This patch fix the infinite loop of FAT for regular file. By this
patch, I think it covered all of the infinite loop points.
Please apply.
-- OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>diff -urN linux-2.5.3-pre2/fs/fat/dir.c fat_loop/fs/fat/dir.c --- linux-2.5.3-pre2/fs/fat/dir.c Sun Jan 20 20:55:42 2002 +++ fat_loop/fs/fat/dir.c Sun Jan 20 21:01:05 2002 @@ -744,8 +744,8 @@ if ((dir->i_ino == MSDOS_ROOT_INO) && (MSDOS_SB(sb)->fat_bits != 32)) return -ENOSPC; new_bh = fat_extend_dir(dir); - if (!new_bh) - return -ENOSPC; + if (IS_ERR(new_bh)) + return PTR_ERR(new_bh); fat_brelse(sb, new_bh); do { fat_get_entry(dir, &curr, bh, de, ino); @@ -761,7 +761,10 @@ struct msdos_dir_entry *de; __u16 date, time; - if ((bh = fat_extend_dir(dir)) == NULL) return -ENOSPC; + bh = fat_extend_dir(dir); + if (IS_ERR(bh)) + return PTR_ERR(bh); + /* zeroed out, so... */ fat_date_unix2dos(dir->i_mtime,&time,&date); de = (struct msdos_dir_entry*)&bh->b_data[0]; diff -urN linux-2.5.3-pre2/fs/fat/file.c fat_loop/fs/fat/file.c --- linux-2.5.3-pre2/fs/fat/file.c Fri Dec 28 01:17:43 2001 +++ fat_loop/fs/fat/file.c Sun Jan 20 21:01:05 2002 @@ -65,8 +65,11 @@ return -EIO; } if (!(iblock % MSDOS_SB(inode->i_sb)->cluster_size)) { - if (fat_add_cluster(inode) < 0) - return -ENOSPC; + int error; + + error = fat_add_cluster(inode); + if (error < 0) + return error; } MSDOS_I(inode)->mmu_private += sb->s_blocksize; phys = fat_bmap(inode, iblock); diff -urN linux-2.5.3-pre2/fs/fat/misc.c fat_loop/fs/fat/misc.c --- linux-2.5.3-pre2/fs/fat/misc.c Sun Jan 20 20:55:42 2002 +++ fat_loop/fs/fat/misc.c Sun Jan 20 21:01:05 2002 @@ -131,14 +131,13 @@ { struct super_block *sb = inode->i_sb; int count, nr, limit, last, curr, file_cluster; - int cluster_size = MSDOS_SB(sb)->cluster_size; - int res = -ENOSPC; + int cluster_bits = MSDOS_SB(sb)->cluster_bits; lock_fat(sb); if (MSDOS_SB(sb)->free_clusters == 0) { unlock_fat(sb); - return res; + return -ENOSPC; } limit = MSDOS_SB(sb)->clusters; nr = limit; /* to keep GCC happy */ @@ -150,7 +149,7 @@ if (count >= limit) { MSDOS_SB(sb)->free_clusters = 0; unlock_fat(sb); - return res; + return -ENOSPC; } MSDOS_SB(sb)->prev_free = (count + MSDOS_SB(sb)->prev_free + 1) % limit; @@ -174,13 +173,20 @@ */ last = file_cluster = 0; if ((curr = MSDOS_I(inode)->i_start) != 0) { + int max_cluster = MSDOS_I(inode)->mmu_private >> cluster_bits; + fat_cache_lookup(inode, INT_MAX, &last, &curr); file_cluster = last; - while (curr && curr != -1){ + while (curr && curr != -1) { file_cluster++; - if (!(curr = fat_access(sb, last = curr,-1))) { + if (!(curr = fat_access(sb, last = curr, -1))) { fat_fs_panic(sb, "File without EOF"); - return res; + return -EIO; + } + if (file_cluster > max_cluster) { + fat_fs_panic(sb,"inode %lu: bad cluster counts", + inode->i_ino); + return -EIO; } } } @@ -192,14 +198,12 @@ MSDOS_I(inode)->i_logstart = nr; mark_inode_dirty(inode); } - if (file_cluster - != inode->i_blocks / cluster_size / (sb->s_blocksize / 512)) { + if (file_cluster != (inode->i_blocks >> (cluster_bits - 9))) { printk ("file_cluster badly computed!!! %d <> %ld\n", - file_cluster, - inode->i_blocks / cluster_size / (sb->s_blocksize / 512)); + file_cluster, inode->i_blocks >> (cluster_bits - 9)); fat_cache_inval_inode(inode); } - inode->i_blocks += (1 << MSDOS_SB(sb)->cluster_bits) / 512; + inode->i_blocks += (1 << cluster_bits) >> 9; return nr; } @@ -213,24 +217,23 @@ if (MSDOS_SB(sb)->fat_bits != 32) { if (inode->i_ino == MSDOS_ROOT_INO) - return res; + return ERR_PTR(-ENOSPC); } nr = fat_add_cluster(inode); if (nr < 0) - return res; + return ERR_PTR(nr); sector = MSDOS_SB(sb)->data_start + (nr - 2) * cluster_size; last_sector = sector + cluster_size; - if (MSDOS_SB(sb)->cvf_format && MSDOS_SB(sb)->cvf_format->zero_out_cluster) + if (MSDOS_SB(sb)->cvf_format + && MSDOS_SB(sb)->cvf_format->zero_out_cluster) { + res = ERR_PTR(-EIO); MSDOS_SB(sb)->cvf_format->zero_out_cluster(inode, nr); - else { + } else { for ( ; sector < last_sector; sector++) { -#ifdef DEBUG - printk("zeroing sector %d\n", sector); -#endif if (!(bh = fat_getblk(sb, sector))) - printk("getblk failed\n"); + printk("FAT: fat_getblk() failed\n"); else { memset(bh->b_data, 0, sb->s_blocksize); fat_set_uptodate(sb, bh, 1); @@ -241,6 +244,8 @@ fat_brelse(sb, bh); } } + if (res == NULL) + res = ERR_PTR(-EIO); } if (inode->i_size & (sb->s_blocksize - 1)) { fat_fs_panic(sb, "Odd directory size"); @@ -484,12 +489,11 @@ **res_de) { int count, cluster; - unsigned long dir_size; + unsigned long dir_size = 0; #ifdef DEBUG printk("raw_scan_nonroot: start=%d\n",start); #endif - dir_size = 0; do { for (count = 0; count < MSDOS_SB(sb)->cluster_size; count++) { if ((cluster = raw_scan_sector(sb,(start-2)* - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/