sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

cpu: in SMP pthread_cancel occasionally deadlock

Browse Source 
In smp when cpu0 calls up_cpu_resume to release the cpu1 lock, another locked cpu1 did not execute immediately,
and soon cpu0 called up_cpu_resume again, now cpu1 unable to respond to the interrupt at this time, resulting in a deadlock.
Our solution is to restore cpu1 execution from asynchronous to synchronous to ensure that cpu1 is restored.

Signed-off-by: hujun5 <hujun5@xiaomi.com>
This commit is contained in:
hujun5 2023-04-07 17:11:33 +08:00 committed by Xiang Xiao
parent 9971dfd0c1
commit 6063c0516e
1 changed files with 13 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
arch/arm64/src/common/arm64_cpupause.c
View File

@ -57,6 +57,7 @@
static volatile spinlock_t g_cpu_wait[CONFIG_SMP_NCPUS];
static volatile spinlock_t g_cpu_paused[CONFIG_SMP_NCPUS];
static volatile spinlock_t g_cpu_resumed[CONFIG_SMP_NCPUS];
/****************************************************************************
* Public Functions
@ -135,6 +136,10 @@ int up_cpu_paused(int cpu)
spin_unlock(&g_cpu_paused[cpu]);
/* Ensure the CPU has been resumed to avoid causing a deadlock */
spin_lock(&g_cpu_resumed[cpu]);
/* Wait for the spinlock to be released. The requesting CPU will release
* the spinlock when the CPU is resumed.
*/
@ -163,6 +168,7 @@ int up_cpu_paused(int cpu)
arm64_restorestate(tcb->xcp.regs);
spin_unlock(&g_cpu_wait[cpu]);
spin_unlock(&g_cpu_resumed[cpu]);
return OK;
}
@ -331,5 +337,12 @@ int up_cpu_resume(int cpu)
!spin_islocked(&g_cpu_paused[cpu]));
spin_unlock(&g_cpu_wait[cpu]);
/* Ensure the CPU has been resumed to avoid causing a deadlock */
spin_lock(&g_cpu_resumed[cpu]);
spin_unlock(&g_cpu_resumed[cpu]);
return OK;
}