Commit Graph

434 Commits

Author SHA1 Message Date
wangbowen6
e9ccab2db3 fs/poll: using callback mechanism to implement poll notification
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-10-03 00:00:56 +08:00
Xiang Xiao
bdeaea3742 Remove the unnessary empty line after label
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-09-30 17:54:56 +02:00
wangbowen6
5e46a9908a vfs/fs_poll: not clear POLLIN event if POLLHUP or POLLERR set
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-09-27 08:24:33 +09:00
wangbowen6
344c8be049 poll: add poll_notify() api and call it in all drivers
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-09-26 12:06:32 +08:00
Xiang Xiao
db518bf0df fs: Allocate unique serial number for the root pseduo file system node
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-09-20 16:21:06 +08:00
Xiang Xiao
493152a45f fs/eventfd: Remove the extra space
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-09-20 16:20:17 +09:00
Nathan Hartman
09a186727c fs/vfs/fs_open.c: Improve documentation blocks of functions 2022-09-08 09:01:36 +08:00
Xiang Xiao
9726be616a fs: Run the default action of FIONBIO/FIOCLEX/FIONCLEX in success path
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-09-06 23:16:15 +08:00
Jiuzhu Dong
4dbf4555eb fs/dir: support fchdir and dirfd
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-08-09 23:30:01 +08:00
Jiuzhu Dong
fe17f747a7 fs/directory: move private directory information to filesystem
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-08-09 23:30:01 +08:00
Jiuzhu Dong
90db4daca9 fs/directory: update readdir interface for all filesystem
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-08-09 23:30:01 +08:00
Jiuzhu Dong
3a70962b7a fs/directory: use file mode to manage directory
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-08-09 23:30:01 +08:00
ligd
5f68aa56fd poll: defalut set POLLERR POLLHUP to events
Signed-off-by: ligd <liguiding1@xiaomi.com>
2022-07-26 09:36:55 +08:00
Jiuzhu Dong
978530f5f6 fs/ioctl: add BLKSSZGET cmd to get block sector size
refer to:
https://sites.uclouvain.be/SystInfo/usr/include/sys/mount.h.html

Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-07-22 09:07:57 +03:00
Jiuzhu Dong
b385c48a30 fs/fs_fsync: add file sync operation by ioctl
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-07-22 09:07:57 +03:00
ligd
cda9cfbc80 unlink: don't do unlink with inode_semtake
Signed-off-by: ligd <liguiding1@xiaomi.com>
2022-07-22 11:00:26 +08:00
Nathan Hartman
849f760b77 Fix various typos 2022-07-08 02:15:54 +08:00
Xiang Xiao
22e4f1c59a sched: Remove start from nxsem_tickwait[_uninterruptible]
to simplify both caller and callee

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-05-15 13:55:58 +03:00
Xiang Xiao
f14eed6432 vfs/fcntl: Minor style fix
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-04-14 17:21:25 +03:00
liuhaitao
05cc6fb610 fs/fcntl: add O_APPEND flag judge in fcntl
Signed-off-by: liuhaitao <liuhaitao@xiaomi.com>
2022-04-14 17:21:25 +03:00
Jukka Laitinen
dbc163f1b0 fs/vfs/fs_truncate.c: Use ioctl to truncate on non-mountpoint inode
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
2022-04-13 09:40:18 +08:00
yinshengkai
4b6743591a fs/vfs: fix st_mode mask check
The full mask for st_mode is 0177777
Now modify any file permissions in hostfs and all will fail
2022-04-08 15:06:41 +03:00
Jiuzhu Dong
b03f2e34b8 lseek: use type:off_t for return value
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-03-28 22:49:37 +08:00
Petro Karashchenko
98ba65c422 c89: get rid of designated initializers in common code
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2022-03-28 13:39:27 +08:00
ligd
0d365f6fb9 epoll: fix epoll close error, report by kasan
-#9  0xf7abf899 in __asan::__asan_report_load2 (addr=4072681776) at ../../../../../src/libsanitizer/asan/asan_rtl.cc:117
-#10 0x5693f718 in inode_release (node=0xf2c03124) at inode/fs_inoderelease.c:69
-#11 0x568ea61b in file_close (filep=0xf55fedd0) at vfs/fs_close.c:79
-#12 0x568e7e56 in nx_close (fd=3) at inode/fs_files.c:528
-#13 0x568e7f0e in close (fd=3) at inode/fs_files.c:562
-#14 0x56e76c39 in epoll_close (epfd=3) at vfs/fs_epoll.c:252
-#15 0x56c33829 in sensor_service_delete (ctrl=0x578b8540 <control>) at src/common.c:439
-#16 0x56a0561e in sensor_middle_service_main (argc=1, argv=0xf55de820) at sensor_main.c:118
-#17 0x56878675 in nxtask_startup (entrypt=0x56a054cc <sensor_middle_service_main>, argc=1, argv=0xf55de820) at sched/task_startup.c:70
-#18 0x5684427a in nxtask_start () at task/task_start.c:133
-#19 0xdeadbeef in ?? ()

