From: Darrick J. Wong <darrick.wong@xxxxxxxxxx> Add to xfs_db the ability to list a directory. Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx> --- db/namei.c | 380 ++++++++++++++++++++++++++++++++++++++++++++++ libxfs/libxfs_api_defs.h | 1 man/man8/xfs_db.8 | 14 ++ 3 files changed, 395 insertions(+) diff --git a/db/namei.c b/db/namei.c index 3c9889d62338..b2c036e6777a 100644 --- a/db/namei.c +++ b/db/namei.c @@ -221,8 +221,388 @@ static const cmdinfo_t path_cmd = { .help = path_help, }; +/* List a directory's entries. */ + +static const char *filetype_strings[XFS_DIR3_FT_MAX] = { + [XFS_DIR3_FT_UNKNOWN] = N_("unknown"), + [XFS_DIR3_FT_REG_FILE] = N_("regular"), + [XFS_DIR3_FT_DIR] = N_("directory"), + [XFS_DIR3_FT_CHRDEV] = N_("chardev"), + [XFS_DIR3_FT_BLKDEV] = N_("blkdev"), + [XFS_DIR3_FT_FIFO] = N_("fifo"), + [XFS_DIR3_FT_SOCK] = N_("socket"), + [XFS_DIR3_FT_SYMLINK] = N_("symlink"), + [XFS_DIR3_FT_WHT] = N_("whiteout"), +}; + +static const char * +get_dstr( + struct xfs_mount *mp, + uint8_t filetype) +{ + if (!xfs_sb_version_hasftype(&mp->m_sb)) + return filetype_strings[XFS_DIR3_FT_UNKNOWN]; + + if (filetype >= XFS_DIR3_FT_MAX) + return filetype_strings[XFS_DIR3_FT_UNKNOWN]; + + return filetype_strings[filetype]; +} + +static void +dir_emit( + struct xfs_mount *mp, + char *name, + ssize_t namelen, + xfs_ino_t ino, + uint8_t dtype) +{ + char *display_name; + struct xfs_name xname = { .name = name }; + const char *dstr = get_dstr(mp, dtype); + xfs_dahash_t hash; + bool good; + + if (namelen < 0) { + /* Negative length means that name is null-terminated. */ + display_name = name; + xname.len = strlen(name); + good = true; + } else { + /* + * Otherwise, name came from a directory entry, so we have to + * copy the string to a buffer so that we can add the null + * terminator. + */ + display_name = malloc(namelen + 1); + memcpy(display_name, name, namelen); + display_name[namelen] = 0; + xname.len = namelen; + good = libxfs_dir2_namecheck(name, namelen); + } + hash = libxfs_dir2_hashname(mp, &xname); + + dbprintf("%-18llu %-14s 0x%08llx %3d %s", ino, dstr, hash, xname.len, + display_name); + if (!good) + dbprintf(_(" (corrupt)")); + dbprintf("\n"); + + if (display_name != name) + free(display_name); +} + +static int +list_sfdir( + struct xfs_da_args *args) +{ + struct xfs_inode *dp = args->dp; + struct xfs_mount *mp = dp->i_mount; + struct xfs_dir2_sf_entry *sfep; + struct xfs_dir2_sf_hdr *sfp; + xfs_ino_t ino; + unsigned int i; + uint8_t filetype; + + sfp = (struct xfs_dir2_sf_hdr *)dp->i_df.if_u1.if_data; + + /* . and .. entries */ + dir_emit(args->dp->i_mount, ".", -1, dp->i_ino, XFS_DIR3_FT_DIR); + + ino = libxfs_dir2_sf_get_parent_ino(sfp); + dir_emit(args->dp->i_mount, "..", -1, ino, XFS_DIR3_FT_DIR); + + /* Walk everything else. */ + sfep = xfs_dir2_sf_firstentry(sfp); + for (i = 0; i < sfp->count; i++) { + ino = libxfs_dir2_sf_get_ino(mp, sfp, sfep); + filetype = libxfs_dir2_sf_get_ftype(mp, sfep); + + dir_emit(args->dp->i_mount, (char *)sfep->name, sfep->namelen, + ino, filetype); + sfep = libxfs_dir2_sf_nextentry(mp, sfp, sfep); + } + + return 0; +} + +/* List entries in block format directory. */ +static int +list_blockdir( + struct xfs_da_args *args) +{ + struct xfs_inode *dp = args->dp; + struct xfs_mount *mp = dp->i_mount; + struct xfs_buf *bp; + struct xfs_da_geometry *geo = mp->m_dir_geo; + unsigned int offset; + unsigned int end; + int error; + + error = xfs_dir3_block_read(NULL, dp, &bp); + if (error) + return error; + + end = xfs_dir3_data_end_offset(geo, bp->b_addr); + for (offset = geo->data_entry_offset; offset < end;) { + struct xfs_dir2_data_unused *dup = bp->b_addr + offset; + struct xfs_dir2_data_entry *dep = bp->b_addr + offset; + uint8_t filetype; + + if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) { + /* Unused entry */ + offset += be16_to_cpu(dup->length); + continue; + } + + /* Real entry */ + offset += libxfs_dir2_data_entsize(mp, dep->namelen); + filetype = libxfs_dir2_data_get_ftype(dp->i_mount, dep); + dir_emit(mp, (char *)dep->name, dep->namelen, + be64_to_cpu(dep->inumber), filetype); + } + + libxfs_trans_brelse(args->trans, bp); + return error; +} + +/* List entries in leaf format directory. */ +static int +list_leafdir( + struct xfs_da_args *args) +{ + struct xfs_bmbt_irec map; + struct xfs_iext_cursor icur; + struct xfs_inode *dp = args->dp; + struct xfs_mount *mp = dp->i_mount; + struct xfs_buf *bp = NULL; + struct xfs_ifork *ifp = XFS_IFORK_PTR(dp, XFS_DATA_FORK); + struct xfs_da_geometry *geo = mp->m_dir_geo; + xfs_dablk_t dabno = 0; + int error = 0; + + /* Read extent map. */ + if (!(ifp->if_flags & XFS_IFEXTENTS)) { + error = -libxfs_iread_extents(NULL, dp, XFS_DATA_FORK); + if (error) + return error; + } + + while (dabno < geo->leafblk) { + unsigned int offset; + unsigned int length; + + /* Find mapping for leaf block. */ + if (!xfs_iext_lookup_extent(dp, ifp, dabno, &icur, &map)) + break; + if (map.br_startoff >= geo->leafblk) + break; + libxfs_trim_extent(&map, dabno, geo->leafblk - dabno); + + /* Read the directory block of that first mapping. */ + error = xfs_dir3_data_read(NULL, dp, map.br_startoff, 0, &bp); + if (error) + break; + + for (offset = geo->data_entry_offset; offset < geo->blksize;) { + struct xfs_dir2_data_entry *dep; + struct xfs_dir2_data_unused *dup; + uint8_t filetype; + + dup = bp->b_addr + offset; + dep = bp->b_addr + offset; + + if (be16_to_cpu(dup->freetag) == + XFS_DIR2_DATA_FREE_TAG) { + /* Skip unused entry */ + length = be16_to_cpu(dup->length); + offset += length; + continue; + } + + offset += libxfs_dir2_data_entsize(mp, dep->namelen); + filetype = libxfs_dir2_data_get_ftype(mp, dep); + + dir_emit(mp, (char *)dep->name, dep->namelen, + be64_to_cpu(dep->inumber), filetype); + } + + dabno += XFS_DADDR_TO_FSB(mp, bp->b_length); + libxfs_buf_relse(bp); + bp = NULL; + } + + if (bp) + libxfs_buf_relse(bp); + + return error; +} + +/* Read the directory, display contents. */ +int +listdir( + struct xfs_inode *dp) +{ + struct xfs_da_args args = { + .dp = dp, + .geo = dp->i_mount->m_dir_geo, + }; + int error; + int isblock; + + if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) + return list_sfdir(&args); + + error = -libxfs_dir2_isblock(&args, &isblock); + if (error) + return error; + + if (isblock) + return list_blockdir(&args); + return list_leafdir(&args); +} + +/* If the io cursor points to a directory, list its contents. */ +static int +ls_cur( + char *tag, + bool direct) +{ + struct xfs_inode *dp; + int ret = 0; + + if (iocur_top->typ != &typtab[TYP_INODE]) { + dbprintf(_("current object is not an inode.