nuttx/nuttx-mirror
1
2
Fork 0
mirror of https://github.com/apache/nuttx.git synced 2025-01-13 09:49:21 +08:00
Code Issues Projects Releases Packages Wiki Activity Actions
57394 commits 18 branches 187 tags 413 MiB
Commit graph

53 commits

Author SHA1 Message Date
YAMAMOTO Takashi
aa8aa444c2 drivers/syslog/syslog_channel.c: remove a duplicated comment 2024-11-18 19:10:24 +08:00
YAMAMOTO Takashi
637a0f5af2 SYSLOG_DEFAULT: avoid using critical section in certain cases 
Note that SYSLOG_DEFAULT can be used in extreme situations,
including very early in the boot.
 2024-11-18 13:52:53 +08:00
YAMAMOTO Takashi
b990b25b8e Reapply "SYSLOG_DEFAULT: wrap up_putc/up_nputs calls with critical section" 
This reverts commit 35240d77fa.
 2024-11-18 13:52:53 +08:00
YAMAMOTO Takashi
35240d77fa Revert "SYSLOG_DEFAULT: wrap up_putc/up_nputs calls with critical section" 
This reverts commit f2aeb5e56f.

Because regressions are reported:
* https://github.com/apache/nuttx/pull/14722#issuecomment-2470710236
* https://github.com/apache/nuttx/pull/14722#issuecomment-2470778673
* https://github.com/apache/nuttx/issues/14749
 2024-11-13 11:15:34 +08:00
YAMAMOTO Takashi
f2aeb5e56f SYSLOG_DEFAULT: wrap up_putc/up_nputs calls with critical section 
This would avoid the undesirable intertactions with the serial driver
described in https://github.com/apache/nuttx/issues/14662.

Although I'm not entirely happy with this fix because it assumes
the particular implementations of up_putc/up_nputc and its association
to the serial devices, I haven't come up with better ideas for now.

An alternative is to place some serializations inside the target
specific serial (and/or whatever provides up_putc api) implementaitons.
But it isn't too attractive to put potentially complex logic into the
low-level machinaries, especially when we have a lot of similar copies
of it.

Another alternative is to deprecate up_putc. (at least for the purpose
of syslog.) But it seems at least some of users are relying on what
the current implementation provides heavily.

This commit also removes g_lowputs_lock because the critical section
would serve the purpose of the lock as well.
 2024-11-12 09:49:57 +08:00
Alin Jerpelea
286d37026c drivers: migrate to SPDX identifier 
Most tools used for compliance and SBOM generation use SPDX identifiers
This change brings us a step closer to an easy SBOM generation.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
 2024-11-06 18:02:25 +08:00
chao an
c6591c0f49 driver/serial: remove return value of up_putc() 
modify the prototype of up_putc(): remove the return value

The architecture code does not care about the return value of up_putc(), so removing it saves two statements:

Before:                                                    After:
de4c: e52de004  push  {lr}    @ (str lr, [sp, #-4]!)    |  de4c: e52de004  push  {lr}    @ (str lr, [sp, #-4]!)
de50: e24dd014  sub sp, sp, #20                         |  de50: e24dd014  sub sp, sp, #20
de54: e58d0004  str r0, [sp, #4]                        |  de54: e58d0004  str r0, [sp, #4]
de58: e30030f8  movw  r3, #248  @ 0xf8                  |  de58: e30030f8  movw  r3, #248  @ 0xf8
de5c: e3423000  movt  r3, #8192 @ 0x2000                |  de5c: e3423000  movt  r3, #8192 @ 0x2000
de60: e58d300c  str r3, [sp, #12]                       |  de60: e58d300c  str r3, [sp, #12]
de64: e59d1004  ldr r1, [sp, #4]                        |  de64: e59d1004  ldr r1, [sp, #4]
de68: e59d000c  ldr r0, [sp, #12]                       |  de68: e59d000c  ldr r0, [sp, #12]
de6c: ebfffe66  bl  d80c <pl011_putc>                   |  de6c: ebfffe66  bl  d80c <pl011_putc>
de70: e59d3004  ldr r3, [sp, #4]                        |  de70: e28dd014  add sp, sp, #20
de74: e1a00003  mov r0, r3                              |  de74: e49df004  pop {pc}    @ (ldr pc, [sp], #4)
de78: e28dd014  add sp, sp, #20                         |
de7c: e49df004  pop {pc}    @ (ldr pc, [sp], #4)        |

