The task files should consult the "spawn action" and "O_CLOEXEC flags"
to determine further whether the file should be duplicated.
This PR will further optimize file list duplicating to avoid the performance
regression caused by additional file operations.
Signed-off-by: chao an <anchao@xiaomi.com>
VELAPLATFO-18473
refs:
https://man7.org/linux/man-pages/man2/fcntl.2.html
If the FD_CLOEXEC bit is set, the file descriptor will automatically
be closed during a successful execve(2).
(If the execve(2) fails, the file descriptor is left open.)
modify:
1. Ensure that the child task copies all fds of the parent task,
including those with O_CLOEXE.
2. Make sure spawn_file_action is executed under fd with O_CLOEXEC,
otherwise it will fail.
3. When a new task is activated or exec is called, close all fds
with O_CLOEXEC flags.
Signed-off-by: dongjiuzhu1 <dongjiuzhu1@xiaomi.com>
In embedded development environments, due to the lack of address isolation between processes,
fd may be passed between processes and lead to misuse,
We have designed an fd cross-process automatic detection tool,
fdcheck_protect returns the fd containing the pid information,
indicating that the ownership of the current fd belongs to the pid and is not allowed to be used by other processes.
fdcheck_restore will obtain the true fd and check if the ownership of the fd is legal
For ease of understanding, let's give an example where
the following information is represented in 32-bit binary format
fd 00000000 00000000 00000000 10001010
pid 00000000 00000000 00000011 01010101
ret 00000000 00000011 01010101 10001010
Signed-off-by: hujun5 <hujun5@xiaomi.com>
When the file pointer is reopened, keep f_tag unchanged as fd remains unchanged
fix issue https://github.com/apache/nuttx/issues/9494
Signed-off-by: hujun5 <hujun5@xiaomi.com>
FD (file descriptor) is widely used in system software development,
and almost all implementations of posix os (including nuttx) use FD as an index.
the value of fd needs to be allocated starting from the minimum available value of 3, and each process has a copy,
so the same fd value is very easy to reuse in the program.
In multi threaded or multi process environments without address isolation,
If the ownership, global variables, and competition relationships of fd are not properly handled,
there may be issues with fd duplication or accidental closure.
Further leading to the following issues, which are difficult to troubleshoot.
1. Security vulnerability: the fd we wrote is not the expected fd and will be accessed by hackers to obtain data
2. Program exceptions or crashes: write or read fd failures, and program logic errors
3. The structured file XML or database is damaged: the data format written to the database is not the expected format.
The implementation principle of fdsan is based on the implementation of Android
https://android.googlesource.com/platform/bionic/+/master/docs/fdsan.md
Signed-off-by: hujun5 <hujun5@xiaomi.com>
The spawn proxy thread is a special existence in NuttX, usually some developers
spend a lot of time on stack overflow of spawn proxy thread:
https://github.com/apache/nuttx/issues/9046https://github.com/apache/nuttx/pull/9081
In order to avoid similar issues, this PR will remove spawn proxy thread to simplify
the process of task/posix_spawn().
1. Postpone the related processing of spawn file actions until after task_init()
2. Delete the temporary thread of spawn proxy and related global variables
Signed-off-by: chao an <anchao@xiaomi.com>
apps/examples/usrsocktest/usrsocktest_basic_daemon.c:
321 static void basic_daemon_dup2(FAR struct usrsocktest_daemon_conf_s *dconf)
322 {
...
335 ret = dup2(sd2, sd);
352 }
Usrsocktest Task hold the file group lock and send the close request to usrsock deamon :
| #0 net_lockedwait_uninterruptible (sem=0x5555555f8ba2 <g_usrsockdev+34>) at utils/net_lock.c:427
| #1 0x000055555557489c in usrsockdev_do_request (conn=0x5555555f8800 <g_usrsock_connections>, iov=0x7ffff3f36040, iovcnt=1) at usrsock/usrsock_dev.c:1185
| --> send close request to usrsock deamon
|
| #2 0x00005555555d0439 in do_close_request (conn=0x5555555f8800 <g_usrsock_connections>) at usrsock/usrsock_close.c:109
| #3 0x00005555555d04f5 in usrsock_close (conn=0x5555555f8800 <g_usrsock_connections>) at usrsock/usrsock_close.c:157
| #4 0x00005555555cf100 in usrsock_sockif_close (psock=0x7ffff3ea4a60) at usrsock/usrsock_sockif.c:234
| #5 0x00005555555c7b2f in psock_close (psock=0x7ffff3ea4a60) at socket/net_close.c:102
| #6 0x000055555557a518 in sock_file_close (filep=0x7ffff3f253d0) at socket/socket.c:115
| #7 0x000055555557678f in file_close (filep=0x7ffff3f253d0) at vfs/fs_close.c:74
| #8 0x000055555557694c in file_dup2 (filep1=0x7ffff3f253e8, filep2=0x7ffff3f253d0) at vfs/fs_dup2.c:129
| ---> hold group file list lock ( _files_semtake(list) )
|
| #9 0x0000555555575aab in nx_dup2 (fd1=7, fd2=6) at inode/fs_files.c:451
| #10 0x0000555555575af3 in dup2 (fd1=7, fd2=6) at inode/fs_files.c:473
