![通过 pacman 安装 nvidia 驱动程序后,EndeavourOS 随机滞后](https://linux22.com/image/195083/%E9%80%9A%E8%BF%87%20pacman%20%E5%AE%89%E8%A3%85%20nvidia%20%E9%A9%B1%E5%8A%A8%E7%A8%8B%E5%BA%8F%E5%90%8E%EF%BC%8CEndeavourOS%20%E9%9A%8F%E6%9C%BA%E6%BB%9E%E5%90%8E.png)
我在 Lenovo Ideapad 3 15arh05 上使用 EndeavourOS,内核版本为 5.13.13.arch1-1。
pacman -Syy
我最近刚刚通过和更新了系统-Syu
,之后我通过 安装了 nvidia 驱动程序pacman -S nvidia
。现在系统有时会随机完全冻结。
一堆 kworker 进程突然出现在顶部,几乎总是占用 100% 的 CPU,同时dmesg
消息journalctl
如下所示:
[ 2194.247872] eno1: io_base_l = 0xffff, should be 0x3001
.
[ 2194.247877] eno1: mem_base_l = 0xffff, should be 0x4004
.
[ 2194.247880] eno1: mem_base_h = 0xffff, should be 0xc370
.
[ 2194.247884] eno1: resv_0x1c_l = 0xffff, should be 0x0000
.
[ 2194.247887] eno1: resv_0x1c_h = 0xffff, should be 0x0000
.
[ 2194.247891] eno1: resv_0x20_l = 0xffff, should be 0x0004
.
[ 2194.247895] eno1: resv_0x20_h = 0xffff, should be 0xc370
.
[ 2194.247898] eno1: resv_0x24_l = 0xffff, should be 0x0000
.
[ 2194.247901] eno1: resv_0x24_h = 0xffff, should be 0x0000
.
[ 2194.247905] eno1: ilr = 0xff, should be 0x04
.
[ 2194.247908] eno1: resv_0x2c_l = 0xffff, should be 0x17aa
.
[ 2194.247912] eno1: resv_0x2c_h = 0xffff, should be 0x38f7
.
[ 2194.248038] eno1: pci_sn_l = 0xffffffff, should be 0x684ce000
.
[ 2194.249247] eno1: pci_sn_h = 0xffffffff, should be 0x01000000
.
我猜这与网络管理器有关,并且
[ 2153.504278] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out!
我可以做什么来解决这个问题?
答案1
那个错误(“[drm:amdgpu_dm_atomic_commit_tail [amdgpu]]错误等待栅栏超时!”)已报告给 amdgpu 开发人员:https://gitlab.freedesktop.org/drm/amd/-/issues/934