currently, nuttx implements readv/writev on the top of read/write.
while it might work for the simplest cases, it's broken by design.
for example, it's impossible to make it work correctly for files
which need to preserve data boundaries without allocating a single
contiguous buffer. (udp socket, some character devices, etc)
this change is a start of the migration to a better design.
that is, implement read/write on the top of readv/writev.
to avoid a single huge change, following things will NOT be done in
this commit:
* fix actual bugs caused by the original readv-based-on-read design.
(cf. https://github.com/apache/nuttx/pull/12674)
* adapt filesystems/drivers to actually benefit from the new interface.
(except a few trivial examples)
* eventually retire the old interface.
* retire read/write syscalls. implement them in libc instead.
* pread/pwrite/preadv/pwritev (except the introduction of struct uio,
which is a preparation to back these variations with the new
interface.)
Blocking while running a signal handler is not advisable, instead write
the log string character by character.
There is also a potential for a deadlock, as discussed in #6618
Note: querying for rtcb->sigdeliver is not 100% ideal, as it only tells
_if_ a signal handler has been queued, not if it is running. However, it
makes syslog safe / usable which is a debug feature anyhow.
Using ONESHOT_CURRENT retrieves the tick number multiplied by tick time; thus
it doesn't give the accurate monotonic time - it is quantized by
the tick time. This cannot be used as a ndelay timer, it would always loop
at least to the end of the ongoing tick.
Revert the up_udelay to use the original "coarse" looping. The "accurate" udelay,
if such is needed, should either be done under arch specific code, or there should be
a function for getting the accurate time that is available for all the platforms.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
CC: clk/clk_fixed_rate.c clk/clk_divider.c:177:14: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(rate - now) < abs(rate - best);
^
clk/clk_divider.c:177:14: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(rate - now) < abs(rate - best);
^~~
clk/clk_divider.c:177:32: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(rate - now) < abs(rate - best);
^
clk/clk_divider.c:177:32: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(rate - now) < abs(rate - best);
^~~
CC: mm_heap/mm_initialize.c 2 warnings generated.
clk/clk.c:1324:11: warning: variable 'irqflags' is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized]
if (clk->parents == NULL)
^~~~~~~~~~~~~~~~~~~~
clk/clk.c:1341:19: note: uninitialized use occurs here
clk_list_unlock(irqflags);
^~~~~~~~
clk/clk.c:1324:7: note: remove the 'if' if its condition is always false
if (clk->parents == NULL)
^~~~~~~~~~~~~~~~~~~~~~~~~
clk/clk.c:1255:22: note: initialize the variable 'irqflags' to silence this warning
irqstate_t irqflags;
^
= 0
CC: clk/clk_mux.c clk/clk_multiplier.c:110:14: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(rate - new) < abs(rate - best);
^
clk/clk_multiplier.c:110:14: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(rate - new) < abs(rate - best);
^~~
clk/clk_multiplier.c:110:32: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(rate - new) < abs(rate - best);
^
clk/clk_multiplier.c:110:32: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(rate - new) < abs(rate - best);
^~~
clk/clk_mux.c:47:14: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(now - rate) < abs(best - rate);
^
clk/clk_mux.c:47:14: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(now - rate) < abs(best - rate);
^~~
clk/clk_mux.c:47:32: warning: taking the absolute value of unsigned type 'unsigned int' has no effect [-Wabsolute-value]
return abs(now - rate) < abs(best - rate);
^
clk/clk_mux.c:47:32: note: remove the call to 'abs' since unsigned values cannot be negative
return abs(now - rate) < abs(best - rate);
^~~
AS: sim/sim_fork_x86.S 2 warnings generated.
1 warning2 warnings generated.
generated.
iperf.c:325:14: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'unsigned long' [-Wformat]
now_len -last_len,
^~~~~~~~~~~~~~~~~
iperf.c:340:14: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'uintmax_t' (aka 'unsigned long') [-Wformat]
now_len,
^~~~~~~
CC: misc/rpmsgblk_server.c 2 warnings generated.
