^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * linux/fs/ext4/ialloc.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright (C) 1992, 1993, 1994, 1995
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * Remy Card (card@masi.ibp.fr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * Laboratoire MASI - Institut Blaise Pascal
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) * Universite Pierre et Marie Curie (Paris VI)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * BSD ufs-inspired inode and directory allocation by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) * Stephen Tweedie (sct@redhat.com), 1993
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) * Big-endian to little-endian byte-swapping/bitmaps by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) * David S. Miller (davem@caip.rutgers.edu), 1995
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/time.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/stat.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/string.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/quotaops.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/buffer_head.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/random.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/bitops.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/blkdev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/cred.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <asm/byteorder.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include "ext4.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include "ext4_jbd2.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include "xattr.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include "acl.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #include <trace/events/ext4.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * ialloc.c contains the inodes allocation and deallocation routines
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) * The free inodes are managed by bitmaps. A file system contains several
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * block for inodes, N blocks for the inode table and data blocks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * The file system contains group descriptors which are located after the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * super block. Each descriptor contains the number of the bitmap block and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) * the free blocks count in the block.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) * To avoid calling the atomic setbit hundreds or thousands of times, we only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) * need to use it within a single byte (to ensure we get endianness right).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) * We can use memset for the rest of the bitmap as there are no other users.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) void ext4_mark_bitmap_end(int start_bit, int end_bit, char *bitmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) if (start_bit >= end_bit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) ext4_debug("mark end bits +%d through +%d used\n", start_bit, end_bit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) ext4_set_bit(i, bitmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) if (i < end_bit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) void ext4_end_bitmap_read(struct buffer_head *bh, int uptodate)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) if (uptodate) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) set_buffer_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) set_bitmap_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) unlock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) put_bh(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) static int ext4_validate_inode_bitmap(struct super_block *sb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) struct ext4_group_desc *desc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) ext4_group_t block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) struct buffer_head *bh)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) ext4_fsblk_t blk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) struct ext4_group_info *grp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) if (EXT4_SB(sb)->s_mount_state & EXT4_FC_REPLAY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) grp = ext4_get_group_info(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) if (buffer_verified(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) if (EXT4_MB_GRP_IBITMAP_CORRUPT(grp))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) return -EFSCORRUPTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) ext4_lock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) if (buffer_verified(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) goto verified;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) blk = ext4_inode_bitmap(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) if (!ext4_inode_bitmap_csum_verify(sb, block_group, desc, bh,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) EXT4_INODES_PER_GROUP(sb) / 8) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) ext4_simulate_fail(sb, EXT4_SIM_IBITMAP_CRC)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) ext4_error(sb, "Corrupt inode bitmap - block_group = %u, "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) "inode_bitmap = %llu", block_group, blk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) ext4_mark_group_bitmap_corrupted(sb, block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) return -EFSBADCRC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) set_buffer_verified(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) verified:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) * Read the inode allocation bitmap for a given block_group, reading
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) * into the specified slot in the superblock's bitmap cache.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) * Return buffer_head of bitmap on success, or an ERR_PTR on error.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) static struct buffer_head *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) struct ext4_group_desc *desc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) struct ext4_sb_info *sbi = EXT4_SB(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) struct buffer_head *bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) ext4_fsblk_t bitmap_blk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) desc = ext4_get_group_desc(sb, block_group, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) if (!desc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) return ERR_PTR(-EFSCORRUPTED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) bitmap_blk = ext4_inode_bitmap(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) if ((bitmap_blk <= le32_to_cpu(sbi->s_es->s_first_data_block)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) (bitmap_blk >= ext4_blocks_count(sbi->s_es))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) ext4_error(sb, "Invalid inode bitmap blk %llu in "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) "block_group %u", bitmap_blk, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) ext4_mark_group_bitmap_corrupted(sb, block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) return ERR_PTR(-EFSCORRUPTED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) bh = sb_getblk(sb, bitmap_blk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) if (unlikely(!bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) ext4_warning(sb, "Cannot read inode bitmap - "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) "block_group = %u, inode_bitmap = %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) block_group, bitmap_blk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) if (bitmap_uptodate(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) goto verify;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) lock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) if (bitmap_uptodate(bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) unlock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) goto verify;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) ext4_lock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) if (ext4_has_group_desc_csum(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) if (block_group == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) unlock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) ext4_error(sb, "Inode bitmap for bg 0 marked "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) "uninitialized");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) err = -EFSCORRUPTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) memset(bh->b_data, 0, (EXT4_INODES_PER_GROUP(sb) + 7) / 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) sb->s_blocksize * 8, bh->b_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) set_bitmap_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) set_buffer_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) set_buffer_verified(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) unlock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) return bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) if (buffer_uptodate(bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) * if not uninit if bh is uptodate,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) * bitmap is also uptodate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) set_bitmap_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) unlock_buffer(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) goto verify;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) * submit the buffer_head for reading
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) trace_ext4_load_inode_bitmap(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) ext4_read_bh(bh, REQ_META | REQ_PRIO, ext4_end_bitmap_read);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) ext4_simulate_fail_bh(sb, bh, EXT4_SIM_IBITMAP_EIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) if (!buffer_uptodate(bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) put_bh(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) ext4_error_err(sb, EIO, "Cannot read inode bitmap - "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) "block_group = %u, inode_bitmap = %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) block_group, bitmap_blk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) ext4_mark_group_bitmap_corrupted(sb, block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) return ERR_PTR(-EIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) verify:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) err = ext4_validate_inode_bitmap(sb, desc, block_group, bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) return bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) put_bh(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) return ERR_PTR(err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) * NOTE! When we get the inode, we're the only people
