From 034dd340b08be1f2f0477ad16131d609f9dbd53c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 30 Aug 2026 09:22:00 -0700 Subject: Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull tracing fixes from Steven Rostedt: - Fix error output of boot instance creation failure Currently if a boot instance creation fails, instead of printing out the name of the instance that failed, it prints "(null)". That is because it prints "cur_str" that had already been processed by strsep(). Print the saved name instead. While at it, print the error code of the failure. - Fix use-after-free for same named historgrams Histograms can be named so that they can be used in multiple events. But if the named histogram has a variable attached, the second event that uses the named histogram which duplicates it and needs to free the original after duplication leaves the old variable in place and still visible. If another histogram uses than variable, it will use the stale one which will try to reference the freed duplicate histogram and crash the kernel. Free the duplicate variables along with the duplicated histogram data. - Check return value of kthread_run() in event self test The events self tests uses a kthread for testing but does not check if it succeeded in creating a kthread. If the kthread creation were to fail, the code will still try to call kthread_stop() on the error returned. - Fix race between reading trace_pipe and updating subbuffer size If a user is reading the trace_pipe file at the same time they update the ring buffer sub-buffer size, can cause the trace_pipe read to read stale data. Add trace_access_lock() around updating the ring buffer sub-buffer size. - Fix eventfs_inode on failure path in creation of the events directory In the creation of the "events" directory, if after allocating the eventfs_inode a failure is detected, it calls cleanup_ei() which calls free_ei(). The free_ei() will test if eventfs_inode being freed has no children. It is a bug if it does. But on the failure case of the creation of the "events" directory, the children lists have not yet been initialized and the free will trigger a warning because list_empty() on an uninitialized list returns false. Move the initialization into init_ei() where it makes more sense and makes sure that a created eventfs_inode has its lists initialized upon creation. - Check return value of kthread_run() in ftrace direct sample code The sample code that shows how to use the ftrace direct calls does not test the return of kthread_run() to see if it succeeds. Return a failure if the kthread_run() doesn't succeed. - Clear user events state on fork in case of alloc failure On fork, the child gets a pointer to the parent's user events state. It makes a copy of it then updates the child's pointer to it. But if the allocation fails, the duplication function leaves the child with a pointer to its parent's descriptor. When the child cleans up its data, it will free the parent's descriptor while the parent is still using it. In the duplication function, set the child's user_event_mm to NULL before testing if the allocation succeeded, and when it exits it will not free the parent's descriptor. - Fix retry exhaustion in simple ring buffer reader swap simple_ring_buffer_swap_reader_page() starts with retry set to 8 and post-decrements it only after a failed link replacement. On the final attempt, a successful replacement leaves retry at zero, while a failed replacement leaves it at -1. But the check for success expects the retry value to be