linux/fs/ntfs3/namei.c

540 lines
11 KiB

fs/ntfs3: Use kernel ALIGN macros over driver specific The static checkers (Smatch) were complaining because QuadAlign() was buggy. If you try to align something higher than UINT_MAX it got truncated to a u32. Smatch warning was: fs/ntfs3/attrib.c:383 attr_set_size_res() warn: was expecting a 64 bit value instead of '~7' So that this will not happen again we will change all these macros to kernel made ones. This can also help some other static analyzing tools to give us better warnings. Patch was generated with Coccinelle script and after that some style issue was hand fixed. Coccinelle script: virtual patch @alloc depends on patch@ expression x; @@ ( - #define QuadAlign(n) (((n) + 7u) & (~7u)) | - QuadAlign(x) + ALIGN(x, 8) | - #define IsQuadAligned(n) (!((size_t)(n)&7u)) | - IsQuadAligned(x) + IS_ALIGNED(x, 8) | - #define Quad2Align(n) (((n) + 15u) & (~15u)) | - Quad2Align(x) + ALIGN(x, 16) | - #define IsQuad2Aligned(n) (!((size_t)(n)&15u)) | - IsQuad2Aligned(x) + IS_ALIGNED(x, 16) | - #define Quad4Align(n) (((n) + 31u) & (~31u)) | - Quad4Align(x) + ALIGN(x, 32) | - #define IsSizeTAligned(n) (!((size_t)(n) & (sizeof(size_t) - 1))) | - IsSizeTAligned(x) + IS_ALIGNED(x, sizeof(size_t)) | - #define DwordAlign(n) (((n) + 3u) & (~3u)) | - DwordAlign(x) + ALIGN(x, 4) | - #define IsDwordAligned(n) (!((size_t)(n)&3u)) | - IsDwordAligned(x) + IS_ALIGNED(x, 4) | - #define WordAlign(n) (((n) + 1u) & (~1u)) | - WordAlign(x) + ALIGN(x, 2) | - #define IsWordAligned(n) (!((size_t)(n)&1u)) | - IsWordAligned(x) + IS_ALIGNED(x, 2) | ) Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Kari Argillander <kari.argillander@gmail.com> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
2021-08-26 11:56:29 +03:00
fs/ntfs3: Fix null-ptr-deref on inode->i_op in ntfs_lookup() Syzbot reported a null-ptr-deref bug: ntfs3: loop0: Different NTFS' sector size (1024) and media sector size (512) ntfs3: loop0: Mark volume as dirty due to NTFS errors general protection fault, probably for non-canonical address 0xdffffc0000000001: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] RIP: 0010:d_flags_for_inode fs/dcache.c:1980 [inline] RIP: 0010:__d_add+0x5ce/0x800 fs/dcache.c:2796 Call Trace: <TASK> d_splice_alias+0x122/0x3b0 fs/dcache.c:3191 lookup_open fs/namei.c:3391 [inline] open_last_lookups fs/namei.c:3481 [inline] path_openat+0x10e6/0x2df0 fs/namei.c:3688 do_filp_open+0x264/0x4f0 fs/namei.c:3718 do_sys_openat2+0x124/0x4e0 fs/open.c:1310 do_sys_open fs/open.c:1326 [inline] __do_sys_open fs/open.c:1334 [inline] __se_sys_open fs/open.c:1330 [inline] __x64_sys_open+0x221/0x270 fs/open.c:1330 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd If the MFT record of ntfs inode is not a base record, inode->i_op can be NULL. And a null-ptr-deref may happen: ntfs_lookup() dir_search_u() # inode->i_op is set to NULL d_splice_alias() __d_add() d_flags_for_inode() # inode->i_op->get_link null-ptr-deref Fix this by adding a Check on inode->i_op before calling the d_splice_alias() function. Fixes: 4342306f0f0d ("fs/ntfs3: Add file operations and implementation") Reported-by: syzbot+a8f26a403c169b7593fe@syzkaller.appspotmail.com Signed-off-by: ZhangPeng <zhangpeng362@huawei.com> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
2022-11-25 10:21:59 +00:00