Signed-off-by: chao an <anchao@lixiang.com>
 2024-10-26 13:21:29 +08:00
yangsong8
8c13b8df1d syslog: convert \n to \r\n in syslog framework layer 
Signed-off-by: yangsong8 <yangsong8@xiaomi.com>
 2024-10-17 02:29:51 +08:00
chao an
a04e44ea75 syslog/channel: move syslog channel map into rodata 
add SYSLOG_REGISTER to support disable syslog channel register

Signed-off-by: chao an <anchao@lixiang.com>
 2024-09-26 16:10:29 +08:00
chao an
9abe737ef3 syslog/channel: add constant attribute if SYSLOG_IOCTL is not enabled 
move all private channel define from data to rodata

Signed-off-by: chao an <anchao@lixiang.com>
 2024-09-26 16:10:29 +08:00
chao an
a5251161c6 syslog/channel: rename syslog_channel() to syslog_channel_register() 
Change syslog API naming more reasonable:

1. rename syslog_channel() to syslog_channel_register()
2. rename syslog_channel_remove() to syslog_channel_unregister()

Signed-off-by: chao an <anchao@lixiang.com>
 2024-09-24 19:28:49 +08:00
chao an
85c39e96d6 global/variables: add g_ prefix to some global variables 
Signed-off-by: chao an <anchao@lixiang.com>
 2024-09-02 18:23:57 +08:00
yinshengkai
ce98f186c0 syslog: add syslog channel filtering function 
support to control the opening or closing of the specified channel through the syslogmask command at runtime

Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
 2023-07-13 01:04:07 +08:00
Xiang Xiao
efd1b838e6 syslog: Add sc_write_force callback 
It is used to write the log message to the channel immediately
when the log message is generated in the interrupt context, which
is faster than the normal force callback.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2023-06-23 09:55:19 -03:00
Tiago Medicci Serrano
71c8265e2d drivers/syslog: add mutex to syslog_default_write 
This commit reverts 19f269e54b and
substitutes the semaphore to a mutex, used as a locking mechanism
to avoid syslog messages being messed.

Considering SMP, threads running on different CPUs could call
syslog_default_write simultaneously. It then calls `up_nputs` that,
in most of the implementations, calls `up_putc` multiple times to
write to the serial. So, calling it at the same would cause syslog
messages to interfere with each other.

`up_nputs` and `up_putc` are low-level functions that could be used
during initialization and by the interrupt handler. Hence, it isn't
advisable to implement any locking mechanism on them. On the other
hand, syslog can also be used from the interrupt: if the interrupt
buffer is selected (`CONFIG_SYSLOG_INTBUFFER=y`), the syslog
messages will be flushed when we aren't serving an interrupt.
On the contrary, if it isn't set, it would call the registered
`sc_force` callback, which calls `syslog_default_putc` instead of
the lock-protected `syslog_default_write`.
 2023-06-02 10:17:54 +08:00
Xiang Xiao
af2b491420 driver/segger: Move SYSLOG_RTT config from drivers/syslog to drivers/segger 
and remove include/nuttx/syslog/syslog_rtt.h

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-11-27 23:10:39 +01:00
Xiang Xiao
d04ed9587b drivers/syslog: Call up_nputs in syslog_default_write 
since the buffer may not be terminated by null

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-07-21 21:26:10 +03:00
Xiang Xiao
19f269e54b syslog: Remove the lock from syslog_default_write 
to avoid the problem when is called from interrupt/signal handler