non-zero and exits with an error on zero. This is the opposite result. Fix it. - Fail nicely when the remote swap_reader_page() returns an error Currently, if the swap_reader_page() of a remote buffer fails, it triggers a WARN_ON_ONCE() and continues normally. Instead, have it exit with an error and a pr_warn() print instead of a full WARNING. * tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: ring-buffer: Stop remote reader update when page swap fails tracing: Fix retry exhaustion in simple ring buffer reader swap tracing/user_events: Clear copied tracing state before fork duplication samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify eventfs: Initialize ei->children and ei->list in init_ei() tracing: Fix use-after-free in trace_pipe read on sub-buffer order change tracing: Fix crash passing ERR_PTR to kthread_stop() tracing: Fix use-after-free with same-name named triggers tracing: Fix logged instance name on creation failure --- fs/ntfs3/dir.c | 747 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 747 insertions(+) create mode 100644 fs/ntfs3/dir.c (limited to 'fs/ntfs3/dir.c') diff --git a/fs/ntfs3/dir.c b/fs/ntfs3/dir.c new file mode 100644 index 000000000..eb9152e9f --- /dev/null +++ b/fs/ntfs3/dir.c @@ -0,0 +1,747 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * + * Copyright (C) 2019-2021 Paragon Software GmbH, All rights reserved. + * + * Directory handling functions for NTFS-based filesystems. + * + */ + +#include +#include +#include + +#include "debug.h" +#include "ntfs.h" +#include "ntfs_fs.h" + +/* Convert little endian UTF-16 to NLS string. */ +int ntfs_utf16_to_nls(struct ntfs_sb_info *sbi, const __le16 *name, u32 len, + u8 *buf, int buf_len) +{ + int ret, warn; + u8 *op; + struct nls_table *nls = sbi->options->nls; + + static_assert(sizeof(wchar_t) == sizeof(__le16)); + + if (buf_len <= 0) + return -EINVAL; + + buf_len -= 1; + + if (!nls) { + /* UTF-16 -> UTF-8 */ + ret = utf16s_to_utf8s((wchar_t *)name, len, UTF16_LITTLE_ENDIAN, + buf, buf_len); + buf[ret] = '\0'; + return ret; + } + + op = buf; + warn = 0; + + while (len--) { + u16 ec; + int charlen; + char dump[5]; + + if (buf_len < NLS_MAX_CHARSET_SIZE) { + ntfs_warn(sbi->sb, + "filename was truncated while converting."); + break; + } + + ec = le16_to_cpu(*name++); + charlen = nls->uni2char(ec, op, buf_len); + + if (charlen > 0) { + op += charlen; + buf_len -= charlen; + continue; + } + + *op++ = '_'; + buf_len -= 1; + if (warn) + continue; + + warn = 1; + hex_byte_pack(&dump[0], ec >> 8); + hex_byte_pack(&dump[2], ec); + dump[4] = 0; + + ntfs_err(sbi->sb, "failed to convert \"%s\" to %s", dump, + nls->charset); + } + + *op = '\0'; + return op - buf; +} + +// clang-format off +#define PLANE_SIZE 0x00010000 + +#define SURROGATE_PAIR 0x0000d800 +#define SURROGATE_LOW 0x00000400 +#define SURROGATE_BITS 0x000003ff +// clang-format on + +/* + * put_utf16 - Modified version of put_utf16 from fs/nls/nls_base.c + * + * Function is sparse warnings free. + */ +static inline void put_utf16(wchar_t *s, unsigned int c, + enum utf16_endian endian) +{ + static_assert(sizeof(wchar_t) == sizeof(__le16)); + static_assert(sizeof(wchar_t) == sizeof(__be16)); + + switch (endian) { + default: + *s = (wchar_t)c; + break; + case UTF16_LITTLE_ENDIAN: + *(__le16 *)s = __cpu_to_le16(c); + break; + case UTF16_BIG_ENDIAN: + *(__be16 *)s = __cpu_to_be16(c); + break; + } +} + +/* + * _utf8s_to_utf16s + * + * Modified version of 'utf8s_to_utf16s' allows to + * detect -ENAMETOOLONG without writing out of expected maximum. + */ +static int _utf8s_to_utf16s(const u8 *s, int inlen, enum utf16_endian endian, + wchar_t *pwcs, int maxout) +{ + u16 *op; + int size; + unicode_t u; + + op = pwcs; + while (inlen > 0 && *s) { + if (*s & 0x80) { + size = utf8_to_utf32(s, inlen, &u); + if (size < 0) + return -EINVAL; + s += size; + inlen -= size; + + if (u >= PLANE_SIZE) { + if (maxout < 2) + return -ENAMETOOLONG; + + u -= PLANE_SIZE; + put_utf16(op++, + SURROGATE_PAIR | + ((u >> 10) & SURROGATE_BITS), + endian); + put_utf16(op++, + SURROGATE_PAIR | SURROGATE_LOW | + (u & SURROGATE_BITS), + endian); + maxout -= 2; + } else { + if (maxout < 1) + return -ENAMETOOLONG; + + put_utf16(op++, u, endian); + maxout--; + } + } else { + if (maxout < 1) + return -ENAMETOOLONG; + + put_utf16(op++, *s++, endian); + inlen--; + maxout--; + } + } + return op - pwcs; +} + +/* + * ntfs_nls_to_utf16 - Convert input string to UTF-16. + * @name: Input name. + * @name_len: Input name length. + * @uni: Destination memory. + * @max_ulen: Destination memory. + * @endian: Endian of target UTF-16 string. + * + * This function is called: + * - to create NTFS name + * - to create symlink + * + * Return: UTF-16 string length or error (if negative). + */ +int ntfs_nls_to_utf16(struct ntfs_sb_info *sbi, const u8 *name, u32 name_len, + struct cpu_str *uni, u32 max_ulen, + enum utf16_endian endian) +{ + int ret, slen, i; + const u8 *end; + struct nls_table *nls = sbi->options->nls; + u16 *uname = uni->name; + + static_assert(sizeof(wchar_t) == sizeof(u16)); + + if (!nls) { + /* utf8 -> utf16 */ + ret = _utf8s_to_utf16s(name, name_len, endian, uname, max_ulen); + } else { + for (ret = 0, end = name + name_len; name < end; + ret++, name += slen) { + if (ret >= max_ulen) + return -ENAMETOOLONG; + + slen = nls->char2uni(name, end - name, uname + ret); + if (!slen) + return -EINVAL; + if (slen < 0) + return slen; + } + +#ifdef __BIG_ENDIAN + if (endian == UTF16_LITTLE_ENDIAN) { + i = ret; + + while (i--) { + __cpu_to_le16s(uname); + uname++; + } + } +#else + if (endian == UTF16_BIG_ENDIAN) { + i = ret; + + while (i--) { + __cpu_to_be16s(uname); + uname++; + } + } +#endif + } + + uni->len = ret; + uni->ads_len = 0; + if (ret > 0 && sbi->options->ads) { + uname = uni->name; + /* Find delimiter in range [1 : ret-2). */ + for (i = 1; i + 1 < ret; i++) { + if (uname[i] == ':') { + uni->ads_len = ret - i - 1; + uni->len = i; + uname[i] = 0; + ret = i; + + uname += i + 1; + i = uni->ads_len; + /* Return ADS name as little endian. Always */ +#ifdef __BIG_ENDIAN + if (endian == UTF16_LITTLE_ENDIAN) { + while (i--) { + __cpu_to_le16s(uname); + uname++; + } + } +#else + if (endian == UTF16_BIG_ENDIAN) { + while (i--) { + __cpu_to_be16s(uname); + uname++; + } + } +#endif + break; + } + } + } + + return ret; +} + +/* + * dir_search_u - Helper function. + */ +struct inode *dir_search_flags(struct inode *dir, const struct cpu_str *uni, + struct ntfs_fnd *fnd, u32 flags) +{ + int err = 0; + struct super_block *sb = dir->i_sb; + struct ntfs_sb_info *sbi = sb->s_fs_info; + struct ntfs_inode *ni = ntfs_i(dir); + struct NTFS_DE *e; + int diff; + struct inode *inode = NULL; + struct ntfs_fnd *fnd_a = NULL; + + if (!fnd) { + fnd_a = fnd_get(); + if (!fnd_a) { + err = -ENOMEM; + goto out; + } + fnd = fnd_a; + } + + err = indx_find(&ni->dir, ni, NULL, uni, 0, sbi, &diff, &e, fnd); + + if (err) + goto out; + + if (diff) { + err = -ENOENT; + goto out; + } + + inode = ntfs_iget5_flags(sb, &e->ref, uni, flags); + if (!IS_ERR(inode) && is_bad_inode(inode)) { + iput(inode); + err = -EINVAL; + } +out: + fnd_put(fnd_a); + + return err == -ENOENT ? NULL : err ? ERR_PTR(err) : inode; +} + +static inline bool de_fname_fits(const struct NTFS_DE *e, u32 e_size, + const struct ATTR_FILE_NAME *fname) +{ + return sizeof(struct NTFS_DE) + fname_full_size(fname) <= e_size; +} + +/* + * returns false if 'ctx' if full + */ +static inline bool ntfs_dir_emit(struct ntfs_sb_info *sbi, + struct ntfs_inode *ni, const struct NTFS_DE *e, + u8 *name, struct dir_context *ctx) +{ + const struct ATTR_FILE_NAME *fname; + u64 ino; + int name_len; + u32 dt_type; + + fname = Add2Ptr(e, sizeof(struct NTFS_DE)); + + if (fname->type == FILE_NAME_DOS) + return true; + + if (!mi_is_ref(&ni->mi, &fname->home)) + return true; + + ino = ino_get(&e->ref); + + if (ino == MFT_REC_ROOT) + return true; + + /* Skip meta files. Unless option to show metafiles is set. */ + if (!sbi->options->showmeta && ntfs_is_meta_file(sbi, ino)) + return true; + + if (sbi->options->nohidden && (fname->dup.fa & FILE_ATTRIBUTE_HIDDEN)) + return true; + + if (!de_fname_fits(e, le16_to_cpu(e->size), fname)) + return true; + + name_len = ntfs_utf16_to_nls(sbi, fname->name, fname->name_len, name, + PATH_MAX); + if (name_len <= 0) { + ntfs_warn(sbi->sb, "failed to convert name for inode %llx.", + ino); + return true; + } + + /* + * NTFS: symlinks are "dir + reparse" or "file + reparse" + * Unfortunately reparse attribute is used for many purposes (several dozens). + * It is not possible here to know is this name symlink or not. + * To get exactly the type of name we should to open inode (read mft). + * getattr for opened file (fstat) correctly returns symlink. + */ + dt_type = (fname->dup.fa & FILE_ATTRIBUTE_DIRECTORY) ? DT_DIR : DT_REG; + + /* + * It is not reliable to detect the type of name using duplicated information + * stored in parent directory. + * The only correct way to get the type of name - read MFT record and find ATTR_STD. + * The code below is not good idea. + * It does additional locks/reads just to get the type of name. + * Should we use additional mount option to enable branch below? + */ + if (fname->dup.extend_data && ino != ni->mi.rno) { + struct inode *inode = ntfs_iget5(sbi->sb, &e->ref, NULL); + if (!IS_ERR_OR_NULL(inode)) { + dt_type = fs_umode_to_dtype(inode->i_mode); + iput(inode); + } + } + + return dir_emit(ctx, (s8 *)name, name_len, ino, dt_type); +} + +/* + * ntfs_read_hdr - Helper function for ntfs_readdir(). + * + * returns 0 if ok. + * returns -EINVAL if directory is corrupted. + * returns +1 if 'ctx' is full. + */ +static int ntfs_read_hdr(struct ntfs_sb_info *sbi, struct ntfs_inode *ni, + const struct INDEX_HDR *hdr, u64 vbo, u64 pos, + u8 *name, struct dir_context *ctx) +{ + const struct NTFS_DE *e; + u32 e_size; + u32 end = le32_to_cpu(hdr->used); + u32 off = le32_to_cpu(hdr->de_off); + + for (;; off += e_size) { + if (off + sizeof(struct NTFS_DE) > end) + return -EINVAL; + + e = Add2Ptr(hdr, off); + e_size = le16_to_cpu(e->size); + if (e_size < sizeof(struct NTFS_DE) || off + e_size > end) + return -EINVAL; + + if (de_is_last(e)) + return 0; + + /* Skip already enumerated. */ + if (vbo + off < pos) + continue; + + if (le16_to_cpu(e->key_size) < SIZEOF_ATTRIBUTE_FILENAME) + return -EINVAL; + + ctx->pos = vbo + off; + + /* Submit the name to the filldir callback. */ + if (!ntfs_dir_emit(sbi, ni, e, name, ctx)) { + /* ctx is full. */ + return +1; + } + } +} + +/* + * ntfs_readdir - file_operations::iterate_shared + * + * Use non sorted enumeration. + * Sorted enumeration may result infinite loop if