fs/ntfs3: Fix null-ptr-deref on inode->i_op in ntfs_lookup() Syzbot reported a null-ptr-deref bug: ntfs3: loop0: Different NTFS' sector size (1024) and media sector size (512) ntfs3: loop0: Mark volume as dirty due to NTFS errors general protection fault, probably for non-canonical address 0xdffffc0000000001: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] RIP: 0010:d_flags_for_inode fs/dcache.c:1980 [inline] RIP: 0010:__d_add+0x5ce/0x800 fs/dcache.c:2796 Call Trace: <TASK> d_splice_alias+0x122/0x3b0 fs/dcache.c:3191 lookup_open fs/namei.c:3391 [inline] open_last_lookups fs/namei.c:3481 [inline] path_openat+0x10e6/0x2df0 fs/namei.c:3688 do_filp_open+0x264/0x4f0 fs/namei.c:3718 do_sys_openat2+0x124/0x4e0 fs/open.c:1310 do_sys_open fs/open.c:1326 [inline] __do_sys_open fs/open.c:1334 [inline] __se_sys_open fs/open.c:1330 [inline] __x64_sys_open+0x221/0x270 fs/open.c:1330 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd If the MFT record of ntfs inode is not a base record, inode->i_op can be NULL. And a null-ptr-deref may happen: ntfs_lookup() dir_search_u() # inode->i_op is set to NULL d_splice_alias() __d_add() d_flags_for_inode() # inode->i_op->get_link null-ptr-deref Fix this by adding a Check on inode->i_op before calling the d_splice_alias() function. Fixes: 4342306f0f0d ("fs/ntfs3: Add file operations and implementation") Reported-by: syzbot+a8f26a403c169b7593fe@syzkaller.appspotmail.com Signed-off-by: ZhangPeng <zhangpeng362@huawei.com> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
2022-11-25 10:21:59 +00:00
Change inode_operations.mkdir to return struct dentry * Some filesystems, such as NFS, cifs, ceph, and fuse, do not have complete control of sequencing on the actual filesystem (e.g. on a different server) and may find that the inode created for a mkdir request already exists in the icache and dcache by the time the mkdir request returns. For example, if the filesystem is mounted twice the directory could be visible on the other mount before it is on the original mount, and a pair of name_to_handle_at(), open_by_handle_at() calls could instantiate the directory inode with an IS_ROOT() dentry before the first mkdir returns. This means that the dentry passed to ->mkdir() may not be the one that is associated with the inode after the ->mkdir() completes. Some callers need to interact with the inode after the ->mkdir completes and they currently need to perform a lookup in the (rare) case that the dentry is no longer hashed. This lookup-after-mkdir requires that the directory remains locked to avoid races. Planned future patches to lock the dentry rather than the directory will mean that this lookup cannot be performed atomically with the mkdir. To remove this barrier, this patch changes ->mkdir to return the resulting dentry if it is different from the one passed in. Possible returns are: NULL - the directory was created and no other dentry was used ERR_PTR() - an error occurred non-NULL - this other dentry was spliced in This patch only changes file-systems to return "ERR_PTR(err)" instead of "err" or equivalent transformations. Subsequent patches will make further changes to some file-systems to return a correct dentry. Not all filesystems reliably result in a positive hashed dentry: - NFS, cifs, hostfs will sometimes need to perform a lookup of the name to get inode information. Races could result in this returning something different. Note that this lookup is non-atomic which is what we are trying to avoid. Placing the lookup in filesystem code means it only happens when the filesystem has no other option. - kernfs and tracefs leave the dentry negative and the ->revalidate operation ensures that lookup will be called to correctly populate the dentry. This could be fixed but I don't think it is important to any of the users of vfs_mkdir() which look at the dentry. The recommendation to use d_drop();d_splice_alias() is ugly but fits with current practice. A planned future patch will change this. Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: NeilBrown <neilb@suse.de> Link: https://lore.kernel.org/r/20250227013949.536172-2-neilb@suse.de Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-02-27 12:32:53 +11:00
