Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #273189 > unrolled thread

Re: old kernel appears more stable than latest

Back to article view | Back to linux.debian.user

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: old kernel appears more stable than latest Jeffrey Walton <noloader@gmail.com> - 2024-09-21 14:30 +0200
    Re: old kernel appears more stable than latest Jeffrey Walton <noloader@gmail.com> - 2024-09-25 15:00 +0200

#273189 — Re: old kernel appears more stable than latest

On Fri, Sep 13, 2024 at 11:33 PM hlyg <hlyg2023@outlook.com> wrote:
>
> i have installed latest 12.7, after running about 10 hours, it doesn't
> respond to my keyboard pressing, power LED on front panel of pc case
> become red, i have to reboot
>
> i run journalctl, i am not sure if msg below are related to my problem
>
> Sep 13 18:06:54 debian kernel: INFO: task RTW_CMD_THREAD:561 blocked for
> more than 604 seconds.
> Sep 13 18:06:54 debian kernel:       Tainted: G         C 6.1.0-25-amd64
> #1 Debian 6.1.106-3
> Sep 13 18:06:54 debian kernel: "echo 0 >
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> Sep 13 18:06:54 debian kernel: task:RTW_CMD_THREAD  state:D stack:0
> pid:561   ppid:2      flags:0x00004000
> Sep 13 18:06:54 debian kernel: Call Trace:
> Sep 13 18:06:54 debian kernel:  <TASK>
> Sep 13 18:06:54 debian kernel:  __schedule+0x34d/0x9e0
> Sep 13 18:06:54 debian kernel:  schedule+0x5a/0xd0
> Sep 13 18:06:54 debian kernel:  schedule_timeout+0x118/0x150
> Sep 13 18:06:54 debian kernel:  wait_for_completion+0x86/0x160
> Sep 13 18:06:54 debian kernel:  ?
> rtw_setassocsta_cmdrsp_callback+0x80/0x80 [r8188eu]
> Sep 13 18:06:54 debian kernel:  rtw_cmd_thread+0x3f/0x180 [r8188eu]
> Sep 13 18:06:54 debian kernel:  kthread+0xda/0x100
> Sep 13 18:06:54 debian kernel:  ? kthread_complete_and_exit+0x20/0x20
> Sep 13 18:06:54 debian kernel:  ret_from_fork+0x22/0x30
> Sep 13 18:06:54 debian kernel:  </TASK>
>
> 12.5 which use vmlinuz-6.1.0-18-amd64 appears more stable
> 12.7 use vmlinuz-6.1.0-25-amd64

When I experience weird power related problems (and problems with
reboot), I go right to the BIOS or UEFI and ensure the firmware is the
latest. UEFI is where the ACPI tables are located, and that affects
all power behavior. So I think your first step is to ensure the
firmware is up to date.

I also disable S3 and S4 sleep states in the firmware. In this case, I
don't want both the firmware and OS to try to control power states.

Jeff

[toc] | [next] | [standalone]

#273272

On Tue, Sep 24, 2024 at 10:38 PM hlyg <hlyg2023@outlook.com> wrote:
>
> it seems to be caused by wrong usb configuration
>
> kernel msg during boot: "usb 1-3: device descriptor read/64, error -71"
>
> if usb hub is connected to wrong usb port, after running about 12 hours,
> 12.7 fails while 12.5 is OK
>
> connecting to correct usb port can fix it
>
> i get clue from freebsd, its installer refuse to work if wrongly connected

That is a very interesting finding. I thought the external hubs would
auto-configure to the highest version provided by the host. I would
not expect the machine to hang or crash in that particular
configuration. Do you think it is due to cheap hardware?

I should read the USB spec to try to understand what is happening, but
I am too lazy at the moment...

Jeff

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web