marked unused. bg_checksum: Now that we depend on bg_flags and bg_itable_unused to determine the block and inode usage, we need to make sure group descriptor is not corrupt. We add checksum to group descriptor to detect corruption. If the descriptor is found to be corrupt, we mark all the blocks and inodes in the group used. Signed-off-by: Avantika Mathur <mathur@us.ibm.com> Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
2007-10-17 06:38:25 +08:00
inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
/* This is the optimal IO size (for stat), not the fs block size */
inode->i_blocks = 0;
inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime =
ext4_current_time(inode);
memset(ei->i_data, 0, sizeof(ei->i_data));
ei->i_dir_start_lookup = 0;
ei->i_disksize = 0;
/*
* Don't inherit extent flag from directory. We set extent flag on
* newly created directory and file only if -o extent mount option is
* specified
*/
ei->i_flags = EXT4_I(dir)->i_flags & ~(EXT4_INDEX_FL|EXT4_EXTENTS_FL);
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL);
/* dirsync only applies to directories */
if (!S_ISDIR(mode))
ei->i_flags &= ~EXT4_DIRSYNC_FL;
ei->i_file_acl = 0;
ei->i_dtime = 0;
ei->i_block_group = group;
ext4_set_inode_flags(inode);
if (IS_DIRSYNC(inode))
ext4_handle_sync(handle);
if (insert_inode_locked(inode) < 0) {
err = -EINVAL;
goto fail_drop;
}
spin_lock(&sbi->s_next_gen_lock);
inode->i_generation = sbi->s_next_generation++;
spin_unlock(&sbi->s_next_gen_lock);
ei->i_state = EXT4_STATE_NEW;
ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize;
ret = inode;
if (DQUOT_ALLOC_INODE(inode)) {
err = -EDQUOT;
goto fail_drop;
}
err = ext4_init_acl(handle, inode, dir);
if (err)
goto fail_free_drop;
err = ext4_init_security(handle, inode, dir);
if (err)
goto fail_free_drop;
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
/* set extent flag only for directory, file and normal symlink*/
if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) {
EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
ext4_ext_tree_init(handle, inode);
}
}
err = ext4_mark_inode_dirty(handle, inode);
if (err) {
ext4_std_error(sb, err);
goto fail_free_drop;
}
ext4_debug("allocating inode %lu\n", inode->i_ino);
trace_mark(ext4_allocate_inode, "dev %s ino %lu dir %lu mode %d",
sb->s_id, inode->i_ino, dir->i_ino, mode);
goto really_out;
fail:
ext4_std_error(sb, err);
out:
iput(inode);
ret = ERR_PTR(err);
really_out:
brelse(inode_bitmap_bh);
return ret;
fail_free_drop:
DQUOT_FREE_INODE(inode);
fail_drop:
DQUOT_DROP(inode);
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
unlock_new_inode(inode);
iput(inode);
brelse(inode_bitmap_bh);
return ERR_PTR(err);
}
/* Verify that we are loading a valid orphan from disk */
struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
{
unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
ext4_group_t block_group;
int bit;
struct buffer_head *bitmap_bh;
struct inode *inode = NULL;
long err = -EIO;
/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
ext4_warning(sb, __func__,
"bad orphan ino %lu! e2fsck was run?", ino);
goto error;
}
block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
if (!bitmap_bh) {
ext4_warning(sb, __func__,
"inode bitmap error for orphan %lu", ino);
goto error;
}
/* Having the inode bit set should be a 100% indicator that this
* is a valid orphan (no e2fsck run on fs). Orphans also include
* inodes that were being truncated, so we can't check i_nlink==0.
*/
if (!ext4_test_bit(bit, bitmap_bh->b_data))
goto bad_orphan;
inode = ext4_iget(sb, ino);
if (IS_ERR(inode))
goto iget_failed;
/*
* If the orphans has i_nlinks > 0 then it should be able to be
* truncated, otherwise it won't be removed from the orphan list
* during processing and an infinite loop will result.
*/
if (inode->i_nlink && !ext4_can_truncate(inode))
goto bad_orphan;
if (NEXT_ORPHAN(inode) > max_ino)
goto bad_orphan;
brelse(bitmap_bh);
return inode;
iget_failed:
err = PTR_ERR(inode);
inode = NULL;
bad_orphan:
ext4_warning(sb, __func__,
"bad orphan inode %lu! e2fsck was run?", ino);
printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n",
bit, (unsigned long long)bitmap_bh->b_blocknr,
ext4_test_bit(bit, bitmap_bh->b_data));
printk(KERN_NOTICE "inode=%p\n", inode);
if (inode) {
printk(KERN_NOTICE "is_bad_inode(inode)=%d\n",
is_bad_inode(inode));
printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
NEXT_ORPHAN(inode));
printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink);
/* Avoid freeing blocks if we got a bad deleted inode */
if (inode->i_nlink == 0)
inode->i_blocks = 0;
iput(inode);
}
brelse(bitmap_bh);
error:
return ERR_PTR(err);
}
unsigned long ext4_count_free_inodes(struct super_block *sb)
{
unsigned long desc_count;
struct ext4_group_desc *gdp;
ext4_group_t i;
#ifdef EXT4FS_DEBUG
struct ext4_super_block *es;
unsigned long bitmap_count, x;
struct buffer_head *bitmap_bh = NULL;
es = EXT4_SB(sb)->s_es;
desc_count = 0;
bitmap_count = 0;
gdp = NULL;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
desc_count += ext4_free_inodes_count(sb, gdp);
brelse(bitmap_bh);
bitmap_bh = ext4_read_inode_bitmap(sb, i);
if (!bitmap_bh)
continue;
x = ext4_count_free(bitmap_bh, EXT4_INODES_PER_GROUP(sb) / 8);
printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n",
i, ext4_free_inodes_count(sb, gdp), x);
bitmap_count += x;
}
brelse(bitmap_bh);
printk(KERN_DEBUG "ext4_count_free_inodes: "
"stored = %u, computed = %lu, %lu\n",
le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
return desc_count;
#else
desc_count = 0;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
desc_count += ext4_free_inodes_count(sb, gdp);
cond_resched();
}
return desc_count;
#endif
}
/* Called at mount-time, super-block is locked */
unsigned long ext4_count_dirs(struct super_block * sb)
{
unsigned long count = 0;
ext4_group_t i;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
count += ext4_used_dirs_count(sb, gdp);
}
return count;
}