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) * that have access to it, and as such there are no
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) * race conditions we have to worry about. The inode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) * is not on the hash-lists, and it cannot be reached
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) * through the filesystem because the directory entry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) * has been deleted earlier.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) * HOWEVER: we must make sure that we get no aliases,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) * which means that we have to call "clear_inode()"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) * _before_ we mark the inode not in use in the inode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) * bitmaps. Otherwise a newly created file might use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) * the same inode number (not actually the same pointer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) * though), and then we'd have two inodes sharing the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * same inode number and space on the harddisk.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) void ext4_free_inode(handle_t *handle, struct inode *inode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) struct super_block *sb = inode->i_sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) int is_directory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) unsigned long ino;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) struct buffer_head *bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) struct buffer_head *bh2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) ext4_group_t block_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) unsigned long bit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) struct ext4_group_desc *gdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) struct ext4_super_block *es;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) struct ext4_sb_info *sbi;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) int fatal = 0, err, count, cleared;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) struct ext4_group_info *grp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) if (!sb) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) printk(KERN_ERR "EXT4-fs: %s:%d: inode on "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) "nonexistent device\n", __func__, __LINE__);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) if (atomic_read(&inode->i_count) > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) ext4_msg(sb, KERN_ERR, "%s:%d: inode #%lu: count=%d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) __func__, __LINE__, inode->i_ino,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) atomic_read(&inode->i_count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) if (inode->i_nlink) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) ext4_msg(sb, KERN_ERR, "%s:%d: inode #%lu: nlink=%d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) __func__, __LINE__, inode->i_ino, inode->i_nlink);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) sbi = EXT4_SB(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) ino = inode->i_ino;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) ext4_debug("freeing inode %lu\n", ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) trace_ext4_free_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) dquot_initialize(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) dquot_free_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) is_directory = S_ISDIR(inode->i_mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) /* Do this BEFORE marking the inode not in use or returning an error */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) ext4_clear_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) es = sbi->s_es;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) ext4_error(sb, "reserved or nonexistent inode %lu", ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) goto error_return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) /* Don't bother if the inode bitmap is corrupt. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) if (IS_ERR(bitmap_bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) fatal = PTR_ERR(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) goto error_return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) if (!(sbi->s_mount_state & EXT4_FC_REPLAY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) grp = ext4_get_group_info(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) if (unlikely(EXT4_MB_GRP_IBITMAP_CORRUPT(grp))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) fatal = -EFSCORRUPTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) goto error_return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) BUFFER_TRACE(bitmap_bh, "get_write_access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) fatal = ext4_journal_get_write_access(handle, bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) if (fatal)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) goto error_return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) fatal = -ESRCH;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) gdp = ext4_get_group_desc(sb, block_group, &bh2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) if (gdp) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) BUFFER_TRACE(bh2, "get_write_access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) fatal = ext4_journal_get_write_access(handle, bh2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) ext4_lock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) cleared = ext4_test_and_clear_bit(bit, bitmap_bh->b_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) if (fatal || !cleared) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) count = ext4_free_inodes_count(sb, gdp) + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) ext4_free_inodes_set(sb, gdp, count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) if (is_directory) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) count = ext4_used_dirs_count(sb, gdp) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) ext4_used_dirs_set(sb, gdp, count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) if (percpu_counter_initialized(&sbi->s_dirs_counter))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) percpu_counter_dec(&sbi->s_dirs_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) ext4_inode_bitmap_csum_set(sb, block_group, gdp, bitmap_bh,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) EXT4_INODES_PER_GROUP(sb) / 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) ext4_group_desc_csum_set(sb, block_group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) ext4_unlock_group(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) if (percpu_counter_initialized(&sbi->s_freeinodes_counter))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) percpu_counter_inc(&sbi->s_freeinodes_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) if (sbi->s_log_groups_per_flex) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) struct flex_groups *fg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) fg = sbi_array_rcu_deref(sbi, s_flex_groups,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) ext4_flex_group(sbi, block_group));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) atomic_inc(&fg->free_inodes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) if (is_directory)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) atomic_dec(&fg->used_dirs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) fatal = ext4_handle_dirty_metadata(handle, NULL, bh2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) if (cleared) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) BUFFER_TRACE(bitmap_bh, "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) if (!fatal)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) fatal = err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) ext4_error(sb, "bit already cleared for inode %lu", ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) ext4_mark_group_bitmap_corrupted(sb, block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) error_return:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) ext4_std_error(sb, fatal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) struct orlov_stats {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) __u64 free_clusters;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) __u32 free_inodes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) __u32 used_dirs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) * Helper function for Orlov's allocator; returns critical information
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) * for a particular block group or flex_bg. If flex_size is 1, then g
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) * is a block group number; otherwise it is flex_bg number.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) static void get_orlov_stats(struct super_block *sb, ext4_group_t g,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) int flex_size, struct orlov_stats *stats)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) struct ext4_group_desc *desc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) if (flex_size > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) struct flex_groups *fg = sbi_array_rcu_deref(EXT4_SB(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) s_flex_groups, g);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) stats->free_inodes = atomic_read(&fg->free_inodes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) stats->free_clusters = atomic64_read(&fg->free_clusters);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) stats->used_dirs = atomic_read(&fg->used_dirs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) desc = ext4_get_group_desc(sb, g, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) if (desc) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) stats->free_inodes = ext4_free_inodes_count(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) stats->free_clusters = ext4_free_group_clusters(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) stats->used_dirs = ext4_used_dirs_count(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) stats->free_inodes = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) stats->free_clusters = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) stats->used_dirs = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) * Orlov's allocator for directories.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) * We always try to spread first-level directories.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) * If there are blockgroups with both free inodes and free clusters counts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) * not worse than average we return one with smallest directory count.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) * Otherwise we simply return a random group.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) * For the rest rules look so:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) * It's OK to put directory into a group unless
