developers must be careful to hold spinlocks and ensure all
of protected code is under control, so remove support for nested
spinlocks to improve performance.
Signed-off-by: chao an <anchao@lixiang.com>
When coredump to mtd, it maybe cost lots of time, do flush syslog can
make user access all log when coredump processing, should be better.
Signed-off-by: buxiasen <buxiasen@xiaomi.com>
set CONFIG_PRIORITY_INHERITANCE=y
set CONFIG_SEM_PREALLOCHOLDERS=0 or CONFIG_SEM_PREALLOCHOLDERS=8
#24 0x4dcab71 in __assert assert/lib_assert.c:37
#25 0x4d6b0e9 in nxsem_destroyholder semaphore/sem_holder.c:602
#26 0x4d80cf7 in nxsem_destroy semaphore/sem_destroy.c:80
#27 0x4d80db9 in sem_destroy semaphore/sem_destroy.c:120
#28 0x4dcb077 in nxmutex_destroy misc/lib_mutex.c:122
#29 0x4dc6611 in pipecommon_freedev pipes/pipe_common.c:117
#30 0x4dc7fdc in pipecommon_close pipes/pipe_common.c:397
#31 0x4ed4f6d in file_close vfs/fs_close.c:78
#32 0x6a91133 in local_free local/local_conn.c:184
#33 0x6a92a9c in local_release local/local_release.c:129
#34 0x6a91d1a in local_subref local/local_conn.c:271
#35 0x6a75767 in local_close local/local_sockif.c:797
#36 0x4e978f6 in psock_close socket/net_close.c:102
#37 0x4eed1b9 in sock_file_close socket/socket.c:115
#38 0x4ed4f6d in file_close vfs/fs_close.c:78
#39 0x4ed1459 in nx_close_from_tcb inode/fs_files.c:754
#40 0x4ed1501 in nx_close inode/fs_files.c:781
#41 0x4ed154a in close inode/fs_files.c:819
#42 0x6bcb9ce in property_get kvdb/client.c:307
#43 0x6bcd465 in property_get_int32 kvdb/common.c:270
#44 0x5106c9a in tz_offset_restore app/miwear_bluetooth.c:745
#45 0x510893f in miwear_bluetooth_main app/miwear_bluetooth.c:1033
#46 0x4dcf5c8 in nxtask_startup sched/task_startup.c:70
#47 0x4d70873 in nxtask_start task/task_start.c:134
#48 0x4e04a07 in pre_start sim/sim_initialstate.c:52
Signed-off-by: ligd <liguiding1@xiaomi.com>
set CONFIG_PRIORITY_INHERITANCE=y
set CONFIG_SEM_PREALLOCHOLDERS=0
semaphore/sem_holder.c:320:34: runtime error: member access within null pointer of type 'struct tcb_s'
#0 0xd8b540 in nxsem_boostholderprio semaphore/sem_holder.c:320
#1 0xd8c1cf in nxsem_boost_priority semaphore/sem_holder.c:703
#2 0xda5dfa in nxsem_wait semaphore/sem_wait.c:145
#3 0xda61d9 in nxsem_wait_uninterruptible semaphore/sem_wait.c:248
#4 0x12f2477 in media_service_thread0 /home/ligd/platform/dev/apps/examples/hello/hello_main.c:44
#5 0x1204154 in pthread_startup pthread/pthread_create.c:59
#6 0x1cd906f in pthread_start pthread/pthread_create.c:139
#7 0xe72fcb in pre_start sim/sim_initialstate.c:52
Signed-off-by: ligd <liguiding1@xiaomi.com>
When statistics when sched_cpuload_critMonitor, thread switching only calculates the current CPU's running time. Other CPU running threads are updated for updates
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>
gprof can analyze code hot spots based on scheduled sampling.
After adding the "-pg" parameter when compiling, you can view the code call graph.
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
If the write lock is already held by oneself and sine the write
lock can be recursively held, so, this operation can be converted to a write
lock to avoid deadlock.