names tree contains loop. + */ +static int ntfs_readdir(struct file *file, struct dir_context *ctx) +{ + const struct INDEX_ROOT *root; + size_t bit; + int err = 0; + struct inode *dir = file_inode(file); + struct ntfs_inode *ni = ntfs_i(dir); + struct super_block *sb = dir->i_sb; + struct ntfs_sb_info *sbi = sb->s_fs_info; + loff_t i_size = i_size_read(dir); + u64 pos = ctx->pos; + u8 *name = NULL; + struct indx_node *node = NULL; + u8 index_bits = ni->dir.index_bits; + size_t max_bit = i_size >> ni->dir.index_bits; + loff_t eod = i_size + sbi->record_size; + + /* Name is a buffer of PATH_MAX length. */ + static_assert(NTFS_NAME_LEN * 4 < PATH_MAX); + + if (!pos) { + /* + * ni->dir.version increments each directory change. + * Save the initial value of ni->dir.version. + */ + file->private_data = (void *)ni->dir.version; + } + + if (pos >= eod) { + if (file->private_data == (void *)ni->dir.version) { + /* No changes since first readdir. */ + return 0; + } + + /* + * Handle directories that changed after the initial readdir(). + * + * Some user space code implements recursive removal like this instead + * of calling rmdir(2) directly: + * + * fd = opendir(path); + * while ((dent = readdir(fd))) + * unlinkat(dirfd(fd), dent->d_name, 0); + * closedir(fd); + * + * POSIX leaves unspecified what readdir() should return once the + * directory has been modified after opendir()/rewinddir(), so this + * pattern is not guaranteed to work on all filesystems or platforms. + * + * In ntfs3 the internal name tree may be reshaped while entries are + * being removed, so there is no stable anchor for continuing a + * single-pass walk based on the original readdir() order. + * + * In practice some widely used tools (for example certain rm(1) + * implementations) have used this readdir()/unlink() loop, and some + * filesystems behave in a way that effectively makes it work in the + * common case. + * + * The code below follows that practice and tries to provide + * "rmdir-like" behaviour for such callers on ntfs3, even though the + * situation is not strictly defined by the APIs. + * + * Apple documents the same readdir()/unlink() issue and a workaround + * for HFS file systems in: + * https://web.archive.org/web/20220122122948/https:/support.apple.com/kb/TA21420?locale=en_US + */ + ctx->pos = pos = 3; + file->private_data = (void *)ni->dir.version; + } + + if (!dir_emit_dots(file, ctx)) + return 0; + + name = kmalloc(PATH_MAX, GFP_KERNEL); + if (!name) + return -ENOMEM; + + if (!ni->mi_loaded && ni->attr_list.size) { + /* + * Directory inode is locked for read. + * Load all subrecords to avoid 'write' access to 'ni' during + * directory reading. + */ + ni_lock(ni); + if (!ni->mi_loaded && ni->attr_list.size) { + err = ni_load_all_mi(ni); + if (!err) + ni->mi_loaded = true; + } + ni_unlock(ni); + if (err) + goto out; + } + + /* + * Keep directory metadata stable for the whole walk. Loading subrecords + * once is not enough if concurrent writeback can still compact ATTR_LIST + * entries and free the record that ntfs_read_hdr() is currently walking. + */ + ni_lock(ni); + + root = indx_get_root(&ni->dir, ni, NULL, NULL); + if (!root) { + err = -EINVAL; + goto out_unlock; + } + + if (pos >= sbi->record_size) { + bit = (pos - sbi->record_size) >> index_bits; + } else { + /* + * Add each name from root in 'ctx'. + */ + err = ntfs_read_hdr(sbi, ni, &root->ihdr, 0, pos, name, ctx); + if (err) + goto out_unlock; + bit = 0; + } + + /* + * Enumerate indexes until the end of dir. + */ + for (; bit < max_bit; bit += 