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-07-17 17:31:19 +03:00
Xiang Xiao
8254ad9159 drivers/syslog: Call up_puts instead up_putc one by one 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-07-17 17:31:19 +03:00
Nathan Hartman
1867bc2210 Fix compiler warnings (-Wunused-parameter) in various functions 
Fixes the -Wunused-parameter warning in:
* group/group_signal.c: group_signal()
* irq/irq_unexpectedisr.c: irq_unexpected_isr()
* task/task_spawn.c: nxtask_spawn_proxy()
* timer/timer_getoverrun.c: timer_getoverrun()
* misc/dev_null.c: devnull_read(), devnull_write(), devnull_poll()
* misc/dev_zero.c: devzero_read(), devzero_write(), devzero_poll()
* syslog/syslog_channel.c: syslog_default_write()
* syslog/syslog_device.c: syslog_dev_flush()
* grp/lib_initgroups.c: initgroups()
* misc/lib_mknod.c: mknod()
* misc/lib_glob.c: ignore_err()
* pthread/pthread_barrierinit.c: pthread_barrier_init()
* pthread/pthread_atfork.c: pthread_atfork()
* semaphore/sem_init.c: nxsem_init()
* stream/lib_nullinstream.c: nullinstream_getc()
* stream/lib_nulloutstream.c: nulloutstream_putc()
* stream/lib_libnoflush.c: lib_noflush()
* stream/lib_libsnoflush.c: lib_snoflush()
* string/lib_strerror.c: strerror()
* time/lib_gettimeofday.c: gettimeofday()
* time/lib_settimeofday.c: settimeofday()
* unistd/lib_pathconf.c: fpathconf(), pathconf()
* unistd/lib_getrusage.c: getrusage()
* unistd/lib_setrlimit.c: setrlimit()
* unistd/lib_getrlimit.c: getrlimit()
* unistd/lib_setpriority.c: setpriority()
 2022-07-12 11:42:34 +08:00
gaojiawei
335fc3dde2 syslog: Fixed a potential buffer overflow issue 
The `CONFIG_SYSLOG_MAX_CHANNELS` is user-specified, however, if the user
defines more channels than what `CONFIG_SYSLOG_MAX_CHANNELS` was defined as,
a potential buffer overflow occurred. Although the compiler does warn us about
that, we should explicitly tell the user this is an error.

Signed-off-by: gaojiawei <gaojiawei@xiaomi.com>
 2022-05-18 01:36:16 +08:00
Xiang Xiao
8391ad2721 syslog: Change g_syslog_default_sem to static function variable 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-03-20 12:37:09 +02:00
chao.an
40f63453f4 syslog/channel: minor fix to avoid unreachabled return 
Signed-off-by: chao.an <anchao@xiaomi.com>
 2022-03-19 11:13:44 +02:00
chao.an
4e08b1df93 drviers/syslog: correct the return value of default channel write 
Signed-off-by: chao.an <anchao@xiaomi.com>
 2022-03-14 08:59:34 +02:00
Xiang Xiao
01b791d773 drivers/syslog: Implement RTT based log channel 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2022-02-01 11:26:46 +01:00
ligd
ff0cd0baac Revert "drivers/syslog: Call up_puts in syslog_default_write instad up_putc" 
This reverts commit 174b240325.

Revert "drivers/syslog: Ensure the buffer zero terminate"

This reverts commit 1692aa7894.
 2021-11-01 06:54:05 -05:00
Xiang Xiao
174b240325 drivers/syslog: Call up_puts in syslog_default_write instad up_putc 
since some drivers(e.g. semihosting) have more fast implementation.

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2021-09-18 07:33:44 -03:00
David Sidrane
3e4b3ada5e syslog:syslog_channel:Fix compiler warnings 
syslog_channel.c:98:8: error: unknown type name 'sem_t'
   syslog/syslog_channel.c:99:14: error: 'g_syslog_default_sem'
   defined but not used
 2021-07-14 17:38:30 -07:00
Jiuzhu Dong
817259ec2d syslog/ramlog_channel: fix log confusion when multi task writing together 
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
 2021-07-03 11:24:21 -05:00
Jiuzhu Dong
b30bf4ff06 syslog/default_channel: fix log confusion when multi task writing together 
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
 2021-07-03 11:24:21 -05:00
Fotis Panagiotopoulos
b16de69e56 syslog: Added channel close callback. 2021-06-18 00:47:09 -05:00
Jiuzhu Dong
ed6b257fa4 syslog: support ramlog, up_putc, rpmsg_syslog coexist 
N/A

