linux_old1/fs/readdir.c

558 lines
14 KiB
C
Raw Normal View History

License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 22:07:57 +08:00
// SPDX-License-Identifier: GPL-2.0
/*
* linux/fs/readdir.c
*
* Copyright (C) 1995 Linus Torvalds
*/
#include <linux/stddef.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/time.h>
#include <linux/mm.h>
#include <linux/errno.h>
#include <linux/stat.h>
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/fsnotify.h>
#include <linux/dirent.h>
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/unistd.h>
#include <linux/compat.h>
#include <linux/uaccess.h>
#include <asm/unaligned.h>
/*
* Note the "unsafe_put_user() semantics: we goto a
* label for errors.
*/
#define unsafe_copy_dirent_name(_dst, _src, _len, label) do { \
char __user *dst = (_dst); \
const char *src = (_src); \
size_t len = (_len); \
uaccess: implement a proper unsafe_copy_to_user() and switch filldir over to it In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I made filldir() use unsafe_put_user(), which improves code generation on x86 enormously. But because we didn't have a "unsafe_copy_to_user()", the dirent name copy was also done by hand with unsafe_put_user() in a loop, and it turns out that a lot of other architectures didn't like that, because unlike x86, they have various alignment issues. Most non-x86 architectures trap and fix it up, and some (like xtensa) will just fail unaligned put_user() accesses unconditionally. Which makes that "copy using put_user() in a loop" not work for them at all. I could make that code do explicit alignment etc, but the architectures that don't like unaligned accesses also don't really use the fancy "user_access_begin/end()" model, so they might just use the regular old __copy_to_user() interface. So this commit takes that looping implementation, turns it into the x86 version of "unsafe_copy_to_user()", and makes other architectures implement the unsafe copy version as __copy_to_user() (the same way they do for the other unsafe_xyz() accessor functions). Note that it only does this for the copying _to_ user space, and we still don't have a unsafe version of copy_from_user(). That's partly because we have no current users of it, but also partly because the copy_from_user() case is slightly different and cannot efficiently be implemented in terms of a unsafe_get_user() loop (because gcc can't do asm goto with outputs). It would be trivial to do this using "rep movsb", which would work really nicely on newer x86 cores, but really badly on some older ones. Al Viro is looking at cleaning up all our user copy routines to make this all a non-issue, but for now we have this simple-but-stupid version for x86 that works fine for the dirent name copy case because those names are short strings and we simply don't need anything fancier. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Reported-by: Guenter Roeck <linux@roeck-us.net> Reported-and-tested-by: Tony Luck <tony.luck@intel.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Max Filippov <jcmvbkbc@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-08 03:56:48 +08:00
unsafe_put_user(0, dst+len, label); \
unsafe_copy_to_user(dst, src, len, label); \
} while (0)
int iterate_dir(struct file *file, struct dir_context *ctx)
{
struct inode *inode = file_inode(file);
bool shared = false;
int res = -ENOTDIR;
if (file->f_op->iterate_shared)
shared = true;
else if (!file->f_op->iterate)
goto out;
res = security_file_permission(file, MAY_READ);
if (res)
goto out;
if (shared)
res = down_read_killable(&inode->i_rwsem);
else
res = down_write_killable(&inode->i_rwsem);
if (res)
goto out;
res = -ENOENT;
if (!IS_DEADDIR(inode)) {
ctx->pos = file->f_pos;
if (shared)
res = file->f_op->iterate_shared(file, ctx);
else
res = file->f_op->iterate(file, ctx);
file->f_pos = ctx->pos;
fsnotify_access(file);
file_accessed(file);
}
if (shared)
inode_unlock_shared(inode);
else
inode_unlock(inode);
out:
return res;
}
EXPORT_SYMBOL(iterate_dir);
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
/*
* POSIX says that a dirent name cannot contain NULL or a '/'.
*
* It's not 100% clear what we should really do in this case.
