RK3568 PCIe 异常排查
本帖最后由 firefly_zhongw 于 2021-8-24 10:49 编辑PCIe3.0控制器初始化设备系统异常
[ 21.202444] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
[ 21.202467] rcu: 3-...0: (1 GPs behind) idle=21a/1/0x4000000000000000 softirq=23/25 fqs=1856
[ 21.202473] rcu: (detected by 2, t=6302 jiffies, g=-1163, q=24)
[ 21.202482] Task dump for CPU 3:
[ 21.202488] rk-pcie Rrunning task 0 54 2 0x0000002a
[ 21.202497] Call trace:
[ 21.202511]__switch_to+0xe4/0x138
[ 21.202518]0x4
[ 21.205777] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[ 21.205796] rcu: 3-...!: (24 ticks this GP) idle=21a/1/0x4000000000000000 softirq=5/25 fqs=1
[ 21.205801] rcu: (detected by 2, t=6302 jiffies, g=-1183, q=2)
[ 21.205810] Task dump for CPU 3:
[ 21.205815] rk-pcie Rrunning task 0 54 2 0x0000002a
[ 21.205823] Call trace:
[ 21.205831]__switch_to+0xe4/0x138
[ 21.205837]0x4
[ 21.205846] rcu: rcu_sched kthread starved for 6299 jiffies! g-1183 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 ->cpu=0
[ 21.205851] rcu: RCU grace-period kthread stack dump:
[ 21.205856] rcu_sched I 0 11 2 0x00000028
[ 21.205863] Call trace:
[ 21.205869]__switch_to+0xe4/0x138
[ 21.205879]__schedule+0x2f4/0x930
[ 21.205886]schedule+0x38/0xa0
[ 21.205892]schedule_timeout+0x194/0x478
[ 21.205901]rcu_gp_kthread+0x4fc/0x840
[ 21.205909]kthread+0x12c/0x158
[ 21.205916]ret_from_fork+0x10/0x18异常原因:
如果系统卡住此 log 附近,则表明 PCIe3.0 的 PHY 工作异常。
解决方法:
硬件上使用PCIe
1. 外部晶振芯片的时钟输入是否异常,如果无时钟或者幅度异常,将导致 phy 无法锁定。
2. 检查 PCIE30_AVDD_0V9 和 PCIE30_AVDD_1V8 电压是否满足要求。
硬件上不使用PCIe
kernel 的 dts 里把 PCIe disabled。
&pcie30phy {
status = "disabled";
};
&pcie3x2 {
status = "disabled";
};
我们又遇到如下:
[ 1.206302] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 2.219630] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 3.232964] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 4.242959] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x1
[ 5.256292] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 6.269624] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 7.286295] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 8.296290] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 9.309624] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 10.326290] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 11.336288] rk-pcie 3c0800000.pcie: PCIe Link Fail
[ 11.336308] rk-pcie 3c0800000.pcie: failed to initialize host
[ 21.592941] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
[ 21.592963] rcu: 2-...0: (2 GPs behind) idle=7c2/1/0x4000000000000000 softirq=57/59 fqs=1811
[ 21.592969] rcu: (detected by 0, t=6302 jiffies, g=-1039, q=289)
[ 21.592978] Task dump for CPU 2:
[ 21.592983] swapper/0 Rrunning task 0 1 0 0x0000002a
[ 21.592992] Call trace:
[ 21.593007]__switch_to+0xe4/0x138
[ 21.593025]enqueue_task_fair+0x48c/0x1868
[ 84.609593] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
[ 84.609615] rcu: 2-...0: (2 GPs behind) idle=7c2/1/0x4000000000000000 softirq=57/59 fqs=7335
[ 84.609620] rcu: (detected by 0, t=25207 jiffies, g=-1039, q=289)
[ 84.609628] Task dump for CPU 2:
[ 84.609633] swapper/0 Rrunning task 0 1 0 0x0000002a
[ 84.609640] Call trace:
[ 84.609649]__switch_to+0xe4/0x138
[ 84.609658]enqueue_task_fair+0x48c/0x1868
rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
rcu: 2-...0: (2 GPs behind) idle=7c2/1/0x4000000000000000 softirq=57/59 fqs=12974
rcu: (detected by 1, t=44112 jiffies, g=-1039, q=289)
Task dump for CPU 2:
swapper/0 Rrunning task 0 1 0 0x0000002a
Call trace:
__switch_to+0xe4/0x138
enqueue_task_fair+0x48c/0x1868,然后机卡死了,这个能办看下么 自己做的底板,由于PCIe导致的问题,优先排查PCIe相关的硬件电路以及信号。 楼主,请问一下,您的问题解决了吗?遇到相似的问题 [ 6.299285] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
start rknn server, version:1.3.0 (121b661 build: 2022-04-29 11:11:47)
I NPUTransfer: Starting NPU Transfer Server, Transfer version 2.1.0 (b5861e7@2020-11-23T11:50:51)
[ 7.312616] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
[ 8.325947] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
[ 9.339280] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
[ 10.355946] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
[ 11.365986] rk-pcie 3c0800000.pcie: PCIe Link Fail
[ 11.366020] rk-pcie 3c0800000.pcie: failed to initialize host
我调试pcie也遇到了类似的错误,该如何解决呀
1786472307 发表于 2024-6-18 20:32
[ 6.299285] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
start rknn server, version:1.3. ...
您好,解决了没?我也遇到这个问题了
珂望5826 发表于 2024-6-19 11:40
您好,解决了没?我也遇到这个问题了
没有
1786472307 发表于 2024-6-18 20:32
[ 6.299285] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x3
start rknn server, version:1.3. ...
你好,这个问题解决了吗? 喔亮晶晶 发表于 2024-6-29 16:25
你好,这个问题解决了吗?
硬件存在问题,重新还接一遍就解决了
liuwen1987 发表于 2021-9-1 19:04
我们又遇到如下:
[ 1.206302] rk-pcie 3c0800000.pcie: PCIe Linking... LTSSM is 0x0
[ 2.219630 ...
您好,请问题解决了嘛
页:
[1]