fs/ntfs3: Don't use uni1 uninitialized in ntfs_d_compare() Clang warns: fs/ntfs3/namei.c:445:7: error: variable 'uni1' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] if (toupper(c1) != toupper(c2)) { ^~~~~~~~~~~~~~~~~~~~~~~~~~ ./include/linux/ctype.h:64:20: note: expanded from macro 'toupper' #define toupper(c) __toupper(c) ^ fs/ntfs3/namei.c:487:12: note: uninitialized use occurs here __putname(uni1); ^~~~ ./include/linux/fs.h:2789:65: note: expanded from macro '__putname' #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) ^~~~ fs/ntfs3/namei.c:445:3: note: remove the 'if' if its condition is always false if (toupper(c1) != toupper(c2)) { ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ fs/ntfs3/namei.c:434:7: error: variable 'uni1' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] if (!lm--) { ^~~~~ fs/ntfs3/namei.c:487:12: note: uninitialized use occurs here __putname(uni1); ^~~~ ./include/linux/fs.h:2789:65: note: expanded from macro '__putname' #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) ^~~~ fs/ntfs3/namei.c:434:3: note: remove the 'if' if its condition is always false if (!lm--) { ^~~~~~~~~~~~ fs/ntfs3/namei.c:430:22: note: initialize the variable 'uni1' to silence this warning struct cpu_str *uni1, *uni2; ^ = NULL 2 errors generated. There is no point in calling __putname() in these particular error paths, as there has been no corresponding __getname() call yet. Just return directly in these blocks to clear up the warning. Fixes: a3a956c78efa ("fs/ntfs3: Add option "nocase"") Link: https://github.com/ClangBuiltLinux/linux/issues/1729 Signed-off-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
2022-10-04 07:41:45 -07:00
fs/ntfs3: Don't use uni1 uninitialized in ntfs_d_compare() Clang warns: fs/ntfs3/namei.c:445:7: error: variable 'uni1' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] if (toupper(c1) != toupper(c2)) { ^~~~~~~~~~~~~~~~~~~~~~~~~~ ./include/linux/ctype.h:64:20: note: expanded from macro 'toupper' #define toupper(c) __toupper(c) ^ fs/ntfs3/namei.c:487:12: note: uninitialized use occurs here __putname(uni1); ^~~~ ./include/linux/fs.h:2789:65: note: expanded from macro '__putname' #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) ^~~~ fs/ntfs3/namei.c:445:3: note: remove the 'if' if its condition is always false if (toupper(c1) != toupper(c2)) { ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ fs/ntfs3/namei.c:434:7: error: variable 'uni1' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] if (!lm--) { ^~~~~ fs/ntfs3/namei.c:487:12: note: uninitialized use occurs here __putname(uni1); ^~~~ ./include/linux/fs.h:2789:65: note: expanded from macro '__putname' #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) ^~~~ fs/ntfs3/namei.c:434:3: note: remove the 'if' if its condition is always false if (!lm--) { ^~~~~~~~~~~~ fs/ntfs3/namei.c:430:22: note: initialize the variable 'uni1' to silence this warning struct cpu_str *uni1, *uni2; ^ = NULL 2 errors generated. There is no point in calling __putname() in these particular error paths, as there has been no corresponding __getname() call yet. Just return directly in these blocks to clear up the warning. Fixes: a3a956c78efa ("fs/ntfs3: Add option "nocase"") Link: https://github.com/ClangBuiltLinux/linux/issues/1729 Signed-off-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
2022-10-04 07:41:45 -07:00