* The filesystem is clearly corrupted, but returning a hard
* error means that you now don't see any of the other names
* either, so that isn't a perfect alternative.
*
* And if you return an error, what error do you use? Several
* filesystems seem to have decided on EUCLEAN being the error
* code for EFSCORRUPTED, and that may be the error to use. Or
* just EIO, which is perhaps more obvious to users.
*
* In order to see the other file names in the directory, the
* caller might want to make this a "soft" error: skip the
* entry, and return the error at the end instead.
*
* Note that this should likely do a "memchr(name, 0, len)"
* check too, since that would be filesystem corruption as
* well. However, that case can't actually confuse user space,
* which has to do a strlen() on the name anyway to find the
* filename length, and the above "soft error" worry means
* that it's probably better left alone until we have that
* issue clarified.
*
* Note the PATH_MAX check - it's arbitrary but the real
* kernel limit on a possible path component, not NAME_MAX,
* which is the technical standard limit.
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
*/
static int verify_dirent_name(const char *name, int len)
{
if (len <= 0 || len >= PATH_MAX)
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
return -EIO;
if (memchr(name, '/', len))
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
return -EIO;
return 0;
}
/*
* Traditional linux readdir() handling..
*
* "count=1" is a special case, meaning that the buffer is one
* dirent-structure in size and that the code can't handle more
* anyway. Thus the special "fillonedir()" function for that
* case (the low-level handlers don't need to care about this).
*/
#ifdef __ARCH_WANT_OLD_READDIR
struct old_linux_dirent {
unsigned long d_ino;
unsigned long d_offset;
unsigned short d_namlen;
char d_name[1];
};
struct readdir_callback {
struct dir_context ctx;
struct old_linux_dirent __user * dirent;
int result;
};
static int fillonedir(struct dir_context *ctx, const char *name, int namlen,
loff_t offset, u64 ino, unsigned int d_type)
{
struct readdir_callback *buf =
container_of(ctx, struct readdir_callback, ctx);
struct old_linux_dirent __user * dirent;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
unsigned long d_ino;
if (buf->result)
return -EINVAL;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
d_ino = ino;
if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) {
buf->result = -EOVERFLOW;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
return -EOVERFLOW;
}
buf->result++;
dirent = buf->dirent;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(dirent,
(unsigned long)(dirent->d_name + namlen + 1) -
(unsigned long)dirent))
goto efault;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
if ( __put_user(d_ino, &dirent->d_ino) ||
__put_user(offset, &dirent->d_offset) ||
__put_user(namlen, &dirent->d_namlen) ||
__copy_to_user(dirent->d_name, name, namlen) ||
__put_user(0, dirent->d_name + namlen))
goto efault;
return 0;
efault:
buf->result = -EFAULT;
return -EFAULT;
}
SYSCALL_DEFINE3(old_readdir, unsigned int, fd,
struct old_linux_dirent __user *, dirent, unsigned int, count)
{
int error;
struct fd f = fdget_pos(fd);
struct readdir_callback buf = {
.ctx.actor = fillonedir,
.dirent = dirent
};
if (!f.file)
return -EBADF;
error = iterate_dir(f.file, &buf.ctx);
if (buf.result)
error = buf.result;
fdput_pos(f);
return error;
}
#endif /* __ARCH_WANT_OLD_READDIR */
/*
* New, all-improved, singing, dancing, iBCS2-compliant getdents()
* interface.