reason:
epoll_close -> close -> epoll_do_close (free inode)
                     -> inode_release  (reuse inode, crash)

fix:
use the global inode to match the fd which will return to user.
like the g_sock_inode in fs/socket/socket.c

Signed-off-by: ligd <liguiding1@xiaomi.com>
2022-03-27 08:51:21 +03:00
Xiang Xiao
b2c86f808d net: Remove psock_fcntl related code
since the nonblocking mode set through psock_ioctl now

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-03-04 22:03:55 +02:00
Xiang Xiao
19e796305a vfs: Forward fcntl(F_SETFD...) to ioctl(FIOCLEX|FIONCLEX...)
this is follow change to:
commit 37730a1fce
Author: fangzhenwei <fangzhenwei@xiaomi.com>
Date:   Thu Nov 25 13:43:10 2021 +0800

    nuttx/fcntl:pass O_NONBLOCK flag to ioctl

    1. fix pty fcntl F_SETFL(O_NONBLOCK) fail issue

    Signed-off-by: fangzhenwei <fangzhenwei@xiaomi.com>

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-03-04 22:03:55 +02:00
Xiang Xiao
43f57240e0 Replece clock_gettime(CLOCK_MONOTONIC) with clock_systime_timespec
it's better to call the kernrel api insteaad user space api in kernel

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-02-24 01:06:36 +08:00
Xiang Xiao
f1ed349dd9 sched/clock: Remove CLOCK_MONOTONIC option from Kconfig
here is the reason:
1.clock_systime_timespec(core function) always exist regardless the setting
2.CLOCK_MONOTONIC is a foundamental clock type required by many places

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-02-23 01:21:26 +08:00
fangzhenwei
37730a1fce nuttx/fcntl:pass O_NONBLOCK flag to ioctl
1. fix pty fcntl F_SETFL(O_NONBLOCK) fail issue

Signed-off-by: fangzhenwei <fangzhenwei@xiaomi.com>
2022-02-22 13:42:44 +08:00
Jiuzhu Dong
02ba5f4103 fs/epoll_fd: support dup file descriptor
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-11 13:52:35 +08:00
Xiang Xiao
0eec2f5251 fs: Remove the unnecessary Kconfig definition
since all these options should always have the default value

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-02-05 23:58:01 +01:00
YAMAMOTO Takashi
39ab0f4836 Do not require write-access for fsync
* fsync doesn't modify the file. It doesn't make sense to require
  write-access.

* This matches what ~all other systems do.
  At least Linux, macOS, and NetBSD.
2022-01-14 09:01:08 +01:00
Petro Karashchenko
a743fed63d file_operations: get back C89 compatible initializer
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2022-01-11 02:14:00 +08:00
zhouliang3
1f53a058fa nuttx:Change fs strncpy to strlcpy to avoid losing'\0'
Signed-off-by: zhouliang3 <zhouliang3@xiaomi.com>
2021-12-30 18:06:00 +08:00
Petro Karashchenko
d445282566 fs/vfs: Add file descriptor based timers support
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2021-12-20 04:00:20 -06:00
Xiang Xiao
0896d0680b eventfd: Change mode_semaphore from uint8_t to bool
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-12-19 01:41:11 -06:00
Xiang Xiao
76a4ca01a3 eventfd: Change minor type from size_t to unsigned int
to avoid the potential overflow devpath buffer on 64bit platform

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-12-19 01:41:11 -06:00
Xiang Xiao
8cb24a2964 eventfd: Fix the typo error
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-12-19 01:41:11 -06:00
Xiang Xiao
f9c1117c88 eventfd: Remove the unused and private eventfd_get_minor
since it isn't defined by Linux kernel too