Signed-off-by: dongjiuzhu1 <dongjiuzhu1@xiaomi.com>
after
text data bss dec hex filename
269732 51065 63335 384132 5dc84 nuttx
before
text data bss dec hex filename
269784 51065 63335 384184 5dcb8 nuttx
size -50
Signed-off-by: hujun5 <hujun5@xiaomi.com>
In the original implementation, if wdog is not ACTIVE, -EINVAL will be returned. The rp2040 i2c driver relies on this semantics to correctly release the semaphore.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@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>
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>
Some app with same code runs on different cores in AMP mode,
need known physical core id on which the function is called.
Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
bug:
thread 1: get mutexA
wait a sem
thread 2: wait mutexA
This situation will be mistakenly judged as a deadlock.
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
Signed-off-by: xuxingliang <xuxingliang@xiaomi.com>
reason:
To remove the "sync pause" and decouple the critical section from the dependency on enabling interrupts,
after that we need to further implement "schedlock + spinlock".
changelist
1 Modify the implementation of critical sections to no longer involve enabling interrupts or handling synchronous pause events.
2 GIC_SMP_CPUCALL attach to pause handler to remove arch interface up_cpu_paused_restore up_cpu_paused_save
3 Completely remove up_cpu_pause, up_cpu_resume, up_cpu_paused, and up_cpu_pausereq
4 change up_cpu_pause_async to up_send_cpu_sgi
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Change mqueue_msg_s to a dynamic array so that
mqueue msg can be malloc based on the msg size
of the current data when applied dynamically.
Signed-off-by: zhangyuan29 <zhangyuan29@xiaomi.com>
Add heap current used to note.
Plot it in segger sysview data plot.
Signed-off-by: xuxingliang <xuxingliang@xiaomi.com>
Signed-off-by: Neo Xu <neo.xu1990@gmail.com>
Record all memory allocation and release, save to ram, used to analyze memory allocation rate and memory usage
Its absolute value is not trustworthy because the memory will be allocated in thread A and released in thread B
netinit-5 [0] 0.105984392: tracing_mark_write: C|5|Heap Usage|96|free: heap: 0x606000000020 size:24, address: 0x603000000370
netinit-5 [0] 0.105996874: tracing_mark_write: C|5|Heap Usage|24|free: heap: 0x606000000020 size:72, address: 0x6070000008e0
nsh_main-4 [0] 3.825169408: tracing_mark_write: C|4|Heap Usage|2177665|free: heap: 0x606000000020 size:424, address: 0x614000000840
nsh_main-4 [0] 3.825228525: tracing_mark_write: C|4|Heap Usage|14977|free: heap: 0x606000000020 size:2162688, address: 0x7f80a639f800
nsh_main-4 [0] 3.825298789: tracing_mark_write: C|4|Heap Usage|15189|malloc: heap: 0x606000000020 size:20, address: 0x6030000003a0
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Signed-off-by: Neo Xu <neo.xu1990@gmail.com>
reason:
In the kernel, we are planning to remove all occurrences of up_cpu_pause as one of the steps to
simplify the implementation of critical sections. The goal is to enable spin_lock_irqsave to encapsulate critical sections,
thereby facilitating the replacement of critical sections(big lock) with smaller spin_lock_irqsave(small lock)
Signed-off-by: hujun5 <hujun5@xiaomi.com>
1. use '__attribute__((constructor))' mark initialize function
2. use '__attribute__((destructor))' mark uninitialize function
3. compile module with -fvisibility=hidden. use `__attribute__((visibility("default")))`
mark is need export symbol.so not need module_initialize to initialize export symbol.