*/
struct linux_dirent {
unsigned long d_ino;
unsigned long d_off;
unsigned short d_reclen;
char d_name[1];
};
struct getdents_callback {
struct dir_context ctx;
struct linux_dirent __user * current_dir;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
int prev_reclen;
int count;
int error;
};
static int filldir(struct dir_context *ctx, const char *name, int namlen,
loff_t offset, u64 ino, unsigned int d_type)
{
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
struct linux_dirent __user *dirent, *prev;
struct getdents_callback *buf =
container_of(ctx, struct getdents_callback, ctx);
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
unsigned long d_ino;
int reclen = ALIGN(offsetof(struct linux_dirent, d_name) + namlen + 2,
sizeof(long));
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
int prev_reclen;
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
buf->error = verify_dirent_name(name, namlen);
if (unlikely(buf->error))
return buf->error;
buf->error = -EINVAL; /* only used if we fail.. */
if (reclen > buf->count)
return -EINVAL;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
d_ino = ino;
if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) {
buf->error = -EOVERFLOW;
[PATCH] VFS: Make filldir_t and struct kstat deal in 64-bit inode numbers These patches make the kernel pass 64-bit inode numbers internally when communicating to userspace, even on a 32-bit system. They are required because some filesystems have intrinsic 64-bit inode numbers: NFS3+ and XFS for example. The 64-bit inode numbers are then propagated to userspace automatically where the arch supports it. Problems have been seen with userspace (eg: ld.so) using the 64-bit inode number returned by stat64() or getdents64() to differentiate files, and failing because the 64-bit inode number space was compressed to 32-bits, and so overlaps occur. This patch: Make filldir_t take a 64-bit inode number and struct kstat carry a 64-bit inode number so that 64-bit inode numbers can be passed back to userspace. The stat functions then returns the full 64-bit inode number where available and where possible. If it is not possible to represent the inode number supplied by the filesystem in the field provided by userspace, then error EOVERFLOW will be issued. Similarly, the getdents/readdir functions now pass the full 64-bit inode number to userspace where possible, returning EOVERFLOW instead when a directory entry is encountered that can't be properly represented. Note that this means that some inodes will not be stat'able on a 32-bit system with old libraries where they were before - but it does mean that there will be no ambiguity over what a 32-bit inode number refers to. Note similarly that directory scans may be cut short with an error on a 32-bit system with old libraries where the scan would work before for the same reasons. It is judged unlikely that this situation will occur because modern glibc uses 64-bit capable versions of stat and getdents class functions exclusively, and that older systems are unlikely to encounter unrepresentable inode numbers anyway. [akpm: alpha build fix] Signed-off-by: David Howells <dhowells@redhat.com> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-03 16:13:46 +08:00
return -EOVERFLOW;
}
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
prev_reclen = buf->prev_reclen;
if (prev_reclen && signal_pending(current))
return -EINTR;
dirent = buf->current_dir;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
prev = (void __user *) dirent - prev_reclen;
if (!user_access_begin(prev, reclen + prev_reclen))
goto efault;
/* This might be 'dirent->d_off', but if so it will get overwritten */
unsafe_put_user(offset, &prev->d_off, efault_end);
unsafe_put_user(d_ino, &dirent->d_ino, efault_end);
unsafe_put_user(reclen, &dirent->d_reclen, efault_end);
unsafe_put_user(d_type, (char __user *) dirent + reclen - 1, efault_end);
unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
user_access_end();
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
buf->current_dir = (void __user *)dirent + reclen;
buf->prev_reclen = reclen;
buf->count -= reclen;
return 0;
efault_end:
user_access_end();
efault:
buf->error = -EFAULT;
return -EFAULT;
}
SYSCALL_DEFINE3(getdents, unsigned int, fd,
struct linux_dirent __user *, dirent, unsigned int, count)
{
struct fd f;
struct getdents_callback buf = {
.ctx.actor = filldir,
.count = count,
.current_dir = dirent
};
int error;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(dirent, count))
return -EFAULT;
f = fdget_pos(fd);
if (!f.file)
return -EBADF;
error = iterate_dir(f.file, &buf.ctx);
if (error >= 0)
error = buf.error;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
if (buf.prev_reclen) {
struct linux_dirent __user * lastdirent;
lastdirent = (void __user *)buf.current_dir - buf.prev_reclen;
if (put_user(buf.ctx.pos, &lastdirent->d_off))
error = -EFAULT;
else
error = count - buf.count;
}
fdput_pos(f);
return error;
}
struct getdents_callback64 {
struct dir_context ctx;
struct linux_dirent64 __user * current_dir;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
int prev_reclen;
int count;
int error;
};
static int filldir64(struct dir_context *ctx, const char *name, int namlen,
loff_t offset, u64 ino, unsigned int d_type)
{
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
struct linux_dirent64 __user *dirent, *prev;
struct getdents_callback64 *buf =
container_of(ctx, struct getdents_callback64, ctx);
int reclen = ALIGN(offsetof(struct linux_dirent64, d_name) + namlen + 1,
sizeof(u64));
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
int prev_reclen;
Make filldir[64]() verify the directory entry filename is valid This has been discussed several times, and now filesystem people are talking about doing it individually at the filesystem layer, so head that off at the pass and just do it in getdents{64}(). This is partially based on a patch by Jann Horn, but checks for NUL bytes as well, and somewhat simplified. There's also commentary about how it might be better if invalid names due to filesystem corruption don't cause an immediate failure, but only an error at the end of the readdir(), so that people can still see the filenames that are ok. There's also been discussion about just how much POSIX strictly speaking requires this since it's about filesystem corruption. It's really more "protect user space from bad behavior" as pointed out by Jann. But since Eric Biederman looked up the POSIX wording, here it is for context: "From readdir: The readdir() function shall return a pointer to a structure representing the directory entry at the current position in the directory stream specified by the argument dirp, and position the directory stream at the next entry. It shall return a null pointer upon reaching the end of the directory stream. The structure dirent defined in the <dirent.h> header describes a directory entry. From definitions: 3.129 Directory Entry (or Link) An object that associates a filename with a file. Several directory entries can associate names with the same file. ... 3.169 Filename A name consisting of 1 to {NAME_MAX} bytes used to name a file. The characters composing the name may be selected from the set of all character values excluding the slash character and the null byte. The filenames dot and dot-dot have special meaning. A filename is sometimes referred to as a 'pathname component'." Note that I didn't bother adding the checks to any legacy interfaces that nobody uses. Also note that if this ends up being noticeable as a performance regression, we can fix that to do a much more optimized model that checks for both NUL and '/' at the same time one word at a time. We haven't really tended to optimize 'memchr()', and it only checks for one pattern at a time anyway, and we really _should_ check for NUL too (but see the comment about "soft errors" in the code about why it currently only checks for '/') See the CONFIG_DCACHE_WORD_ACCESS case of hash_name() for how the name lookup code looks for pathname terminating characters in parallel. Link: https://lore.kernel.org/lkml/20190118161440.220134-2-jannh@google.com/ Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jann Horn <jannh@google.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-10-06 02:32:52 +08:00
buf->error = verify_dirent_name(name, namlen);