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-12-19 01:41:11 -06:00
Jiuzhu Dong
e5f8acab97 ftruncate: sanity check for parameter:length
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-11-01 11:08:03 +01:00
chao.an
6dde42d79b vfs/poll: correct the return event if inode has closed
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-09-15 12:02:53 +08:00
Alin Jerpelea
151f6eab3c author: Anton D. Kachalov: update licenses to Apache
Anton D. Kachalov has submitted the ICLA
as a result we can migrate the licenses to Apache.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-09-01 21:02:34 +08:00
chao.an
9701a678bd net/tcp: add nonblock connect(2) support
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-08-19 19:19:05 -07:00
Xiang Xiao
6c61d032db fs: Add utimens and lutimens
https://pubs.opengroup.org/onlinepubs/9699919799/functions/futimens.html
https://www.daemon-systems.org/man/utimens.2.html

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-08-09 17:55:44 -03:00
Xiang Xiao
307cc61893 fs: Add fchstat and chstat callback into mountpt_operations
and implement all status related change function. the individual
file system change will provide in other upcoming patchset.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I8fde9db8eba9586e9c8da078b67e020c26623cf4
2021-07-29 06:33:49 -03:00
Jiuzhu Dong
11faf0cb20 fs/inode: add sanity check for inode to avoid nullpointer
Change-Id: Ib2c74ba308b8f15756fac4e69632c296243eb4ab
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-07-25 23:01:37 -07:00
Xiang Xiao
02a6966ad5 fs: mkdir return EEXIST if the path already exist
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ie1808697b860974367c17e0ba614ca5cae79f324
2021-07-18 10:25:54 -03:00
Xiang Xiao
663104a2e9 fs: Simplify sendfile implementation
and avoid call nx_ file API in the kernel space

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Id951ca161df2c2ee267dc4b5a0d7dfa67df4c1e6
2021-07-17 07:39:15 -03:00
Xiang Xiao
2e0901fcaa net: Add file_socket function
and move the socket special process from fstat/nx_vfcntl/ to file_fstat/file_vfcntl

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ia10341538488ba3a8444df8e73fb5257b2a1f512
2021-07-17 07:39:15 -03:00
Xiang Xiao
a44f62ddf7 fs: should only apply umask to the userspace caller
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I608e3107aaf3d85045958163a742a1ef1107de91
2021-07-17 13:58:58 +09:00
Xiang Xiao
a0f2f6c362 fs: Move umask to task_info_s
and implement the related semantic correctly

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I0587341e2c505e41b5e24e55dba039ed200d65f2
2021-07-14 10:35:15 -03:00
Xiang Xiao
0148e1d501 fs: Support the root file system attributes(mode, uid, gid and time)
Note: all attributes is guarded by PSEUDOFS_ATTRIBUTES to save the space

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I664d60382e356068fd920f08aca5b4a49d8d92a9
2021-07-14 10:35:15 -03:00
Xiang Xiao
30135b5f18 vfs: stat_recursive should return -ENOSYS if stat callback is NULL
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ic342b6fff112d24c7355ddc8aa0937dfc544f1b4
2021-07-13 11:32:52 -03:00
Xiang Xiao
d539998446 fs: fstat shouldn't follow the symbol link
since the decision is already done at open time(from path to fd)

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Idc1df3c1d7fe60927b0643173f3f40895aaffd4c
2021-07-13 10:05:50 -03:00
Xiang Xiao
2fa1e55628 libc: Implement ttyname and ttyname_r
Note: this patch can get file path from root pseudo file handle,
but a general infrastructure is setup for other file system too.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I86cd79ebb741f2f43fdd398bb7498c40687d949b
2021-07-09 18:23:34 -03:00
zhouliang3
b7653e2fae fs: Add parent node of inode
N/A

Signed-off-by: zhouliang3 <zhouliang3@xiaomi.com>
Change-Id: I006c6172ced8edd47b9d4661880e04d3a4274854
2021-07-09 18:23:34 -03:00
Xiang Xiao
ee767021b3 fs: Remove the inernal st_count field from struct stat
it's more simple to reuse the resolve argument as the recursive count

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I826b2fef3af8b12aae230e4766ab09f8de988f8a
2021-07-09 16:38:51 -03:00
Xiang Xiao
643ee82e9c fs: pseudorename should call inode_release in the error path
to avoid the memory leak

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I8f7e4c8fb89188614d03dcf383a1707f25c65174
2021-07-09 15:18:32 -03:00
Xiang Xiao
13651cde99 fs: Move the default ioctl handle from nx_vioctl to file_vioctl
ensure the behaviour is same in all ioctl function variant