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) * it has too many directories already (max_dirs) or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * it has too few free inodes left (min_inodes) or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) * it has too few free clusters left (min_clusters) or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) * Parent's group is preferred, if it doesn't satisfy these
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) * conditions we search cyclically through the rest. If none
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) * of the groups look good we just look for a group with more
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) * free inodes than average (starting at parent's group).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) static int find_group_orlov(struct super_block *sb, struct inode *parent,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) ext4_group_t *group, umode_t mode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) const struct qstr *qstr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) struct ext4_sb_info *sbi = EXT4_SB(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) ext4_group_t real_ngroups = ext4_get_groups_count(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) int inodes_per_group = EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) unsigned int freei, avefreei, grp_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) ext4_fsblk_t freec, avefreec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) unsigned int ndirs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) int max_dirs, min_inodes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) ext4_grpblk_t min_clusters;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) ext4_group_t i, grp, g, ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) struct ext4_group_desc *desc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) struct orlov_stats stats;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) int flex_size = ext4_flex_bg_size(sbi);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) struct dx_hash_info hinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) ngroups = real_ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) if (flex_size > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) ngroups = (real_ngroups + flex_size - 1) >>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) sbi->s_log_groups_per_flex;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) parent_group >>= sbi->s_log_groups_per_flex;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) avefreei = freei / ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) freec = percpu_counter_read_positive(&sbi->s_freeclusters_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) avefreec = freec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) do_div(avefreec, ngroups);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) if (S_ISDIR(mode) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) ((parent == d_inode(sb->s_root)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) (ext4_test_inode_flag(parent, EXT4_INODE_TOPDIR)))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) int best_ndir = inodes_per_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) int ret = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) if (qstr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) if (ext4_hash_in_dirent(parent))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) hinfo.hash_version = DX_HASH_SIPHASH;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) hinfo.hash_version = DX_HASH_HALF_MD4;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) hinfo.seed = sbi->s_hash_seed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) ext4fs_dirhash(parent, qstr->name, qstr->len, &hinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) grp = hinfo.hash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) grp = prandom_u32();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) parent_group = (unsigned)grp % ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) g = (parent_group + i) % ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) get_orlov_stats(sb, g, flex_size, &stats);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) if (!stats.free_inodes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) if (stats.used_dirs >= best_ndir)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) if (stats.free_inodes < avefreei)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) if (stats.free_clusters < avefreec)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) grp = g;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) best_ndir = stats.used_dirs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) goto fallback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) found_flex_bg:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) if (flex_size == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) *group = grp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) * We pack inodes at the beginning of the flexgroup's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) * inode tables. Block allocation decisions will do
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) * something similar, although regular files will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) * start at 2nd block group of the flexgroup. See
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) * ext4_ext_find_goal() and ext4_find_near().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) grp *= flex_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) for (i = 0; i < flex_size; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) if (grp+i >= real_ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) desc = ext4_get_group_desc(sb, grp+i, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) if (desc && ext4_free_inodes_count(sb, desc)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) *group = grp+i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) goto fallback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) max_dirs = ndirs / ngroups + inodes_per_group / 16;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) min_inodes = avefreei - inodes_per_group*flex_size / 4;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) if (min_inodes < 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) min_inodes = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) min_clusters = avefreec - EXT4_CLUSTERS_PER_GROUP(sb)*flex_size / 4;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) * Start looking in the flex group where we last allocated an
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) * inode for this parent directory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) if (EXT4_I(parent)->i_last_alloc_group != ~0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) parent_group = EXT4_I(parent)->i_last_alloc_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) if (flex_size > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) parent_group >>= sbi->s_log_groups_per_flex;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) grp = (parent_group + i) % ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) get_orlov_stats(sb, grp, flex_size, &stats);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) if (stats.used_dirs >= max_dirs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) if (stats.free_inodes < min_inodes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) if (stats.free_clusters < min_clusters)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) goto found_flex_bg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) fallback:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) ngroups = real_ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) avefreei = freei / ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) fallback_retry:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) parent_group = EXT4_I(parent)->i_block_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) grp = (parent_group + i) % ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) desc = ext4_get_group_desc(sb, grp, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) if (desc) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) grp_free = ext4_free_inodes_count(sb, desc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) if (grp_free && grp_free >= avefreei) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) *group = grp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) if (avefreei) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) * The free-inodes counter is approximate, and for really small
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) * filesystems the above test can fail to find any blockgroups
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) avefreei = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) goto fallback_retry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) return -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) static int find_group_other(struct super_block *sb, struct inode *parent,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) ext4_group_t *group, umode_t mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) ext4_group_t i, last, ngroups = ext4_get_groups_count(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) struct ext4_group_desc *desc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) int flex_size = ext4_flex_bg_size(EXT4_SB(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) * Try to place the inode is the same flex group as its
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) * parent. If we can't find space, use the Orlov algorithm to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) * find another flex group, and store that information in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) * parent directory's inode information so that use that flex