if (unlikely(buf->error))
return buf->error;
buf->error = -EINVAL; /* only used if we fail.. */
if (reclen > buf->count)
return -EINVAL;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
prev_reclen = buf->prev_reclen;
if (prev_reclen && signal_pending(current))
return -EINTR;
dirent = buf->current_dir;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
prev = (void __user *)dirent - prev_reclen;
if (!user_access_begin(prev, reclen + prev_reclen))
goto efault;
/* This might be 'dirent->d_off', but if so it will get overwritten */
unsafe_put_user(offset, &prev->d_off, efault_end);
unsafe_put_user(ino, &dirent->d_ino, efault_end);
unsafe_put_user(reclen, &dirent->d_reclen, efault_end);
unsafe_put_user(d_type, &dirent->d_type, efault_end);
unsafe_copy_dirent_name(dirent->d_name, name, namlen, efault_end);
user_access_end();
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
buf->prev_reclen = reclen;
buf->current_dir = (void __user *)dirent + reclen;
buf->count -= reclen;
return 0;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
efault_end:
user_access_end();
efault:
buf->error = -EFAULT;
return -EFAULT;
}
int ksys_getdents64(unsigned int fd, struct linux_dirent64 __user *dirent,
unsigned int count)
{
struct fd f;
struct getdents_callback64 buf = {
.ctx.actor = filldir64,
.count = count,
.current_dir = dirent
};
int error;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(dirent, count))
return -EFAULT;
f = fdget_pos(fd);
if (!f.file)
return -EBADF;
error = iterate_dir(f.file, &buf.ctx);
if (error >= 0)
error = buf.error;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
if (buf.prev_reclen) {
struct linux_dirent64 __user * lastdirent;
typeof(lastdirent->d_off) d_off = buf.ctx.pos;
readdir: make user_access_begin() use the real access range In commit 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") I changed filldir to not do individual __put_user() accesses, but instead use unsafe_put_user() surrounded by the proper user_access_begin/end() pair. That make them enormously faster on modern x86, where the STAC/CLAC games make individual user accesses fairly heavy-weight. However, the user_access_begin() range was not really the exact right one, since filldir() has the unfortunate problem that it needs to not only fill out the new directory entry, it also needs to fix up the previous one to contain the proper file offset. It's unfortunate, but the "d_off" field in "struct dirent" is _not_ the file offset of the directory entry itself - it's the offset of the next one. So we end up backfilling the offset in the previous entry as we walk along. But since x86 didn't really care about the exact range, and used to be the only architecture that did anything fancy in user_access_begin() to begin with, the filldir[64]() changes did something lazy, and even commented on it: /* * Note! This range-checks 'previous' (which may be NULL). * The real range was checked in getdents */ if (!user_access_begin(dirent, sizeof(*dirent))) goto efault; and it all worked fine. But now 32-bit ppc is starting to also implement user_access_begin(), and the fact that we faked the range to only be the (possibly not even valid) previous directory entry becomes a problem, because ppc32 will actually be using the range that is passed in for more than just "check that it's user space". This is a complete rewrite of Christophe's original patch. By saving off the record length of the previous entry instead of a pointer to it in the filldir data structures, we can simplify the range check and the writing of the previous entry d_off field. No need for any conditionals in the user accesses themselves, although we retain the conditional EINTR checking for the "was this the first directory entry" signal handling latency logic. Fixes: 9f79b78ef744 ("Convert filldir[64]() from __put_user() to unsafe_put_user()") Link: https://lore.kernel.org/lkml/a02d3426f93f7eb04960a4d9140902d278cab0bb.1579697910.git.christophe.leroy@c-s.fr/ Link: https://lore.kernel.org/lkml/408c90c4068b00ea8f1c41cca45b84ec23d4946b.1579783936.git.christophe.leroy@c-s.fr/ Reported-and-tested-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-01-23 04:37:25 +08:00