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ie67cfbc579004c8488ef5004c9be7fb179e77041
2021-07-08 09:17:04 -03:00
Jiuzhu Dong
554310bd78 fs/fs_dup2: correct check before dup
Change-Id: I2387ab447f0d8ed9429960922a72de35b2730acb
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-07-02 04:54:43 -05:00
Xiang Xiao
2e54df0f35 Don't include assert.h from public header file
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-06-03 08:36:03 -07:00
Xiang Xiao
d7f96003cf Don't include debug.h from public header file
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-06-01 06:42:02 +09:00
SPRESENSE
0cc0c3d503 fs/vfs: Fix bug that select() did not return
Fix bug that select() did not return when the nfds argument
was set to a negative value. The specification is that -1 is
set to the return value and EINVAL is set to errno.
2021-05-19 06:43:22 -05:00
wangshuo
cb54c838a0 vfs: adjust the error code of write to be consistent with read
According to POSIX, the EACCES will be set when file can not be
written access. Just like fs_read
2021-04-08 21:52:14 -05:00
chao.an
4abf8e6587 fs/epoll: correct the return value of epoll_ctl(2)
Signed-off-by: chao.an <anchao@xiaomi.com>
2021-03-23 00:28:18 -07:00
Jiuzhu Dong
e96c8b9283 fs: allocate file/socket dynamically
Change-Id: I8aea63eaf0275f47f21fc8d5482b51ffecd5c906
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-03-17 06:46:42 -07:00
YAMAMOTO Takashi
5652de3fdf fs: try blockproxy only if BCH is enabled 2021-03-17 01:25:16 -07:00
buyuer
3b23f3f0b8 vfs/epoll: Fix one bug of EPOLL_CTL_DEL.
N/A

Signed-off-by: buyuer <dingddding@163.com>
2021-03-13 10:55:41 -08:00
buyuer
c0f9c7b48c vfs: Fix epoll cannot work under 64-bit operating system
by switching to the real file handle

Signed-off-by: buyuer <dingddding@163.com>
2021-03-13 10:55:41 -08:00
Gustavo Henrique Nihei
330eff36d7 sourcefiles: Fix relative path in file header 2021-03-09 23:18:28 +08:00
ligd
48d49e5a7c mqueue: add poll support
Change-Id: I7e908f6a6c00158c0946587dd79ae3dc5d279d37
Signed-off-by: ligd <liguiding1@xiaomi.com>
2021-03-04 07:11:43 -08:00
Jiuzhu Dong
4d5a964f29 net: unify socket into file descriptor
Change-Id: I9bcd21564e6c97d3edbb38aed1748c114160ea36
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-03-03 19:01:41 -08:00
Xiang Xiao
e7c97ce7f0 fs: Merge fs_poll into poll_fdsetup
to avoid the user misuse the wrong funciton

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-02-23 08:39:48 -03:00
Alin Jerpelea
f9fb182809 Author: Gregory Nutt: update licenses to Apache
Update files from Gregory Nutt to Apache 2.0 license.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-02-05 12:15:56 -03:00
Alin Jerpelea
a544c3f499 fs: nxffs: nxstyle error fix
Fix nxstyle errors to pass CI

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-02-05 12:15:56 -03:00
Juha Niskanen
eaa34b458a fs/driver: update comments about block_proxy()
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>
2021-01-22 17:10:13 +01:00
Xiang Xiao
e49bae1300 fs: Remove fs_dupfd and fs_dupfd2 internal functions
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
2021-01-13 16:39:45 +08:00
Xiang Xiao
1e5bfa623a fs: file_dup2 shouldn't hold the file list lock
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
2021-01-13 16:39:45 +08:00
Xiang Xiao
0dc6990166 Fix nxstyle warning
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-01-13 08:57:58 +01:00
Xiang Xiao
0536953ded Kernel module should prefer functions with nx/kmm prefix
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-01-13 08:57:58 +01:00
Xiang Xiao
fe96250c40 fs: Make nx_vxxx and file_vxxx as internal functions
these functions are the implementation detail and then
don't need expose to external

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ief832391d5b42d1f1645907eb465631650490234
2021-01-12 17:08:22 +01:00
Xiang Xiao
1604fe0b4b fs: Remove file_detach since it is unefficient to call open and file_detach
the kernel user should call file_open directly instead