:28: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'uintmax_t' (aka 'unsigned long') [-Wformat]
(uintmax_t)skip);
^~~~~~~~~~~~~~~
nsh_dbgcmds.c:473:20: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'uintmax_t' (aka 'unsigned long') [-Wformat]
(uintmax_t)position);
^~~~~~~~~~~~~~~~~~~
CC: nsh_main.c 2 warnings generated.
return INTMAX_MIN;
~~~~~~ ^~~~~~~~~~
CC: nsh_system.c note: expanded from macro 'INTMAX_MIN'
^~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:65:41: note: expanded from macro 'INT64_MIN'
~~~~~~~~~~~^~~
inttypes/lib_strtoimax.c:103:37: warning: implicit conversion from 'long long' to 'intmax_t' (aka 'long') changes value from -9223372036854775808 to 0 [-Wconstant-conversion]
return (accum == limit) ? INTMAX_MIN : -(intmax_t)accum;
~~~~~~ ^~~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:136:29: note: expanded from macro 'INTMAX_MIN'
^~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:65:41: note: expanded from macro 'INT64_MIN'
~~~~~~~~~~~^~~
inttypes/lib_strtoimax.c:106:17: warning: result of comparison of constant 9223372036854775807 with expression of type 'uintmax_t' (aka 'unsigned long') is always false [-Wtautological-constant-out-of-range-compare]
if (accum > INTMAX_MAX)
~~~~~ ^ ~~~~~~~~~~
inttypes/lib_strtoimax.c:109:18: warning: implicit conversion from 'long long' to 'intmax_t' (aka 'long') changes value from 9223372036854775807 to -1 [-Wconstant-conversion]
return INTMAX_MAX;
~~~~~~ ^~~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:137:29: note: expanded from macro 'INTMAX_MAX'
^~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:66:29: note: expanded from macro 'INT64_MAX'
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Users/vela/work/nuttx/include/arch/inttypes.h:35:23: note: expanded from macro 'INT64_C'
^~~~~~~
<scratch space>:12:1: note: expanded from here
9223372036854775807ll
^~~~~~~~~~~~~~~~~~~~~
syslog/vsyslog.c:212:32: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'uintmax_t' (aka 'unsigned long') [-Wformat]
, (uintmax_t)ts.tv_sec
^~~~~~~~~~~~~~~~~~~~
CC: iob/iob_free.c 4 warnings generated.
1 warning generated.
CC: mqueue/mq_msgqalloc.c inttypes/lib_strtoumax.c:91:23: warning: implicit conversion from 'unsigned long long' to 'uintmax_t' (aka 'unsigned long') changes value from 18446744073709551615 to 4294967295 [-Wconstant-conversion]
accum = UINTMAX_MAX;
~ ^~~~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:140:29: note: expanded from macro 'UINTMAX_MAX'
^~~~~~~~~~
/Users/vela/work/nuttx/include/stdint.h:67:29: note: expanded from macro 'UINT64_MAX'
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Users/vela/work/nuttx/include/arch/inttypes.h:36:23: note: expanded from macro 'UINT64_C'
^~~~~~~~
<scratch space>:8:1: note: expanded from here
18446744073709551615ull
^~~~~~~~~~~~~~~~~~~~~~~
CC: icmp/icmp_ioctl.c 1 warning generated.
time/lib_strftime.c:584:52: warning: format specifies type 'uintmax_t' (aka 'unsigned long long') but the argument has type 'uintmax_t' (aka 'unsigned long') [-Wformat]
len = snprintf(dest, chleft, "%ju", (uintmax_t)mktime(&tmp));
~~~ ^~~~~~~~~~~~~~~~~~~~~~~
%ju
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
that pr requires chip turn on CONFIG_DRIVERS_BLUETOOTH to use bluetooth,
but not all defconig enable this option, so let's map bt_driver_register
to bt_netdev_register in header file in this case, and revert the unnessary
change in the related chip and board folders.