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) * group for future allocations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) if (flex_size > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) int retry = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) try_again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) parent_group &= ~(flex_size-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) last = parent_group + flex_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) if (last > ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) last = ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) for (i = parent_group; i < last; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) desc = ext4_get_group_desc(sb, i, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) if (desc && ext4_free_inodes_count(sb, desc)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) *group = i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) if (!retry && EXT4_I(parent)->i_last_alloc_group != ~0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) retry = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) parent_group = EXT4_I(parent)->i_last_alloc_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) goto try_again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) * If this didn't work, use the Orlov search algorithm
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) * to find a new flex group; we pass in the mode to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) * avoid the topdir algorithms.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) *group = parent_group + flex_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) if (*group > ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) *group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) return find_group_orlov(sb, parent, group, mode, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) * Try to place the inode in its parent directory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) *group = parent_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) desc = ext4_get_group_desc(sb, *group, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) if (desc && ext4_free_inodes_count(sb, desc) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) ext4_free_group_clusters(sb, desc))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) * We're going to place this inode in a different blockgroup from its
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * parent. We want to cause files in a common directory to all land in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) * the same blockgroup. But we want files which are in a different
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) * directory which shares a blockgroup with our parent to land in a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) * different blockgroup.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) * So add our directory's i_ino into the starting point for the hash.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) *group = (*group + parent->i_ino) % ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) * Use a quadratic hash to find a group with a free inode and some free
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) * blocks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) for (i = 1; i < ngroups; i <<= 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) *group += i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) if (*group >= ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) *group -= ngroups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) desc = ext4_get_group_desc(sb, *group, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) if (desc && ext4_free_inodes_count(sb, desc) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) ext4_free_group_clusters(sb, desc))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) * That failed: try linear search for a free inode, even if that group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) * has no free blocks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) *group = parent_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) if (++*group >= ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) *group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) desc = ext4_get_group_desc(sb, *group, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) if (desc && ext4_free_inodes_count(sb, desc))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) return -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) * In no journal mode, if an inode has recently been deleted, we want
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) * to avoid reusing it until we're reasonably sure the inode table
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) * block has been written back to disk. (Yes, these values are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) * somewhat arbitrary...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) #define RECENTCY_MIN 60
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) #define RECENTCY_DIRTY 300
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) static int recently_deleted(struct super_block *sb, ext4_group_t group, int ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) struct ext4_group_desc *gdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) struct ext4_inode *raw_inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) struct buffer_head *bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) int inodes_per_block = EXT4_SB(sb)->s_inodes_per_block;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) int offset, ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) int recentcy = RECENTCY_MIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) u32 dtime, now;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) gdp = ext4_get_group_desc(sb, group, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) if (unlikely(!gdp))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) bh = sb_find_get_block(sb, ext4_inode_table(sb, gdp) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) (ino / inodes_per_block));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) if (!bh || !buffer_uptodate(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) * If the block is not in the buffer cache, then it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) * must have been written out.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) offset = (ino % inodes_per_block) * EXT4_INODE_SIZE(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) raw_inode = (struct ext4_inode *) (bh->b_data + offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) /* i_dtime is only 32 bits on disk, but we only care about relative
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) * times in the range of a few minutes (i.e. long enough to sync a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) * recently-deleted inode to disk), so using the low 32 bits of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) * clock (a 68 year range) is enough, see time_before32() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) dtime = le32_to_cpu(raw_inode->i_dtime);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) now = ktime_get_real_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) if (buffer_dirty(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) recentcy += RECENTCY_DIRTY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) if (dtime && time_before32(dtime, now) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) time_before32(now, dtime + recentcy))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) brelse(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) static int find_inode_bit(struct super_block *sb, ext4_group_t group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) struct buffer_head *bitmap, unsigned long *ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) bool check_recently_deleted = EXT4_SB(sb)->s_journal == NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) unsigned long recently_deleted_ino = EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) next:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) *ino = ext4_find_next_zero_bit((unsigned long *)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) bitmap->b_data,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) EXT4_INODES_PER_GROUP(sb), *ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) if (*ino >= EXT4_INODES_PER_GROUP(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) goto not_found;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) if (check_recently_deleted && recently_deleted(sb, group, *ino)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) recently_deleted_ino = *ino;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) *ino = *ino + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) if (*ino < EXT4_INODES_PER_GROUP(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) goto next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) goto not_found;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) not_found:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) if (recently_deleted_ino >= EXT4_INODES_PER_GROUP(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) * Not reusing recently deleted inodes is mostly a preference. We don't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) * want to report ENOSPC or skew allocation patterns because of that.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) * So return even recently deleted inode if we could find better in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) * given range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) *ino = recently_deleted_ino;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) int ext4_mark_inode_used(struct super_block *sb, int ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) struct buffer_head *inode_bitmap_bh = NULL, *group_desc_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) struct ext4_group_desc *gdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) ext4_group_t group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) int bit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) int err = -EFSCORRUPTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) if (ino < EXT4_FIRST_INO(sb) || ino > max_ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) inode_bitmap_bh = ext4_read_inode_bitmap(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) if (IS_ERR(inode_bitmap_bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) return PTR_ERR(inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) if (ext4_test_bit(bit, inode_bitmap_bh->b_data)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) gdp = ext4_get_group_desc(sb, group, &group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) if (!gdp || !group_desc_bh) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) ext4_set_bit(bit, inode_bitmap_bh->b_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) BUFFER_TRACE(inode_bitmap_bh, "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) err = ext4_handle_dirty_metadata(NULL, NULL, inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) err = sync_dirty_buffer(inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) /* We may have to initialize the block bitmap if it isn't already */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) if (ext4_has_group_desc_csum(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) struct buffer_head *block_bitmap_bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) block_bitmap_bh = ext4_read_block_bitmap(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) if (IS_ERR(block_bitmap_bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) err = PTR_ERR(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) BUFFER_TRACE(block_bitmap_bh, "dirty block bitmap");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) err = ext4_handle_dirty_metadata(NULL, NULL, block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) sync_dirty_buffer(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) /* recheck and clear flag under lock if we still need to */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) if (ext4_has_group_desc_csum(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) ext4_free_group_clusters_set(sb, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) ext4_free_clusters_after_init(sb, group, gdp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) ext4_block_bitmap_csum_set(sb, group, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) ext4_group_desc_csum_set(sb, group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) brelse(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) /* Update the relevant bg descriptor fields */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) if (ext4_has_group_desc_csum(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) int free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) ext4_lock_group(sb, group); /* while we modify the bg desc */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) free = EXT4_INODES_PER_GROUP(sb) -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) ext4_itable_unused_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) if (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) free = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) * Check the relative inode number against the last used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) * relative inode number in this group. if it is greater
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) * we need to update the bg_itable_unused count
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) if (bit >= free)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) ext4_itable_unused_set(sb, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) (EXT4_INODES_PER_GROUP(sb) - bit - 1));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) ext4_free_inodes_set(sb, gdp, ext4_free_inodes_count(sb, gdp) - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) if (ext4_has_group_desc_csum(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) ext4_inode_bitmap_csum_set(sb, group, gdp, inode_bitmap_bh,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) EXT4_INODES_PER_GROUP(sb) / 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) ext4_group_desc_csum_set(sb, group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) err = ext4_handle_dirty_metadata(NULL, NULL, group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) sync_dirty_buffer(group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) static int ext4_xattr_credits_for_new_inode(struct inode *dir, mode_t mode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) bool encrypt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) struct super_block *sb = dir->i_sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) int nblocks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) #ifdef CONFIG_EXT4_FS_POSIX_ACL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) struct posix_acl *p = get_acl(dir, ACL_TYPE_DEFAULT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) if (IS_ERR(p))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) return PTR_ERR(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) if (p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) int acl_size = p->a_count * sizeof(ext4_acl_entry);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) nblocks += (S_ISDIR(mode) ? 2 : 1) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) __ext4_xattr_set_credits(sb, NULL /* inode */,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) NULL /* block_bh */, acl_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) true /* is_create */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) posix_acl_release(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) #ifdef CONFIG_SECURITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) int num_security_xattrs = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) #ifdef CONFIG_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) num_security_xattrs++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) * We assume that security xattrs are never more than 1k.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) * In practice they are under 128 bytes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) nblocks += num_security_xattrs *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) __ext4_xattr_set_credits(sb, NULL /* inode */,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) NULL /* block_bh */, 1024,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) true /* is_create */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) if (encrypt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) nblocks += __ext4_xattr_set_credits(sb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) NULL /* inode */,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) NULL /* block_bh */,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) FSCRYPT_SET_CONTEXT_MAX_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) true /* is_create */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) return nblocks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) * There are two policies for allocating an inode. If the new inode is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) * a directory, then a forward search is made for a block group with both
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) * free space and a low directory-to-inode ratio; if that fails, then of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) * the groups with above-average free space, that group with the fewest
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) * directories already is chosen.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) * For other inodes, search forward from the parent directory's block
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) * group to find a free inode.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) umode_t mode, const struct qstr *qstr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) __u32 goal, uid_t *owner, __u32 i_flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) int handle_type, unsigned int line_no,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) int nblocks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) struct super_block *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) struct buffer_head *inode_bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) struct buffer_head *group_desc_bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) ext4_group_t ngroups, group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) unsigned long ino = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) struct inode *inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) struct ext4_group_desc *gdp = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) struct ext4_inode_info *ei;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) struct ext4_sb_info *sbi;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) int ret2, err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) struct inode *ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) ext4_group_t i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) ext4_group_t flex_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) struct ext4_group_info *grp = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) bool encrypt = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) /* Cannot create files in a deleted directory */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) if (!dir || !dir->i_nlink)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) return ERR_PTR(-EPERM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) sb = dir->i_sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) sbi = EXT4_SB(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) if (unlikely(ext4_forced_shutdown(sbi)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) return ERR_PTR(-EIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) ngroups = ext4_get_groups_count(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) trace_ext4_request_inode(dir, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) inode = new_inode(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) if (!inode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) ei = EXT4_I(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) * Initialize owners and quota early so that we don't have to account
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) * for quota initialization worst case in standard inode creating
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) * transaction
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) if (owner) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) inode->i_mode = mode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) i_uid_write(inode, owner[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) i_gid_write(inode, owner[1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) } else if (test_opt(sb, GRPID)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) inode->i_mode = mode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) inode->i_uid = current_fsuid();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) inode->i_gid = dir->i_gid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) inode_init_owner(inode, dir, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) if (ext4_has_feature_project(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) ext4_test_inode_flag(dir, EXT4_INODE_PROJINHERIT))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) ei->i_projid = EXT4_I(dir)->i_projid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) ei->i_projid = make_kprojid(&init_user_ns, EXT4_DEF_PROJID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) if (!(i_flags & EXT4_EA_INODE_FL)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) err = dquot_initialize(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) if (!handle && sbi->s_journal && !(i_flags & EXT4_EA_INODE_FL)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) ret2 = ext4_xattr_credits_for_new_inode(dir, mode, encrypt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) if (ret2 < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) err = ret2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) nblocks += ret2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) if (!goal)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) goal = sbi->s_inode_goal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) if (goal && goal <= le32_to_cpu(sbi->s_es->s_inodes_count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) group = (goal - 1) / EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) ino = (goal - 1) % EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) ret2 = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) goto got_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) if (S_ISDIR(mode))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) ret2 = find_group_orlov(sb, dir, &group, mode, qstr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) ret2 = find_group_other(sb, dir, &group, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) got_group:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) EXT4_I(dir)->i_last_alloc_group = group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) err = -ENOSPC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) if (ret2 == -1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) * Normally we will only go through one pass of this loop,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) * unless we get unlucky and it turns out the group we selected