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I5bf7f661006f5d43739bc8618abfb4b983fde78d
2021-01-11 23:47:57 +01:00
Xiang Xiao
0032ddb8bf fs: Reimplement file_open to not depend on nx_open
on the other hand, open/nx_open call file_open instead

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I66990a77cdeb6ff18f7bf48a65bbc7b701dad552
2021-01-11 23:47:57 +01:00
Xiang Xiao
1aa69f4c73 fs: Remove the special hack for pty in nx_vopen
let's replace the content of file in place instead

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I538910d55815c7aec656c05dba4eab2fa1d6d964
2021-01-08 11:04:24 +08:00
Xiang Xiao
86fab49c46 fs: Fix the race condition in file_dup
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
2021-01-08 11:01:46 +08:00
Xiang Xiao
38a068b7e6 fs: Move file_close from fs/inode/fs_fileclose.c to fs/vfs/fs_close.c
just like other vfs public functions

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Id833da690f88013095fbbe1ea425227f808a5608
2021-01-03 15:45:52 +01:00
Xiang Xiao
cfd7390676 fs: Let files_allocate return -EMFILE instead ERROR
since the internal function normally return the error code directly

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ifccbcfb07d2f8f7d5fef1b2a867538f51b73655c
2021-01-03 11:48:35 +01:00
Xiang Xiao
14a3c35917 fs: Ensure fs_dupfd2 always return fd2 in the sucessful path
it's wrong to return file_dup2 directly since file_dup2 never return file handle

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I943537849c75d83b3d646a6a22f035187d9fd521
2021-01-03 00:03:22 +01:00
Xiang Xiao
63bc3efd25 fs: Remove inode null check from file_dup and fs_dupfd2
since the same check already done in file_dup2

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I7ba1309c55be1ac564f798df02fc6725c4a0d469
2021-01-03 00:03:22 +01:00
Xiang Xiao
0defe43282 OS internal function should indicate the error by return negative value
instead to change errno value by calling set_errno

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-12-31 09:37:29 +01:00
Brennan Ashton
1473768ffc vfs: Fix unsed label warning
Signed-off-by: Brennan Ashton <bashton@brennanashton.com>
2020-12-30 21:20:17 -06:00
ligd
3386941a10 fs: remove INODE_IS_SPECIAL() use others instead
Change-Id: I949f1ad012836e6cb233d5362fe2542732b5ecf4
Signed-off-by: ligd <liguiding1@xiaomi.com>
2020-12-30 12:21:02 -06:00
ligd
f5b59e287a fs/vfs: add nx_unlink support
Change-Id: If9009cb7301bb4e49bdce3aea2d56c243256f5a2
Signed-off-by: ligd <liguiding1@xiaomi.com>
2020-12-30 12:21:02 -06:00
chao.an
39146ee367 fs/epoll: add asynchronous epoll control support
In current implementation, the asynchronous call "epoll_ctl()" unable
to wakeup the thread if pending on "epoll_wait()", the newly
added/delete FD cannot be used in the current waiting list,
this patch uses a reserved file object to wake up pending poll internal,
re-traverse the waiting list when a new event comes.

Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-27 11:08:12 -06:00
chao.an
0114f7c58f fs/epoll: add sanity check to handle invalid control option
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-12-27 11:08:12 -06:00
Xiang Xiao
29c6f11239 fs: Skip call fs_checkfd if fd < 3 in fs_fdopen
since the stdin, stdout and stderr may initialize later
in userspace if CONFIG_DEV_CONSOLE isn't enabled.
Note: it isn't bigger issue here to skip the check because
vfs will check the validation again in read and write syscall

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-12-16 07:53:21 +09:00
YAMAMOTO Takashi
8ce63a47fc fs/vfs/fs_epoll.c: Fix syslog formats 2020-11-20 22:22:53 -08:00
chao.an
6aee468d72 eventfd: remove the unique minor limit
The unique minor limit of 255 will overflow easily in
some scenarios where eventfd needs to be create/destroy
frequently:

  while (1)
    {
      fd = eventfd(0, 0);  // minor++
      sleep(1);
      close(fd);
    }

remove the unique minor limit.