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
rootcause: in btuart_rxwork, read data in blocking mode and btuart_read do three times, maybe remote send some packets one time,so it wont read the sencond packets.
Signed-off-by: wangzhi7 <wangzhi7@xiaomi.com>
add bt_driver_register interface, which could handle
these cases:bth4 bth5 btbridge btslip and btuart_lowerhalf_s etc.
Signed-off-by: chengkai <chengkai@xiaomi.com>
add btuart_create interface, which would be more
flexible in complex cases. And extract btuart_register interface.
Signed-off-by: chengkai <chengkai@xiaomi.com>
BLUETOOTH_MAX_FRAMELEN buffer size is only for LE only mode.
then we need to increase HCI buffer size in BR/EDR and LE mode.
Signed-off-by: chengkai <chengkai@xiaomi.com>
when downloading rtk firmware with vendor hci cmd sending to
bt bridge, which would not filter that hci cmd.
Signed-off-by: chengkai <chengkai@xiaomi.com>
h5 ack which send to controller always timeout, which causing
controller would send repend hci data until host send back h5 ack.
Signed-off-by: chengkai <chengkai@xiaomi.com>
bt bridge filter would not match when ACL in data
head with Packet_Boundary_Flag 0b10. in that case connect handle
is the first 3 octets of the packet, which not match
BT_HCI_EVT_LE_CONN_COMPLETE hci event alloc handle.
Signed-off-by: chengkai <chengkai@xiaomi.com>
serial/serial_cmsdk.c: In function 'uart_cmsdk_ioctl':
serial/serial_cmsdk.c:544:10: warning: 'ret' may be used uninitialized in this function [-Wmaybe-uninitialized]
544 | return ret;
| ^~~
Signed-off-by: Peter Bee <bijunda1@xiaomi.com>
A use-after-free problem occurs when there are multiple remotes in the list `g_rpmsg` and the matching remote is not the last item in the list.
Log
# Export the device "/dev/LOCAL_DEV" to remote "REMOTE_CPU"
ap> testdev -d 2 -c "REMOTE_CPU" -l "/dev/LOCAL_DEV"
[ap] kasan_report: kasan detected a read access error, address at 0x3c3d4740,size is 4, return address: 0x2c33620f
[ap] kasan_show_memory: Shadow bytes around the buggy address:
[ap] kasan_show_memory: 0x3c3d46f0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4700: aa aa aa aa cc cc cc cc cc cc cc cc cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4710: 40 47 3d 3c ed 61 33 2c 00 00 00 00 00 00 00 00
[ap] kasan_show_memory: 0x3c3d4720: 00 00 00 00 00 00 00 00 00 00 00 00 cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4730: 55 55 55 55 38 00 00 00 02 2c 00 00 cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4740:[00 00 00 00]66 e0 42 3c cc cc cc cc cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4750: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4760: aa aa aa aa 38 00 00 00 01 2c 00 00 cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4770: 50 57 44 3d 2f 00 cc cc cc cc cc cc cc cc cc cc
[ap] kasan_show_memory: 0x3c3d4780: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc
[ap] dump_assert_info: Current Version: NuttX ****** ***** *** 12.3.0 **********-***** *** ** 2024 **:**:** arm
[ap] dump_assert_info: Assertion failed panic: at file: kasan/hook.c:187 task: testdev process: testdev 0x2ca20495
$ addr2line -fe nuttx/nuttx 0x2c33620f
rpmsgdev_server_created
/workspace/nuttx/drivers/misc/rpmsgdev_server.c:529
# Line 529 => strcmp()
Signed-off-by: wangjianyu3 <wangjianyu3@xiaomi.com>
note/note_driver.c:639:21: warning: 'strlen' reading 1 or more bytes from a region of size 0 [-Wstringop-overread]
639 | namelen = strlen(tcb->name);
| ^~~~~~~~~~~~~~~~~
Signed-off-by: cuiziwei <cuiziwei@xiaomi.com>
note/note_driver.c:1405:11: runtime error: null pointer passed as argument 2, which is declared to never be null