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) * had its last inode grabbed by someone else.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) for (i = 0; i < ngroups; i++, ino = 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) err = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) gdp = ext4_get_group_desc(sb, group, &group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) if (!gdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) * Check free inodes count before loading bitmap.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) if (ext4_free_inodes_count(sb, gdp) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) goto next_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) if (!(sbi->s_mount_state & EXT4_FC_REPLAY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) grp = ext4_get_group_info(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) * Skip groups with already-known suspicious inode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) * tables
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) if (EXT4_MB_GRP_IBITMAP_CORRUPT(grp))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) goto next_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) brelse(inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) inode_bitmap_bh = ext4_read_inode_bitmap(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) /* Skip groups with suspicious inode tables */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) if (((!(sbi->s_mount_state & EXT4_FC_REPLAY))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) && EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) IS_ERR(inode_bitmap_bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) inode_bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) goto next_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) repeat_in_this_group:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) ret2 = find_inode_bit(sb, group, inode_bitmap_bh, &ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) if (!ret2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) goto next_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) if (group == 0 && (ino + 1) < EXT4_FIRST_INO(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) ext4_error(sb, "reserved inode found cleared - "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) "inode=%lu", ino + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) ext4_mark_group_bitmap_corrupted(sb, group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) goto next_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) if ((!(sbi->s_mount_state & EXT4_FC_REPLAY)) && !handle) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) BUG_ON(nblocks <= 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) handle = __ext4_journal_start_sb(dir->i_sb, line_no,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) handle_type, nblocks, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) ext4_trans_default_revoke_credits(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) if (IS_ERR(handle)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) err = PTR_ERR(handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) BUFFER_TRACE(inode_bitmap_bh, "get_write_access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) err = ext4_journal_get_write_access(handle, inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) ret2 = ext4_test_and_set_bit(ino, inode_bitmap_bh->b_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) if (ret2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) /* Someone already took the bit. Repeat the search
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) * with lock held.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) ret2 = find_inode_bit(sb, group, inode_bitmap_bh, &ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) if (ret2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) ext4_set_bit(ino, inode_bitmap_bh->b_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) ret2 = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) ret2 = 1; /* we didn't grab the inode */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) ino++; /* the inode bitmap is zero-based */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) if (!ret2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) goto got; /* we grabbed the inode! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) if (ino < EXT4_INODES_PER_GROUP(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) goto repeat_in_this_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) next_group:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) if (++group == ngroups)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) err = -ENOSPC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) got:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) BUFFER_TRACE(inode_bitmap_bh, "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) err = ext4_handle_dirty_metadata(handle, NULL, inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) BUFFER_TRACE(group_desc_bh, "get_write_access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) err = ext4_journal_get_write_access(handle, group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) /* We may have to initialize the block bitmap if it isn't already */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) if (ext4_has_group_desc_csum(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) struct buffer_head *block_bitmap_bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) block_bitmap_bh = ext4_read_block_bitmap(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) if (IS_ERR(block_bitmap_bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) err = PTR_ERR(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) BUFFER_TRACE(block_bitmap_bh, "get block bitmap access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) err = ext4_journal_get_write_access(handle, block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) brelse(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) BUFFER_TRACE(block_bitmap_bh, "dirty block bitmap");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) err = ext4_handle_dirty_metadata(handle, NULL, block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) /* recheck and clear flag under lock if we still need to */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) if (ext4_has_group_desc_csum(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) ext4_free_group_clusters_set(sb, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) ext4_free_clusters_after_init(sb, group, gdp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) ext4_block_bitmap_csum_set(sb, group, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) ext4_group_desc_csum_set(sb, group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) brelse(block_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) /* Update the relevant bg descriptor fields */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) if (ext4_has_group_desc_csum(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) int free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) struct ext4_group_info *grp = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) if (!(sbi->s_mount_state & EXT4_FC_REPLAY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) grp = ext4_get_group_info(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) down_read(&grp->alloc_sem); /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) * protect vs itable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) * lazyinit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) ext4_lock_group(sb, group); /* while we modify the bg desc */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) free = EXT4_INODES_PER_GROUP(sb) -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) ext4_itable_unused_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) if (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) free = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) * Check the relative inode number against the last used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) * relative inode number in this group. if it is greater
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) * we need to update the bg_itable_unused count