| #11 0x000055555559d937 in basic_daemon_dup2 (dconf=0x5555555f8d80 <usrsocktest_daemon_config>) at usrsocktest_basic_daemon.c:335
| #12 0x000055555559ed80 in usrsocktest_test_basic_daemon_basic_daemon_dup2 () at usrsocktest_basic_daemon.c:612
| #13 0x000055555559f18d in usrsocktest_group_basic_daemon_run () at usrsocktest_basic_daemon.c:666
| #14 0x0000555555599f8d in run_tests (name=0x5555555dc8c3 "basic_daemon", test_fn=0x55555559ef50 <usrsocktest_group_basic_daemon_run>) at usrsocktest_main.c:117
| #15 0x000055555559a06c in run_all_tests () at usrsocktest_main.c:154
| #16 0x000055555559a3d1 in usrsocktest_main (argc=1, argv=0x7ffff3f25450) at usrsocktest_main.c:248
| #17 0x000055555555cad8 in nxtask_startup (entrypt=0x55555559a357 <usrsocktest_main>, argc=1, argv=0x7ffff3f25450) at sched/task_startup.c:70
| #18 0x0000555555559938 in nxtask_start () at task/task_start.c:134
Usrsock Deamon weakup and setup the poll want to perform close request, but locked on fs_getfilep():
| #0 _files_semtake (list=0x7ffff3f250b8) at inode/fs_files.c:51
| --> Request group lock but which hold by close request, deadlock
| #1 0x00005555555758b1 in fs_getfilep (fd=5, filep=0x7ffff3f47190) at inode/fs_files.c:375
| #2 0x00005555555d3064 in poll_fdsetup (fd=5, fds=0x7ffff3f47290, setup=true) at vfs/fs_poll.c:79
| #3 0x00005555555d3243 in poll_setup (fds=0x7ffff3f47290, nfds=2, sem=0x7ffff3f47206) at vfs/fs_poll.c:139
| #4 0x00005555555d39a6 in nx_poll (fds=0x7ffff3f47290, nfds=2, timeout=-1) at vfs/fs_poll.c:383
| #5 0x00005555555d3abd in poll (fds=0x7ffff3f47290, nfds=2, timeout=-1) at vfs/fs_poll.c:501
| --> daemon weak up
| #6 0x00005555555c62c7 in usrsocktest_daemon (param=0x5555555f5360 <g_ub_daemon>) at usrsocktest_daemon.c:1846
| #7 0x000055555559161e in pthread_startup (entry=0x5555555c60d3 <usrsocktest_daemon>, arg=0x5555555f5360 <g_ub_daemon>) at pthread/pthread_create.c:59
| #8 0x00005555555d45f0 in pthread_start () at pthread/pthread_create.c:175
| #9 0x0000000000000000 in ?? ()
Signed-off-by: chao an <anchao@xiaomi.com>
since the temporary kernel spawn proxy need duplicate file handler
from caller, please reference the follow source code for more info:
sched/task/task_posixspawn.c
sched/task/task_spawn.c
This reverts commit 5c5f1dec08.
If the close is blocked(like net close timeout 5s),
then other file operation can't access in this 5s.
Change-Id: Ia6c4ec88d90db0330134d0aaa30e94cb71c8a066
Signed-off-by: ligd <liguiding1@xiaomi.com>
let's call either nx_dup/nx_dup2 or file_dup/file_dup2
instead just like other fs api: xxx->nx_xxx->file_xxx
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I1aacfb9e25dc7b3fcb0345ff7b269b1953a01e5b
the argument passed to file_dup2 doesn't always come from task file list
so it doesn't make sense to hold the file list lock and then it is better
to do the protection in the new function files_dupfd2
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ibf02cea9b0b275e7472f9c04fd66b9242285b957
NULL inode passed to files_allocate doesn't mark file struct in the
allocated state, so other threads which invovle in file allocation
(e.g. open or dup) may allocate the same file struct again.
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I53ff876eae3c7a1e311e7f671686b73a4b4ef891
since the internal function normally return the error code directly
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ifccbcfb07d2f8f7d5fef1b2a867538f51b73655c
Resolution of Issue 619 will require multiple steps, this part of the first step in that resolution: Every call to nxsem_wait_uninterruptible() must handle the return value from nxsem_wait_uninterruptible properly. This commit is only for those files under fs/inode. Utility functions under fs/incode were modified so the changes do extend to other fs/ sub-directories as well.
* Simplify EINTR/ECANCEL error handling
1. Add semaphore uninterruptible wait function
2 .Replace semaphore wait loop with a single uninterruptible wait
3. Replace all sem_xxx to nxsem_xxx
* Unify the void cast usage
1. Remove void cast for function because many place ignore the returned value witout cast
2. Replace void cast for variable with UNUSED macro
This commit backs out most of commit b4747286b1. That change was added because sem_wait() would sometimes cause cancellation points inappropriated. But with these recent changes, nxsem_wait() is used instead and it is not a cancellation point.
In the OS, all calls to sem_wait() changed to nxsem_wait(). nxsem_wait() does not return errors via errno so each place where nxsem_wait() is now called must not examine the errno variable.
In all OS functions (not libraries), change sem_wait() to nxsem_wait(). This will prevent the OS from creating bogus cancellation points and from modifying the per-task errno variable.
sched/semaphore: Add the function nxsem_wait(). This is a new internal OS interface. It is functionally equivalent to sem_wait() except that (1) it is not a cancellation point, and (2) it does not set the per-thread errno value on return.
