reason:
We can utilize percpu storage to hold information about the
current running task. If we intend to implement this feature, we would
need to define two macros that help us manage this percpu information
effectively.
up_this_task: This macro is designed to read the contents of the percpu
register to retrieve information about the current
running task.This allows us to quickly access
task-specific data without having to disable interrupts,
access global variables and obtain the current cpu index.
up_update_task: This macro is responsible for updating the contents of
the percpu register.It is typically called during
initialization or when a context switch occurs to ensure
that the percpu register reflects the information of the
newly running task.
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>
Summary:
1.Add configuration to allocate memory from the specified section
2.Replace all memory operations (kmm_) in the vfs with
fs_heap_. When FS_HEAPSIZE > 0, memory is requested for the file system by specifying a configured heap location. By default (i.e. FS_HEAPSIZE=0) fs_heap_ is equivalent to kmm_
Signed-off-by: chenrun1 <chenrun1@xiaomi.com>
ap> cat pm
=================================================================
==30235==ERROR: AddressSanitizer: heap-buffer-overflow on address 0xf436edd9 at pc 0x03338a48 bp 0x9d1b6ca8 sp 0x9d1b6c98
READ of size 1 at 0xf436edd9 thread T0
#0 0x3338a47 in strncmp string/lib_strncmp.c:42
#1 0x371af87 in pm_get_file_index power/pm/pm_procfs.c:174
#2 0x371b066 in pm_open power/pm/pm_procfs.c:207
#3 0x3640d20 in procfs_open procfs/fs_procfs.c:419
#4 0x359bce2 in file_vopen vfs/fs_open.c:240
#5 0x359c431 in nx_vopen vfs/fs_open.c:312
#6 0x359cb53 in open vfs/fs_open.c:465
#7 0x33bccc9 in nsh_catfile /apps/nshlib/nsh_fsutils.c:140
#8 0x33b28cc in cmd_cat /apps/nshlib/nsh_fscmds.c:556
#9 0x33a434f in nsh_command /apps/nshlib/nsh_command.c:1164
#10 0x3381b8f in nsh_execute /apps/nshlib/nsh_parse.c:845
#11 0x338dc17 in nsh_parse_command /apps/nshlib/nsh_parse.c:2744
#12 0x338e273 in nsh_parse /apps/nshlib/nsh_parse.c:2828
#13 0x3390b47 in nsh_session /apps/nshlib/nsh_session.c:245
#14 0x337e90a in nsh_consolemain /apps/nshlib/nsh_consolemain.c:75
#15 0x337e7f7 in nsh_main /apps/system/nsh/nsh_main.c:74
#16 0x332b6e6 in nxtask_startup sched/task_startup.c:70
#17 0x323ec3f in nxtask_start task/task_start.c:134
#18 0x33636ea in pre_start sim/sim_initialstate.c:52
ap> cat net
=================================================================
==30303==ERROR: AddressSanitizer: heap-buffer-overflow on address 0xf4479a5a at pc 0x03338a48 bp 0x9d2b6ce8 sp 0x9d2b6cd8
READ of size 1 at 0xf4479a5a thread T0
#0 0x3338a47 in strncmp string/lib_strncmp.c:42
#1 0x5395d62 in netprocfs_open procfs/net_procfs.c:215
#2 0x3640d20 in procfs_open procfs/fs_procfs.c:419
#3 0x359bce2 in file_vopen vfs/fs_open.c:240
#4 0x359c431 in nx_vopen vfs/fs_open.c:312
#5 0x359cb53 in open vfs/fs_open.c:465
#6 0x33bccc9 in nsh_catfile /apps/nshlib/nsh_fsutils.c:140
#7 0x33b28cc in cmd_cat /apps/nshlib/nsh_fscmds.c:556
#8 0x33a434f in nsh_command /apps/nshlib/nsh_command.c:1164
#9 0x3381b8f in nsh_execute /apps/nshlib/nsh_parse.c:845
#10 0x338dc17 in nsh_parse_command /apps/nshlib/nsh_parse.c:2744
#11 0x338e273 in nsh_parse /apps/nshlib/nsh_parse.c:2828
#12 0x3390b47 in nsh_session /apps/nshlib/nsh_session.c:245
#13 0x337e90a in nsh_consolemain /apps/nshlib/nsh_consolemain.c:75
#14 0x337e7f7 in nsh_main /apps/system/nsh/nsh_main.c:74
#15 0x332b6e6 in nxtask_startup sched/task_startup.c:70
#16 0x323ec3f in nxtask_start task/task_start.c:134
#17 0x33636ea in pre_start sim/sim_initialstate.c:52
Signed-off-by: dulibo1 <dulibo1@xiaomi.com>
Signed-off-by: buxiasen <buxiasen@xiaomi.com>
There will be a large performance loss after SCHED_CRITMONITOR is enabled.
By isolating thread running time-related functions, CPU load can be run with less overhead.
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Signed-off-by: buxiasen <buxiasen@xiaomi.com>
After the number of threads exceeds the array size, it will not be displayed.
Any number of threads can be displayed using dynamic adaptation
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
This is a memory monitoring interface implemented with reference to Linux's PSI (Pressure Stall Information),
which can send notifications when the system's remaining memory is below the threshold.
The following example code sets two different thresholds.
When the system memory is below 10MB, a notification is triggered.
When the system memory is below 20 MB, a notification (POLLPRI event) is triggered every 1s.
```
int main(int argc, FAR char *argv[])
{
struct pollfd fds[2];
int ret;
if (argc == 2)
{
char *ptr = malloc(1024*1024*atoi(argv[1]));
printf("Allocating %d MB\n", atoi(argv[1]));
ptr[0] = 0;
return 0;
}
fds[0].fd = open("/proc/pressure/memory", O_RDWR);
fds[1].fd = open("/proc/pressure/memory", O_RDWR);
fds[0].events = POLLPRI;
fds[1].events = POLLPRI;
dprintf(fds[0].fd, "%llu -1", 1024LLU*1024 * 10);
dprintf(fds[1].fd, "%llu 1000000", 1024LLU*1024 * 20);
while (1)
{
ret = poll(fds, 2, -1);
if (ret > 0)
{
printf("Memory pressure: POLLPRI, %d\n", ret);
}
}
return 0;
}
```
https://docs.kernel.org/accounting/psi.html
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
This triggers `mm_free_delaylist()` before dumping status in PROTECTED
build, otherwise the `free` command still shows delaylist as `used`.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
Add a common method to format backtrace to buffer, so it can be used by both mm, fs and other possoble modules.
Signed-off-by: fangpeina <fangpeina@xiaomi.com>
Change the type of task group member to single list chain to
avoid accessing the memory allocator to improve the performance
Signed-off-by: chao an <anchao@lixiang.com>
Currently at runtime it is hard to know the exact build config of
the current NuttX instance. Thus it is inconvenient when there are
multiple configs for the same board. This patch attempts to solve
the issue by adding build config to /proc/version.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
These variables will trigger variable 'ret' set but not used warnings due to different configurations.
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Usage:
1. CONFIG_FS_PROCFS_MAX_STACK_RECORD > 0, such as 32,
2. add '-finstrument-functions' to CFLAGS for What you want to check
stack.
3. mount porcfs
4. cat /proc/<pid>/stack will print backtace & size
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
This moves task / thread cancel point logic from the NuttX kernel into
libc, while the data needed by the cancel point logic is moved to TLS.
The change is an enabler to move user-space APIs to libc as well, for
a coherent user/kernel separation.
hujun5