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) if (ino > free)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) ext4_itable_unused_set(sb, gdp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) (EXT4_INODES_PER_GROUP(sb) - ino));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) if (!(sbi->s_mount_state & EXT4_FC_REPLAY))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) up_read(&grp->alloc_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) ext4_free_inodes_set(sb, gdp, ext4_free_inodes_count(sb, gdp) - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) if (S_ISDIR(mode)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) ext4_used_dirs_set(sb, gdp, ext4_used_dirs_count(sb, gdp) + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) if (sbi->s_log_groups_per_flex) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) ext4_group_t f = ext4_flex_group(sbi, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) atomic_inc(&sbi_array_rcu_deref(sbi, s_flex_groups,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) f)->used_dirs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) if (ext4_has_group_desc_csum(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) ext4_inode_bitmap_csum_set(sb, group, gdp, inode_bitmap_bh,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) EXT4_INODES_PER_GROUP(sb) / 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) ext4_group_desc_csum_set(sb, group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) BUFFER_TRACE(group_desc_bh, "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) err = ext4_handle_dirty_metadata(handle, NULL, group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) percpu_counter_dec(&sbi->s_freeinodes_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) if (S_ISDIR(mode))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) percpu_counter_inc(&sbi->s_dirs_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) if (sbi->s_log_groups_per_flex) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) flex_group = ext4_flex_group(sbi, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) atomic_dec(&sbi_array_rcu_deref(sbi, s_flex_groups,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) flex_group)->free_inodes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) /* This is the optimal IO size (for stat), not the fs block size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) inode->i_blocks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) ei->i_crtime = inode->i_mtime;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) memset(ei->i_data, 0, sizeof(ei->i_data));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) ei->i_dir_start_lookup = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) ei->i_disksize = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) /* Don't inherit extent flag from directory, amongst others. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) ei->i_flags =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) ext4_mask_flags(mode, EXT4_I(dir)->i_flags & EXT4_FL_INHERITED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) ei->i_flags |= i_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) ei->i_file_acl = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) ei->i_dtime = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) ei->i_block_group = group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) ei->i_last_alloc_group = ~0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) ext4_set_inode_flags(inode, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) if (IS_DIRSYNC(inode))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) ext4_handle_sync(handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) if (insert_inode_locked(inode) < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) * Likely a bitmap corruption causing inode to be allocated
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) * twice.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) err = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) ext4_error(sb, "failed to insert inode %lu: doubly allocated?",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) inode->i_ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) ext4_mark_group_bitmap_corrupted(sb, group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) EXT4_GROUP_INFO_IBITMAP_CORRUPT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) inode->i_generation = prandom_u32();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) /* Precompute checksum seed for inode metadata */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) if (ext4_has_metadata_csum(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) __u32 csum;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) __le32 inum = cpu_to_le32(inode->i_ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) __le32 gen = cpu_to_le32(inode->i_generation);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) csum = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&inum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) sizeof(inum));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) ei->i_csum_seed = ext4_chksum(sbi, csum, (__u8 *)&gen,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) sizeof(gen));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) ext4_clear_state_flags(ei); /* Only relevant on 32-bit archs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) ext4_set_inode_state(inode, EXT4_STATE_NEW);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) ei->i_extra_isize = sbi->s_want_extra_isize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) ei->i_inline_off = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) if (ext4_has_feature_inline_data(sb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) (!(ei->i_flags & EXT4_DAX_FL) || S_ISDIR(mode)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) ext4_set_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) ret = inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) err = dquot_alloc_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) goto fail_drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) * Since the encryption xattr will always be unique, create it first so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) * that it's less likely to end up in an external xattr block and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) * prevent its deduplication.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) if (encrypt) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) err = fscrypt_set_context(inode, handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) goto fail_free_drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) if (!(ei->i_flags & EXT4_EA_INODE_FL)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) err = ext4_init_acl(handle, inode, dir);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) goto fail_free_drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) err = ext4_init_security(handle, inode, dir, qstr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) goto fail_free_drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) if (ext4_has_feature_extents(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) /* set extent flag only for directory, file and normal symlink*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) ext4_set_inode_flag(inode, EXT4_INODE_EXTENTS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) ext4_ext_tree_init(handle, inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) if (ext4_handle_valid(handle)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) ei->i_sync_tid = handle->h_transaction->t_tid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) ei->i_datasync_tid = handle->h_transaction->t_tid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) err = ext4_mark_inode_dirty(handle, inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) ext4_std_error(sb, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) goto fail_free_drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) ext4_debug("allocating inode %lu\n", inode->i_ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) trace_ext4_allocate_inode(inode, dir, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) brelse(inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) fail_free_drop:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) dquot_free_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) fail_drop:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) clear_nlink(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) unlock_new_inode(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) dquot_drop(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358) inode->i_flags |= S_NOQUOTA;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) iput(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) brelse(inode_bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) return ERR_PTR(err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) /* Verify that we are loading a valid orphan from disk */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) ext4_group_t block_group;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) int bit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) struct buffer_head *bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) struct inode *inode = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) int err = -EFSCORRUPTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) if (ino < EXT4_FIRST_INO(sb) || ino > max_ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) goto bad_orphan;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) if (IS_ERR(bitmap_bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) return ERR_CAST(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) /* Having the inode bit set should be a 100% indicator that this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) * is a valid orphan (no e2fsck run on fs). Orphans also include
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) * inodes that were being truncated, so we can't check i_nlink==0.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) if (!ext4_test_bit(bit, bitmap_bh->b_data))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) goto bad_orphan;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) inode = ext4_iget(sb, ino, EXT4_IGET_NORMAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) if (IS_ERR(inode)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) err = PTR_ERR(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) ext4_error_err(sb, -err,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) "couldn't read orphan inode %lu (err %d)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) ino, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) return inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401) * If the orphans has i_nlinks > 0 then it should be able to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) * be truncated, otherwise it won't be removed from the orphan