\n")); + return -1; + } + + ret = -libxfs_iget(mp, NULL, iocur_top->ino, 0, &dp); + if (ret) { + dbprintf(_("failed to iget directory %llu, error %d\n"), + (unsigned long long)iocur_top->ino, ret); + return -1; + } + + if (S_ISDIR(VFS_I(dp)->i_mode) && !direct) { + /* List the contents of a directory. */ + if (tag) + dbprintf(_("%s:\n"), tag); + + ret = listdir(dp); + if (ret) { + dbprintf(_("failed to list directory %llu: %s\n"), + (unsigned long long)iocur_top->ino, + strerror(ret)); + ret = -1; + goto rele; + } + } else if (direct || !S_ISDIR(VFS_I(dp)->i_mode)) { + /* List the directory entry associated with a single file. */ + char inum[32]; + + if (!tag) { + snprintf(inum, sizeof(inum), "<%llu>", + (unsigned long long)iocur_top->ino); + tag = inum; + } else { + char *p = strrchr(tag, '/'); + + if (p) + tag = p + 1; + } + + dir_emit(mp, tag, -1, iocur_top->ino, + libxfs_mode_to_ftype(VFS_I(dp)->i_mode)); + } else { + dbprintf(_("current inode %llu is not a directory.\n"), + (unsigned long long)iocur_top->ino); + ret = -1; + goto rele; + } + +rele: + libxfs_irele(dp); + return ret; +} + +static void +ls_help(void) +{ + dbprintf(_( +"\n" +" List the contents of the currently selected directory inode.\n" +"\n" +" Options:\n" +" -d -- List directories themselves, not their contents.\n" +"\n" +" Directory contents will be listed in the format:\n" +" inode_number type hash name_length name\n" + )); +} + +static int +ls_f( + int argc, + char **argv) +{ + bool direct = false; + int c; + int ret = 0; + + while ((c = getopt(argc, argv, "d")) != -1) { + switch (c) { + case 'd': + direct = true; + break; + default: + ls_help(); + return 0; + } + } + + if (optind == argc) { + if (ls_cur(NULL, direct)) + exitcode = 1; + return 0; + } + + for (c = optind; c < argc; c++) { + push_cur(); + + ret = path_walk(argv[c]); + if (ret) + goto err_cur; + + ret = ls_cur(argv[c], direct); + if (ret) + goto err_cur; + + pop_cur(); + } + + return 0; +err_cur: + pop_cur(); + if (ret) + exitcode = 1; + return 0; +} + +static const cmdinfo_t ls_cmd = { + .name = "ls", + .altname = "l", + .cfunc = ls_f, + .argmin = 0, + .argmax = -1, + .canpush = 0, + .args = "[-d] [paths...]", + .oneline = N_("list directory contents"), + .help = ls_help, +}; + void namei_init(void) { add_command(&path_cmd); + add_command(&ls_cmd); } diff --git a/libxfs/libxfs_api_defs.h b/libxfs/libxfs_api_defs.h index e7e42e93a07e..7029d0e7daf7 100644 --- a/libxfs/libxfs_api_defs.h +++ b/libxfs/libxfs_api_defs.h @@ -187,6 +187,7 @@ #define xfs_trans_resv_calc libxfs_trans_resv_calc #define xfs_trans_roll_inode libxfs_trans_roll_inode #define xfs_trans_roll libxfs_trans_roll +#define xfs_trim_extent libxfs_trim_extent #define xfs_verify_agbno libxfs_verify_agbno #define xfs_verify_agino libxfs_verify_agino diff --git a/man/man8/xfs_db.8 b/man/man8/xfs_db.8 index d67e108a706a..fefe862b7564 100644 --- a/man/man8/xfs_db.8 +++ b/man/man8/xfs_db.8 @@ -806,6 +806,20 @@ This makes it easier to find discrepancies in the reservation calculations between xfsprogs and the kernel, which will help when diagnosing minimum log size calculation errors. .TP +.BI "ls [\-d] [" dir_path "]..." +List the contents of the given paths. +If a path resolves to a directory, the directory will be listed. +If a path resolves to a file that is not a directory, the entry for the file +within the parent directory will be listed. +If no paths are supplied and the IO cursor points at a directory inode, +the contents of that directory will be listed. +If no paths are supplied and the IO cursor points at a file that is not a +directory, a directory entry will be synthesized for the file. +If the +.B \-d +option is specified, list the directory itself and not its contents. +The output format is: inode number, file type, hash, name length, name. +.TP .BI "metadump [\-egow] " filename Dumps metadata to a file. See .BR xfs_metadump (8)