Change-Id: I0ea1c825ce9b542c883166cb3e72574455ffdd0d
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-11-11 18:26:59 -08:00
Xiang Xiao
84b90e00f0 libc/stdio: Preallocate the stdin, stdout and stderr
to handle the uninitialized stdin/stdout/stderr gracefully
report here:
https://github.com/apache/incubator-nuttx/issues/2203

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-11-11 09:56:10 -08:00
dongjiuzhu
d452a05910 pollnotify: we should send poll events before semaphore incrementes.
There is a good case on sim platform:
When we input some cmd and click enter key to start application in terminal,
this context will change to application from IDLE loop. Althrough entey key '\r'
has been received to recv buffer and complete post semaphore of reader, but
pollnotify may not be called because context change. So when application run
poll function, because no events happend and poll enter wait, context will
again change to IDLE loop, this pollnotify of IDLE loop will run to send poll
events, poll function of applicaton will wake up. It's wrong!

Change-Id: I812a889f2e90781a9c3cb4b0251cccc4d32bebd1
Signed-off-by: dongjiuzhu <dongjiuzhu1@xiaomi.com>
2020-10-26 08:27:09 -03:00
dongjiuzhu
bcf3381d06 epoll_create: mofidy return value for epoll_create1
Change-Id: I422d7640c37105e1120db18b29d56547f447846f
Signed-off-by: dongjiuzhu <dongjiuzhu1@xiaomi.com>
2020-10-25 11:06:37 -07:00
Xiang Xiao
66057a4612 fs: Add the relative path support
all functions which accept the path argument should support the relative path:
https://pubs.opengroup.org/onlinepubs/9699919799/functions/open.html

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-24 03:13:27 -07:00
Xiang Xiao
43f0070582 fs: Simplify the symbol link process in inode_search
avoid the recursive call by moving the path catenation into _inode_search

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-22 16:39:47 +08:00
Xiang Xiao
b76c4672d6 vfs: Create a node as the root of pseudo file system
to remove the special process for root

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-16 11:44:57 -07:00
Yoshinori Sugino
5eefe8609c fs/vfs/fs_ioctl.c: Call F_GETFD to avoid clear other flag we will add in the future 2020-09-16 02:31:33 -07:00
Yoshinori Sugino
d4ec1ababd fs/vfs/fs_ioctl.c: Add FIOCLEX/FIONCLEX support 2020-09-16 02:31:33 -07:00
Xiang Xiao
6584bfa361 fs: Add static to stat_recursive
since it is called in one source file

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-16 01:06:53 -07:00
Xiang Xiao
1e848e5d8b fs: Call RELEASE_SEARCH to free resource in pseudorename
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-16 01:06:53 -07:00
Xiang Xiao
59fbd35609 fs: rename the mountpoint should go through pseudorename branch
to avoid this type of error:
mount -t procfs /a/b
mv /a/b /
mv: rename failed: 88

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-09-15 23:07:20 -07:00
Xiang Xiao
b0797263ca libc/stdio: Allocate file_struct dynamically
1.Reduce the default size of task_group_s(~512B each task)
2.Scale better between simple and complex application

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Ia872137504fddcf64d89c48d6f0593d76d582710
2020-09-11 17:58:17 +08:00
liuhaitao
bf06776f7c fs/vfs: reuse file_dup2 directly in file_dup to fix segfault issue
Or close the fd2 return by dup() would segment fault since filep->f_priv
null access.

Change-Id: I285c86f54bbb486d6c2c5aea53952f69083dca72
Signed-off-by: liuhaitao <liuhaitao@xiaomi.com>
2020-09-09 07:49:40 -07:00
Xiang Xiao
389b45359d epoll: Don't save fd into the field of epoll_data_t
since epoll_data_t is reserved to the caller and
then shouldn't be touched by the implementation

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-08-25 21:06:48 +01:00
Xiang Xiao
a354b9a9d1 fs/vfs: Implement EPOLLONESHOT flag
https://linux.die.net/man/2/epoll_ctl:
EPOLLONESHOT (since Linux 2.6.2)
Sets the one-shot behavior for the associated file descriptor.
This means that after an event is pulled out with epoll_wait(2)
the associated file descriptor is internally disabled and
no other events will be reported by the epoll interface.
The user must call epoll_ctl() with EPOLL_CTL_MOD to
rearm the file descriptor with a new event mask.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I6c0dc93e1cdae0e8cea5b487c7005de2da2c2ec3
2020-08-25 21:06:48 +01:00
Xiang Xiao
6c9ff72b9a fs/vfs: Implement epoll_pwait API
specified here:
https://linux.die.net/man/2/epoll_pwait

