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>
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>
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>
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>
reason:
1 On different architectures, we can utilize more optimized strategies
to implement up_current_regs/up_set_current_regs.
eg. use interrupt registersor percpu registers.
code size
before
text data bss dec hex filename
262848 49985 63893 376726 5bf96 nuttx
after
text data bss dec hex filename
262844 49985 63893 376722 5bf92 nuttx
size change -4
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>
Make this_cpu is arch independent and up_cpu_index do that.
In AMP mode, up_cpu_index() may return the index of the physical core.
Signed-off-by: fangxinyong <fangxinyong@xiaomi.com>
Only in the non-critical region， nuttx can the respond to the irq and not hold the lock
When returning from the irq, there is no need to check whether the lock needs to be released
we also need keep restore_critical_section in svc call
test:
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>
Only in the non-critical region， nuttx can the respond to the irq and not hold the lock
When returning from the irq, there is no need to check whether the lock needs to be restored
test:
We can use qemu for testing.
compiling
make distclean -j20; ./tools/configure.sh -l qemu-armv8a:nsh_smp ;make -j20
running
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>
cpu0 cpu1:
user_main
signest_test
sched_unlock
nxsched_merge_pending
nxsched_add_readytorun
up_cpu_pause
arm_sigdeliver
enter_critical_section
Reason:
In the SMP, cpu0 is already in the critical section and waiting for cpu1 to enter the suspended state.
However, when cpu1 executes arm_sigdeliver, it is in the irq-disabled state but not in the critical section.
At this point, cpu1 is unable to respond to interrupts and
is continuously attempting to enter the critical section, resulting in a deadlock.
Resolve:
adjust the logic, do not entering the critical section when interrupt-disabled.
test:
We can use qemu for testing.
compiling
make distclean -j20; ./tools/configure.sh -l qemu-armv8a:nsh_smp ;make -j20
running
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>
chip/s698pm_cpustart.c: In function 's698pm_cpu_boot':
Error: chip/s698pm_cpustart.c:74:17: error: unused variable 'tcb' [-Werror=unused-variable]
struct tcb_s *tcb = this_task();
^~~
cc1: all warnings being treated as errors
make[1]: *** [Makefile:99: s698pm_cpustart.o] Error 1
make[1]: Target 'libarch.a' not remade because of errors.
make: *** [tools/LibTargets.mk:164: arch/sparc/src/libarch.a] Error 2
make: Target 'all' not remade because of errors.
/github/workspace/sources/nuttx/tools/testbuild.sh: line 370: /github/workspace/sources/nuttx/../nuttx/nuttx.manifest: No such file or directory
Normalize s698pm-dkit/smp
Signed-off-by: hujun5 <hujun5@xiaomi.com>
cpu0 thread0: cpu1:
sched_yield()
nxsched_set_priority()
nxsched_running_setpriority()
nxsched_reprioritize_rtr()
nxsched_add_readytorun()
up_cpu_pause()
IRQ enter
arm64_pause_handler()
enter_critical_section() begin
up_cpu_paused() pick thread0
arm64_restorestate() set thread0 tcb->xcp.regs to CURRENT_REGS
up_switch_context()
thread0 -> thread1
arm64_syscall()
case SYS_switch_context
change thread0 tcb->xcp.regs
restore_critical_section()
enter_critical_section() done
leave_critical_section()
IRQ leave with restore CURRENT_REGS
ERROR !!!
Reason:
As descript above, cpu0 swith task: thread0 -> thread1, and the
syscall() execute slowly, this time cpu1 pick thread0 to run at
up_cpu_paused(). Then cpu0 syscall execute, cpu1 IRQ leave error.
Resolve:
Move arm64_restorestate() after enter_critical_section() done
This is a continued fix with:
https://github.com/apache/nuttx/pull/6833
Signed-off-by: ligd <liguiding1@xiaomi.com>
Newly added logging in `sched/task_exit.c` obsoletes the existing
ones in `arch/up_exit()`, thus remove the latter to reduce duplications.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
When supporting high-priority interrupts, updating the
g_running_tasks within a high-priority interrupt may be
cause problems. The g_running_tasks should only be updated
when it is determined that a task context switch has occurred.
Signed-off-by: zhangyuan21 <zhangyuan21@xiaomi.com>
anjiahao