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) * list during processing and an infinite loop will result.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) * Similarly, it must not be a bad inode.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) if ((inode->i_nlink && !ext4_can_truncate(inode)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) is_bad_inode(inode))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) goto bad_orphan;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) if (NEXT_ORPHAN(inode) > max_ino)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) goto bad_orphan;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) return inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) bad_orphan:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) ext4_error(sb, "bad orphan inode %lu", ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) if (bitmap_bh)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) printk(KERN_ERR "ext4_test_bit(bit=%d, block=%llu) = %d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) bit, (unsigned long long)bitmap_bh->b_blocknr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) ext4_test_bit(bit, bitmap_bh->b_data));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) if (inode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) printk(KERN_ERR "is_bad_inode(inode)=%d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) is_bad_inode(inode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) printk(KERN_ERR "NEXT_ORPHAN(inode)=%u\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) NEXT_ORPHAN(inode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) printk(KERN_ERR "max_ino=%lu\n", max_ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) printk(KERN_ERR "i_nlink=%u\n", inode->i_nlink);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) /* Avoid freeing blocks if we got a bad deleted inode */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) if (inode->i_nlink == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) inode->i_blocks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) iput(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) return ERR_PTR(err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) unsigned long ext4_count_free_inodes(struct super_block *sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) unsigned long desc_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) struct ext4_group_desc *gdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) ext4_group_t i, ngroups = ext4_get_groups_count(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) #ifdef EXT4FS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) struct ext4_super_block *es;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) unsigned long bitmap_count, x;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) struct buffer_head *bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) es = EXT4_SB(sb)->s_es;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) desc_count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) bitmap_count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) gdp = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) gdp = ext4_get_group_desc(sb, i, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) if (!gdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) desc_count += ext4_free_inodes_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) bitmap_bh = ext4_read_inode_bitmap(sb, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) if (IS_ERR(bitmap_bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) bitmap_bh = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) x = ext4_count_free(bitmap_bh->b_data,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) EXT4_INODES_PER_GROUP(sb) / 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) (unsigned long) i, ext4_free_inodes_count(sb, gdp), x);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) bitmap_count += x;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) brelse(bitmap_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) printk(KERN_DEBUG "ext4_count_free_inodes: "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) "stored = %u, computed = %lu, %lu\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) return desc_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) desc_count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) gdp = ext4_get_group_desc(sb, i, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) if (!gdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) desc_count += ext4_free_inodes_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) return desc_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) /* Called at mount-time, super-block is locked */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) unsigned long ext4_count_dirs(struct super_block * sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) unsigned long count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) ext4_group_t i, ngroups = ext4_get_groups_count(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) for (i = 0; i < ngroups; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) if (!gdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) count += ext4_used_dirs_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) return count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) * Zeroes not yet zeroed inode table - just write zeroes through the whole
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) * inode table. Must be called without any spinlock held. The only place
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) * where it is called from on active part of filesystem is ext4lazyinit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) * thread, so we do not need any special locks, however we have to prevent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) * inode allocation from the current group, so we take alloc_sem lock, to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) * block ext4_new_inode() until we are finished.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) int ext4_init_inode_table(struct super_block *sb, ext4_group_t group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) int barrier)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) struct ext4_group_info *grp = ext4_get_group_info(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) struct ext4_sb_info *sbi = EXT4_SB(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) struct ext4_group_desc *gdp = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516) struct buffer_head *group_desc_bh;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) handle_t *handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) ext4_fsblk_t blk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) int num, ret = 0, used_blks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) unsigned long used_inos = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) /* This should not happen, but just to be sure check this */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) if (sb_rdonly(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) gdp = ext4_get_group_desc(sb, group, &group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) if (!gdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) * We do not need to lock this, because we are the only one
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) * handling this flag.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) if (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) if (IS_ERR(handle)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) ret = PTR_ERR(handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) down_write(&grp->alloc_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) * If inode bitmap was already initialized there may be some
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) * used inodes so we need to skip blocks with used inodes in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) * inode table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) if (!(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) used_inos = EXT4_INODES_PER_GROUP(sb) -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) ext4_itable_unused_count(sb, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) used_blks = DIV_ROUND_UP(used_inos, sbi->s_inodes_per_block);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) /* Bogus inode unused count? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) if (used_blks < 0 || used_blks > sbi->s_itb_per_group) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) ext4_error(sb, "Something is wrong with group %u: "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) "used itable blocks: %d; "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) "itable unused count: %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) group, used_blks,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) ext4_itable_unused_count(sb, gdp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) goto err_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) used_inos += group * EXT4_INODES_PER_GROUP(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) * Are there some uninitialized inodes in the inode table
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) * before the first normal inode?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) if ((used_blks != sbi->s_itb_per_group) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) (used_inos < EXT4_FIRST_INO(sb))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) ext4_error(sb, "Something is wrong with group %u: "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) "itable unused count: %u; "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) "itables initialized count: %ld",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) group, ext4_itable_unused_count(sb, gdp),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) used_inos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) goto err_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) blk = ext4_inode_table(sb, gdp) + used_blks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) num = sbi->s_itb_per_group - used_blks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) BUFFER_TRACE(group_desc_bh, "get_write_access");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) ret = ext4_journal_get_write_access(handle,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) goto err_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) * Skip zeroout if the inode table is full. But we set the ZEROED
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) * flag anyway, because obviously, when it is full it does not need
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) * further zeroing.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) if (unlikely(num == 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) goto skip_zeroout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) ext4_debug("going to zero out inode table in group %d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) ret = sb_issue_zeroout(sb, blk, num, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) goto err_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) if (barrier)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) blkdev_issue_flush(sb->s_bdev, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) skip_zeroout:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) ext4_lock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) gdp->bg_flags |= cpu_to_le16(EXT4_BG_INODE_ZEROED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) ext4_group_desc_csum_set(sb, group, gdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) ext4_unlock_group(sb, group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) BUFFER_TRACE(group_desc_bh,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) "call ext4_handle_dirty_metadata");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) ret = ext4_handle_dirty_metadata(handle, NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) group_desc_bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) err_out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) up_write(&grp->alloc_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) ext4_journal_stop(handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) }