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
reason:
In the kernel, we are planning to remove all occurrences of up_cpu_pause as one of the steps to
simplify the implementation of critical sections. The goal is to enable spin_lock_irqsave to encapsulate critical sections,
thereby facilitating the replacement of critical sections(big lock) with smaller spin_lock_irqsave(small lock)
Configuring NuttX and compile:
$ ./tools/configure.sh -l qemu-armv8a:nsh_smp
$ make
Running with qemu
$ qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic
-machine virt,virtualization=on,gic-version=3
-net none -chardev stdio,id=con,mux=on -serial chardev:con
-mon chardev=con,mode=readline -kernel ./nuttx
Signed-off-by: hujun5 <hujun5@xiaomi.com>
reason:
In the kernel, we are planning to remove all occurrences of up_cpu_pause as one of the steps to
simplify the implementation of critical sections. The goal is to enable spin_lock_irqsave to encapsulate critical sections,
thereby facilitating the replacement of critical sections(big lock) with smaller spin_lock_irqsave(small lock)
Configuring NuttX and compile:
$ ./tools/configure.sh -l qemu-armv8a:nsh_smp
$ make
Running with qemu
$ qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic
-machine virt,virtualization=on,gic-version=3
-net none -chardev stdio,id=con,mux=on -serial chardev:con
-mon chardev=con,mode=readline -kernel ./nuttx
Signed-off-by: hujun5 <hujun5@xiaomi.com>
reason:
In the kernel, we are planning to remove all occurrences of up_cpu_pause as one of the steps to
simplify the implementation of critical sections. The goal is to enable spin_lock_irqsave to encapsulate critical sections,
thereby facilitating the replacement of critical sections(big lock) with smaller spin_lock_irqsave(small lock)
Configuring NuttX and compile:
$ ./tools/configure.sh -l qemu-armv8a:nsh_smp
$ make
Running with qemu
$ qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic
-machine virt,virtualization=on,gic-version=3
-net none -chardev stdio,id=con,mux=on -serial chardev:con
-mon chardev=con,mode=readline -kernel ./nuttx
Signed-off-by: hujun5 <hujun5@xiaomi.com>
reason:
1 To improve efficiency, we mimic Linux's behavior where preemption disabling is only applicable to the current CPU and does not affect other CPUs.
2 In the future, we will implement "spinlock+sched_lock", and use it extensively. Under such circumstances, if preemption is still globally disabled, it will seriously impact the scheduling efficiency.
3 We have removed g_cpu_lockset and used irqcount in order to eliminate the dependency of schedlock on critical sections in the future, simplify the logic, and further enhance the performance of sched_lock.
4 We set lockcount to 1 in order to lock scheduling on all CPUs during startup, without the need to provide additional functions to disable scheduling on other CPUs.
5 Cpu1~n must wait for cpu0 to enter the idle state before enabling scheduling because it prevents CPUs1~n from competing with cpu0 for the memory manager mutex, which could cause the cpu0 idle task to enter a wait state and trigger an assert.
size nuttx
before:
text data bss dec hex filename
265396 51057 63646 380099 5ccc3 nuttx
after:
text data bss dec hex filename
265184 51057 63642 379883 5cbeb nuttx
size -216
Configuring NuttX and compile:
$ ./tools/configure.sh -l qemu-armv8a:nsh_smp
$ make
Running with qemu
$ qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic \
-machine virt,virtualization=on,gic-version=3 \
-net none -chardev stdio,id=con,mux=on -serial chardev:con \
-mon chardev=con,mode=readline -kernel ./nuttx
Signed-off-by: hujun5 <hujun5@xiaomi.com>
wd_timer handler will change current task, so need process
scheduler before wd timer handler, Ensure that the current
task is not updated before the RR process
Signed-off-by: zhangyuan29 <zhangyuan29@xiaomi.com>
nxsem_tickwait correctly sleeps more than 1 tick. But nxsem_tickwait_uninterruptible
may wake up to a signal (with -EINTR), in which case the tick + 1 must also
be taken into account. Otherwise the nxsem_tickwait_uninterruptible may
wake up 1 tick too early.
Also fix the nxsem_tickwait to return with -ETIMEDOUT if called with delay 0.
This is similar to e.g. posix sem_timedwait.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
1. extract dump from assert main flow
2. use OSINIT_PANIC for fatal error.
3. fix the method to judge kernel thread.