Change-Id: Ia58c7e195da7ad48f3018a131a78b2f01f94e741
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
 2021-06-02 08:45:20 -05:00
Fotis Panagiotopoulos
1dee243e29 syslog: Added multi device support in syslog_device. 2021-04-06 07:32:23 -05:00
Fotis Panagiotopoulos
f13adbe4bb Added support for multiple syslog channels. 2021-03-23 21:35:22 -07:00
Alin Jerpelea
e5b6305f4a drivers: Author Gregory Nutt: update licenses to Apache 
Gregory Nutt has submitted the SGA and we can migrate the licenses
 to Apache.

Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
 2021-03-04 18:32:27 -08:00
Xiang Xiao
972033a61f syslog: Check sc_flush in't NULL before invocation 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2020-04-27 13:48:56 +08:00
Xiang Xiao
e9773addfa syslog/rpmsg: Fix warning: ‘syslog_rpmsg_write’ defined but not used 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
 2020-04-27 13:48:56 +08:00
Xiang Xiao
1d339eb80c syslog: Make g_default_channel as static 2020-02-24 08:47:27 -06:00
Xiang Xiao
46e22d7fd9 syslog/rpmsg: Don't overwrite up_putc 
Hook into syslog subsystem like ramlog approach
 2020-02-24 08:47:13 -06:00
Juha Niskanen
ebc6f51641 Fix some typos 2019-09-17 10:46:23 -06:00
Xiang Xiao
f579c43754 drivers/syslog: Ensure interrupt log doesn't interlace in normal log. Don't call syslog_putc in syslog_default_write because syslog_putc will try to empty the interrupt buffer every time. 2018-11-09 08:46:32 -06:00
Xiang Xiao
1ee6083de3 Squashed commit of the following: 
drivers/syslog:  Channel configuration.  Decouple SYSLOG_SERIAL_CONSOLE and ARCH_LOWPUTC.  Since some hardware can output log to the special debug channel not serial.

    drivers/syslog/vsyslog.c: Support pre-pending a prefix string to log output if enabled.  This very useful to identify which cpu send out the log in AMP SoC.

    drivers/syslog/vsyslog.c: Make timestamp output same as linux kernel.  It's very useful if NuttX syslog retarget to Linux syslog.
 2018-08-26 06:49:35 -06:00
Gregory Nutt
20727d17c3 SYSLOG: Add option to buffer SYSLOG output to avoid interleaving. 2017-05-10 14:42:43 -06:00
Gregory Nutt
8a9bd9af9d SYSLOG: Some rather complex conditional compilation missed a case. 2016-06-21 15:08:09 -06:00
Gregory Nutt
e6a1d53869 Extend SYSLOG logic so that we can send SYSLOG output to a file. Not verified on initial commit. 2016-06-21 12:54:47 -06:00
Gregory Nutt
8f0c2cdd13 SYSLOG: Change configuration selections to assure that one and only one SYSLOG device can be selected. Also add a check to assure that it is appropriate to use up_putc as the low-level syslog device. Adds CONFIG_SYSLOG_CONSOLE and CONFIG_SYSLOG_SERIAL_CONSOLE. 2016-06-21 09:59:09 -06:00
Gregory Nutt
cf1375daf7 Break syslog_channel.c up into syslog_channel.c, syslog_putc.c, syslog_force.c and syslog_flush.c to limited what is brought into the link. Separate syslog_emergstream.c from syslog_stream.c. Didn't help in the case I was looking at. 2016-06-20 17:10:52 -06:00
Gregory Nutt
505ca542e8 Remove some last traces of lowvsyslog that were missed; Add a SYSLOG emergency channel for handling assertion output more cleanly 2016-06-20 16:11:50 -06:00
Gregory Nutt
f4c47636e7 SYSLOG: Debug output from the IDLE task should be forced like interrupt level output 2016-06-20 07:35:38 -06:00
Gregory Nutt
a43cdbff0b Add missing file header 2016-06-19 09:48:42 -06:00