1) { + /* Get the next used index. */ + err = indx_used_bit(&ni->dir, ni, &bit); + if (err) + goto out_unlock; + + if (bit == MINUS_ONE_T) { + /* no more used indexes. end of dir. */ + break; + } + + if (bit >= max_bit) { + /* Corrupted directory. */ + err = -EINVAL; + goto out_unlock; + } + + err = indx_read_ra(&ni->dir, ni, bit << ni->dir.idx2vbn_bits, + &node, &file->f_ra); + if (err) + goto out_unlock; + + /* + * Add each name from index in 'ctx'. + */ + err = ntfs_read_hdr(sbi, ni, &node->index->ihdr, + ((u64)bit << index_bits) + sbi->record_size, + pos, name, ctx); + if (err) + goto out_unlock; + } + +out_unlock: + ni_unlock(ni); + +out: + kfree(name); + put_indx_node(node); + + if (!err) { + /* End of directory. */ + ctx->pos = eod; + } else if (err == 1) { + /* 'ctx' is full. */ + err = 0; + } else if (err == -ENOENT) { + err = 0; + ctx->pos = pos; + } else if (err < 0) { + if (err == -EINVAL) + _ntfs_bad_inode(dir); + ctx->pos = eod; + } + + return err; +} + +/* + * Return fname when @e passes the same checks as ntfs_dir_emit() before + * exposing an entry (valid key, non-DOS, fname fits in e->size). + */ +static inline const struct ATTR_FILE_NAME * +de_countable_fname(const struct NTFS_DE *e, u32 e_size) +{ + const struct ATTR_FILE_NAME *fname; + + fname = de_get_fname(e); + if (!fname || fname->type == FILE_NAME_DOS || + !de_fname_fits(e, e_size, fname)) + return NULL; + + return fname; +} + +static int ntfs_dir_count(struct inode *dir, bool *is_empty, size_t *dirs, + size_t *files) +{ + int err = 0; + struct ntfs_inode *ni = ntfs_i(dir); + struct NTFS_DE *e = NULL; + struct INDEX_ROOT *root; + struct INDEX_HDR *hdr; + const struct ATTR_FILE_NAME *fname; + u32 e_size, off, end; + size_t drs = 0, fles = 0, bit = 0; + struct indx_node *node = NULL; + size_t max_indx = i_size_read(&ni->vfs_inode) >> ni->dir.index_bits; + + if (is_empty) + *is_empty = true; + + root = indx_get_root(&ni->dir, ni, NULL, NULL); + if (!root) + return -EINVAL; + + hdr = &root->ihdr; + + for (;;) { + end = le32_to_cpu(hdr->used); + off = le32_to_cpu(hdr->de_off); + + for (; off + sizeof(struct NTFS_DE) <= end; off += e_size) { + e = Add2Ptr(hdr, off); + e_size = le16_to_cpu(e->size); + if (e_size < sizeof(struct NTFS_DE) || + off + e_size > end) { + /* Looks like corruption. */ + break; + } + + if (de_is_last(e)) + break; + + fname = de_countable_fname(e, e_size); + if (!fname) + continue; + + if (is_empty) { + *is_empty = false; + if (!dirs && !files) + goto out; + } + + if (fname->dup.fa & FILE_ATTRIBUTE_DIRECTORY) + drs += 1; + else + fles += 1; + } + + if (bit >= max_indx) + goto out; + + err = indx_used_bit(&ni->dir, ni, &bit); + if (err) + goto out; + + if (bit == MINUS_ONE_T) + goto out; + + if (bit >= max_indx) + goto out; + + err = indx_read(&ni->dir, ni, bit << ni->dir.idx2vbn_bits, + &node); + if (err) + goto out; + + hdr = &node->index->ihdr; + bit += 1; + } + +out: + put_indx_node(node); + if (dirs) + *dirs = drs; + if (files) + *files = fles; + + return err; +} + +bool dir_is_empty(struct inode *dir) +{ + bool is_empty = false; + + ntfs_dir_count(dir, &is_empty, NULL, NULL); + + return is_empty; +} + +// clang-format off +const struct file_operations ntfs_dir_operations = { + .llseek = generic_file_llseek, + .read = generic_read_dir, + .iterate_shared = ntfs_readdir, + .fsync = ntfs_file_fsync, + .open = ntfs_file_open, + .unlocked_ioctl = ntfs_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = ntfs_compat_ioctl, +#endif + .setlease = generic_setlease, +}; +// clang-format on -- cgit v1.3.1