#0 0x33bf5cc in sched_note_event_ip note/note_driver.c:1405
#1 0x33bfb57 in note_driver_instrument_enter note/note_initialize.c:55
#2 0x347b084 in __cyg_profile_func_enter misc/lib_instrument.c:68
#3 0x34179de in binder_initialize binder/binder.c:669
#4 0x339a936 in drivers_initialize /home/cuiziwei/vela/happy/nuttx/drivers/drivers_initialize.c:242
#5 0x335a179 in nx_start init/nx_start.c:632
#6 0x32f755c in main sim/sim_head.c:180
#7 0xf6821518 (/lib/i386-linux-gnu/libc.so.6+0x21518) (BuildId: 7f64b917aaa97b9680d8e44931bf7611c5a1f036)
#8 0xf68215f2 in __libc_start_main (/lib/i386-linux-gnu/libc.so.6+0x215f2) (BuildId: 7f64b917aaa97b9680d8e44931bf7611c5a1f036)
#9 0x32b401a in _start (/home/cuiziwei/vela/happy/nuttx/nuttx+0x32b401a) (BuildId: 33f8f7b361d44a008de87fea1bc970b22b48b700)
Signed-off-by: cuiziwei <cuiziwei@xiaomi.com>
avoid the instrument doesn't get registered in some case
Signed-off-by: cuiziwei <cuiziwei@xiaomi.com>
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
There will be compilation errors about undefined reference to `enter_critical_section'.
Fixed by include irq.h
Signed-off-by: fangpeina <fangpeina@xiaomi.com>
ivshmem based driver can use this api to judge weather current
ivshmem device support irq or not, and use polling mode or irq
mode to process the event.
Signed-off-by: wangyongrong <wangyongrong@xiaomi.com>
1. add IS_ALIGNED() definitions for NuttX;
2. replace all the ALIGN_UP() and ALIGN_DOWN() to use common
align implementation;
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
Use bit mask method to represent the command, because it's more
convenient to express multiple characteristics simultaneously.
Signed-off-by: Yongrong Wang <wangyongrong@xiaomi.com>
Usually rpmsg ping logs are needed during testing,
but too many other module logs can easily overwrite it,
so improve the rpmsg ping log level, later we can setlogmask to
a higher log level to mask other modules' logs.
Signed-off-by: Yongrong Wang <wangyongrong@xiaomi.com>
In some cases, we only need to use INSTRUMENTATION_DUMP to record custom data,
and we don't want the thread start and end information
Move this part to INSTRUMENTATION_SWTICH for recording
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
When using alarm_arch implementation, 64-bit time can be returned. Using unsign long will cause precision loss.
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Some app with same code runs on different cores in AMP mode,
need the physical core on which the function is called.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
1. Use critical_section to protect the fds array;
2. Use critical_section and sched lock to protect the poll notify, because
poll notify may cause context switch, delay the context swtich to
sched_unlock();
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
And remove the virtio_alloc/free_buf implementation in virtio.c, because
these two apis has been moved to the OpenAMP
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
notify_wait has been remove in remoteproc ops and virtio_dispatch and
has been moved to the struct rpmsg_virtio_device in new OpenAMP,
so change the rptun too,
Signed-off-by: wangyongrong <wangyongrong@xiaomi.com>
Do not format immediately when calling sched_note_printf, but delay formatting until dump trace.
After turning on SYSTEM_NOTE, similar asynchronous syslog functions can be achieved.
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Rpmsg dev server always open the real char device with nonblock mode,
so let client try to read/wrtie the read device every read/write
operation.
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