lastdirent = (void __user *) buf.current_dir - buf.prev_reclen;
if (__put_user(d_off, &lastdirent->d_off))
error = -EFAULT;
else
error = count - buf.count;
}
fdput_pos(f);
return error;
}
SYSCALL_DEFINE3(getdents64, unsigned int, fd,
struct linux_dirent64 __user *, dirent, unsigned int, count)
{
return ksys_getdents64(fd, dirent, count);
}
#ifdef CONFIG_COMPAT
struct compat_old_linux_dirent {
compat_ulong_t d_ino;
compat_ulong_t d_offset;
unsigned short d_namlen;
char d_name[1];
};
struct compat_readdir_callback {
struct dir_context ctx;
struct compat_old_linux_dirent __user *dirent;
int result;
};
static int compat_fillonedir(struct dir_context *ctx, const char *name,
int namlen, loff_t offset, u64 ino,
unsigned int d_type)
{
struct compat_readdir_callback *buf =
container_of(ctx, struct compat_readdir_callback, ctx);
struct compat_old_linux_dirent __user *dirent;
compat_ulong_t d_ino;
if (buf->result)
return -EINVAL;
d_ino = ino;
if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) {
buf->result = -EOVERFLOW;
return -EOVERFLOW;
}
buf->result++;
dirent = buf->dirent;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(dirent,
(unsigned long)(dirent->d_name + namlen + 1) -
(unsigned long)dirent))
goto efault;
if ( __put_user(d_ino, &dirent->d_ino) ||
__put_user(offset, &dirent->d_offset) ||
__put_user(namlen, &dirent->d_namlen) ||
__copy_to_user(dirent->d_name, name, namlen) ||
__put_user(0, dirent->d_name + namlen))
goto efault;
return 0;
efault:
buf->result = -EFAULT;
return -EFAULT;
}
COMPAT_SYSCALL_DEFINE3(old_readdir, unsigned int, fd,
struct compat_old_linux_dirent __user *, dirent, unsigned int, count)
{
int error;
struct fd f = fdget_pos(fd);
struct compat_readdir_callback buf = {
.ctx.actor = compat_fillonedir,
.dirent = dirent
};
if (!f.file)
return -EBADF;
error = iterate_dir(f.file, &buf.ctx);
if (buf.result)
error = buf.result;
fdput_pos(f);
return error;
}
struct compat_linux_dirent {
compat_ulong_t d_ino;
compat_ulong_t d_off;
unsigned short d_reclen;
char d_name[1];
};
struct compat_getdents_callback {
struct dir_context ctx;
struct compat_linux_dirent __user *current_dir;
struct compat_linux_dirent __user *previous;
int count;
int error;
};
static int compat_filldir(struct dir_context *ctx, const char *name, int namlen,
loff_t offset, u64 ino, unsigned int d_type)
{
struct compat_linux_dirent __user * dirent;
struct compat_getdents_callback *buf =
container_of(ctx, struct compat_getdents_callback, ctx);
compat_ulong_t d_ino;
int reclen = ALIGN(offsetof(struct compat_linux_dirent, d_name) +
namlen + 2, sizeof(compat_long_t));
buf->error = -EINVAL; /* only used if we fail.. */
if (reclen > buf->count)
return -EINVAL;
d_ino = ino;
if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) {
buf->error = -EOVERFLOW;
return -EOVERFLOW;
}
dirent = buf->previous;
if (dirent) {
if (signal_pending(current))
return -EINTR;
if (__put_user(offset, &dirent->d_off))
goto efault;
}
dirent = buf->current_dir;
if (__put_user(d_ino, &dirent->d_ino))
goto efault;
if (__put_user(reclen, &dirent->d_reclen))
goto efault;
if (copy_to_user(dirent->d_name, name, namlen))
goto efault;
if (__put_user(0, dirent->d_name + namlen))
goto efault;
if (__put_user(d_type, (char __user *) dirent + reclen - 1))
goto efault;
buf->previous = dirent;
dirent = (void __user *)dirent + reclen;
buf->current_dir = dirent;
buf->count -= reclen;
return 0;
efault:
buf->error = -EFAULT;
return -EFAULT;
}
COMPAT_SYSCALL_DEFINE3(getdents, unsigned int, fd,
struct compat_linux_dirent __user *, dirent, unsigned int, count)
{
struct fd f;
struct compat_linux_dirent __user * lastdirent;
struct compat_getdents_callback buf = {
.ctx.actor = compat_filldir,
.current_dir = dirent,
.count = count
};
int error;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(dirent, count))
return -EFAULT;
f = fdget_pos(fd);
if (!f.file)
return -EBADF;
error = iterate_dir(f.file, &buf.ctx);
if (error >= 0)
error = buf.error;
lastdirent = buf.previous;
if (lastdirent) {
if (put_user(buf.ctx.pos, &lastdirent->d_off))
error = -EFAULT;
else
error = count - buf.count;
}
fdput_pos(f);
return error;
}
#endif