Signed-off-by: xuxingliang <xuxingliang@xiaomi.com>
1.nxsem_post wake up nxsched_smp_call;
2.stack smp_call_data_s may return;
3.nxsched_smp_call_handler access call_data->lock is not safety;
so adjust the unlock order
Signed-off-by: dulibo1 <dulibo1@xiaomi.com>
1. mainly for long time within critical_section
2. move out memcpy buffer from enter_critical_section()
3. let mq_send() return fail when mq buffer full in ISR
Signed-off-by: ligd <liguiding1@xiaomi.com>
LD: nuttx.elf
ld: /home/ligd/platform/mainline/nuttx/staging/libsched.a(mq_sndinternal.o): in function `wd_start_realtime':
/home/ligd/platform/mainline/nuttx/include/nuttx/wdog.h:274: undefined reference to `clock_realtime2absticks'
/home/ligd/platform/mainline/nuttx/include/nuttx/wdog.h:274:(.text+0xad): relocation truncated to fit: R_X86_64_PLT32 against undefined symbol `clock_realtime2absticks'
ld: /home/ligd/platform/mainline/nuttx/staging/libsched.a(mq_rcvinternal.o): in function `wd_start_realtime':
/home/ligd/platform/mainline/nuttx/include/nuttx/wdog.h:274: undefined reference to `clock_realtime2absticks'
/home/ligd/platform/mainline/nuttx/include/nuttx/wdog.h:274:(.text+0xad): relocation truncated to fit: R_X86_64_PLT32 against undefined symbol `clock_realtime2absticks'
make[1]: *** [Makefile:128: nuttx.elf] Error 1
make: *** [tools/Unix.mk:551: nuttx.elf] Error 2
Signed-off-by: ligd <liguiding1@xiaomi.com>
The feature depends on ARCH_USE_SEPARATED_SECTION
the different memory area has different access speed and cache
capability, so the arch can custom allocate them based on
section names to achieve performance optimization
test:
sim:elf
sim:sotest
Signed-off-by: dongjiuzhu1 <dongjiuzhu1@xiaomi.com>
reason:
Currently, if we need to schedule a task to another CPU, we have to completely halt the other CPU,
manipulate the scheduling linked list, and then resume the operation of that CPU. This process is both time-consuming and unnecessary.
During this process, both the current CPU and the target CPU are inevitably subjected to busyloop.
The improved strategy is to simply send a cross-core interrupt to the target CPU.
The current CPU continues to run while the target CPU responds to the interrupt, eliminating the certainty of a busyloop occurring.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
Signed-off-by: dulibo1 <dulibo1@xiaomi.com>
1.call_data->refcount is not remote_cpus;
2.nxsched_smp_call_add enqueue the call_data;
3.maybe nxsched_smp_call_handler is just doing;
4.dequeue the call_data but call_data->refcount is not corret;
5.unpredictability case will occur;
reason:
1In the scenario of active waiting, context switching is inevitable, and we can eliminate redundant judgments.
code size
before
hujun5@hujun5-OptiPlex-7070:~/downloads1/vela_sim/nuttx$ size nuttx
text data bss dec hex filename
262848 49985 63893 376726 5bf96 nuttx
after
hujun5@hujun5-OptiPlex-7070:~/downloads1/vela_sim/nuttx$ size nuttx
text data bss dec hex filename
263324 49985 63893 377202 5c172 nuttx
reduce code size by -476
Configuring NuttX and compile:
$ ./tools/configure.sh -l qemu-armv8a:nsh_smp
$ make
Running with qemu
$ qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic \
-machine virt,virtualization=on,gic-version=3 \
-net none -chardev stdio,id=con,mux=on -serial chardev:con \
-mon chardev=con,mode=readline -kernel ./nuttx
Signed-off-by: hujun5 <hujun5@xiaomi.com>
This commit simplified thread ID dispatching logic by integrating it into the `nxsig_dispatch` function.
Signed-off-by: ouyangxiangzhen <ouyangxiangzhen@xiaomi.com>