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I20106fdbe4f403ded7eab43f3523edd262e1d9a3
2020-08-25 21:06:48 +01:00
spiriou
0ddefd7c69 fs/vfs/fs_epoll: fix epoll_wait function 2020-08-20 11:49:20 -05:00
chao.an
b89737395b vfs/epoll: add epoll_create1(2) implement
Linux Programmer's Manual

NAME
       epoll_create, epoll_create1 - open an epoll file descriptor

...
SYNOPSIS
       #include <sys/epoll.h>

       int epoll_create1(int flags);
...

   epoll_create1()
       If flags is 0, then, other than the fact that the obsolete
       size argument is dropped, epoll_create1() is the same as
       epoll_create(). The following value can be included in flags
       to obtain different behavior:

       EPOLL_CLOEXEC
              Set the close-on-exec (FD_CLOEXEC) flag on the new file
              descriptor. See the description of the O_CLOEXEC flag in
              open(2) for reasons why this may be useful.

https://man7.org/linux/man-pages/man7/epoll.7.html
2020-08-17 23:41:13 -05:00
Xiang Xiao
85428036ab fs/vfs: Don't check CONFIG_NFILE_STREAMS for mkdir/rename/rmdir/fs_unlink
since these function don't depend on standard input/output function

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I011a387a1e2c1360e8f37716d512bd32f93ad467
2020-08-17 00:16:26 +01:00
spiriou
a4a9eb2f5a fs/vfs: Add file descriptor based events support 2020-07-31 15:09:35 -06:00
Xiang Xiao
91ed14c631 vfs/stat: Make the flag defintion more confirm POSIX standard
specified here:
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_stat.h.html

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-07-07 13:41:10 +01:00
Xiang Xiao
651eefc8f7 fs: Rename link to symlink
since the current implementation is really a symoblic link not hard link

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: I59d30d0a620b0b5714fe91bbe90d4405cf53d187
2020-07-01 11:58:12 +02:00
Xiang Xiao
b598ab43d7 fs: Implement lstat function
specified here:
https://pubs.opengroup.org/onlinepubs/009695399/functions/lstat.html

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
Change-Id: Iefc23a02d425ff84fa4027aea7da1181b01eced7
2020-06-30 13:09:58 -06:00
chao.an
f56e0e004b fs/setfd: correct the return value
Change-Id: I766f3760e7167a579b73673c44b70847ee06850b
Signed-off-by: chao.an <anchao@xiaomi.com>
2020-06-24 13:48:41 +01:00
Matias Nitsche
8c74a31fe2 vfs: fix support for PTY when CONFIG_NET is not enabled 2020-06-16 09:56:21 +08:00
Peter van der Perk
55d9e5f7af net: Add SocketCAN support 2020-06-15 08:07:19 -06:00
Gregory Nutt
1c002e1049 fs/vfs/fs_fstat: Correct fstat() for proxied block and MTD drivers.
Block and MTD drivers may be opened and managed as though they were character drivers.  But this is really sleight of hand; there is a hidden character driver proxy that mediates the interface to the block and MTD drivers in this case.

fstat(), however, did not account for this.  It would report the characteristics of the proxy character driver, not of the underlying block or MTD driver.

This change corrects that.  fstat now checks if the character driver is such a proxy and, if so, reports the characteristics of the underlying block or MTD driver, not the proxy character driver.
2020-05-28 17:34:09 +01:00
Oleg Evseev
3b75201447 fs/vfs/fs_stat.c: correct style 2020-05-26 16:56:10 +08:00
Oleg
dd9d9878ad fs/vfs/fs_stat.c: fill file size for mtd inode
Now when stat() is calling for mtd device it also set size of file as number of bytes of all sectors related to this mtd device.
2020-05-25 13:53:50 -06:00
Oleg
e1dbb6b2a9 Update fs/vfs/fs_stat.c
Co-authored-by: patacongo <spudarnia@yahoo.com>
2020-05-25 10:10:52 -06:00
Oleg
f6c3296f76 fs/vfs/fs_stat.c: fill file size for block inode 2020-05-25 10:10:52 -06:00
Gregory Nutt
154a87993f fs/vfs/fd_open.c: fs_fdopen() must not set errno
Functions within the OS must never set the errno value.  fs_fdopen() was setting the errno value.  Now, after some parameter changes, it reports errors via a negated errno integer return value as do most all other internal OS functions.
2020-05-23 15:22:09 +08:00
Gregory Nutt
a569006fd8 sched/: Make more naming consistent
Rename various functions per the quidelines of https://cwiki.apache.org/confluence/display/NUTTX/Naming+of+OS+Internal+Functions

    nxsem_setprotocol -> nxsem_set_protocol
    nxsem_getprotocol -> nxsem_get_protocol
    nxsem_getvalue -> nxsem_get_value
2020-05-17 14:01:00 -03:00
Xiang Xiao
517974787f Rename clock_systime[r|spec] to clock_systime_[ticks|timespec]
follow up the new naming convention:
https://cwiki.apache.org/confluence/display/NUTTX/Naming+of+OS+Internal+Functions
2020-05-10 14:35:50 -06:00
Gregory Nutt
a4218e2144 include/nuttx/sched.h: Make naming of all internal names consistent:
1. Add internal scheduler functions should begin with nxsched_, not sched_
2. Follow the consistent naming patter of https://cwiki.apache.org/confluence/display/NUTTX/Naming+of+OS+Internal+Functions
2020-05-09 14:19:08 -03:00
Xiang Xiao
4a3d28a957 fs: Move inode_checkflags from include/nuttx/fs/fs.h to fs/inode/inode.h
since this function is only used inside fs subsystem

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-08 07:20:49 -06:00
Xiang Xiao
6604cdb3f2 fs: Remove all LIBC_IOCTL_VARIADIC related stuff
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-05 08:27:58 -06:00
Xiang Xiao
390f9a5fb7 fs/vfs: Add nx_dup and nx_dup2 function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-04 07:20:02 -06:00
Xiang Xiao
1da8cd6b89 fs/vfs: Add nx_poll function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-04 07:20:02 -06:00
Xiang Xiao
4c680bfadc fs/vfs: Rename fdesc_poll to fs_poll
and reorder the function declaration

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-03 12:15:54 -06:00
Xiang Xiao
4b5360362e fs/vfs: Add nx_ioctl/nx_fcntl function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-03 12:15:54 -06:00
Xiang Xiao
65308eabb4 fs/vfs: Add nx_stat function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-03 12:15:54 -06:00
Xiang Xiao
6418d13b92 fs/vfs: Add nx_close function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-03 12:15:54 -06:00
Xiang Xiao
7980673b59 fs/vfs: Add nx_seek function
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-03 12:15:54 -06:00
Xiang Xiao
f8a809eb5b Fix nxstyle issue
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-01 10:43:47 -03:00
Xiang Xiao
eca7059785 Refine __KERNEL__ and CONFIG_BUILD_xxx usage in the code base
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2020-05-01 10:43:47 -03:00
YAMAMOTO Takashi
87f2e578ac Fix typos in comments 2020-04-13 09:50:32 -06:00
Gregory Nutt
ca2b0a3657 Fix more warnings noted in PR checks
arp/arp_notify.c:132:43: warning: for loop has empty body [-Wempty-body]
           prev = curr, curr = curr->nt_flink);
                                              ^

    sixlowpan/sixlowpan_tcpsend.c:806:31: warning: implicit conversion from 'unsigned int' to 'uint16_t' (aka 'unsigned short') changes value from 4294967295 to 65535 [-Wconstant-conversion]
                                  _SO_TIMEOUT(psock->s_sndtimeo));
                                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Numerous warnings like:

    ieee802154_getreq.c:93:3: warning: implicit declaration of function 'memcpy' is invalid in C99 [-Wimplicit-function-declaration]
      IEEE802154_EADDRCOPY(eaddr, req.attrval.mac.eaddr);
      ^

    nxfonts/nxfonts_cache.c:839:35: warning: for loop has empty body [-Wempty-body]
               fcache = fcache->flink);

    bluetooth/bluetooth_finddev.c💯11: warning: implicit declaration of function 'memcmp' is invalid in C99 [-Wimplicit-function-declaration]
          if (BLUETOOTH_ADDRCMP(dev->d_mac.radio.nv_addr, match->bf_addr))
              ^

    rwbuffer.c:559:7: warning: variable 'ret' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
      if (rwb->rhmaxblocks > 0 && rwb->rhnblocks > 0)
          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2020-04-13 02:09:34 +08:00
Gregory Nutt
a4d6af8343 Check return from nxsem_wait_initialize()
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.
2020-03-30 01:33:31 +01:00
Gregory Nutt
547a3cb3d9 Run all .c and .h files in previous commits through nxstyle. 2020-03-22 08:24:07 -05:00
Gregory Nutt
7a871e2f29 fs/: Remove support for CONFIG_FS_WRITABLE 2020-03-22 08:24:07 -05:00