On Wed, May 29, 2019 at 05:54:22PM +0200, Alois Schlögl wrote:
> On 3/26/19 9:03 PM, Romain Perier wrote:
> > On Wed, Mar 20, 2019 at 08:24:33AM +0100, Alois Schlögl wrote:
> >> On 3/18/19 7:46 PM, Romain Perier wrote:
> >>> On Mon, Mar 18, 2019 at 12:43:10PM +0100, Alois Schlögl wrote:
> >>>> On 3/18/19 12:20 PM, Romain Perier wrote:
> >>>>> Hello,
> >>>>>
> >>>>> On Mon, Mar 18, 2019 at 11:27:41AM +0100, Alois Schlögl wrote:
> >>>>>> Source: linux
> >>>>>> Severity: normal
> >>>>>>
> >>>>>> Dear Maintainer,
> >>>>>>
> >>>>>>    On a Lenovo L480 laptop, I've upgraded Debian from 9 (stretch) to 10
> >>>>>> (testing).
> >>>>>>    After the upgrade, the touchpad and the trackpoint was not usable
> >>>>>> anymore.
> >>>>>>
> >>>>>>
> >>>>>>    This already has some bug report here,
> >>>>>>    https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1803600
> >>>>>>
> >>>>>>    As a workaround, one can run the command,
> >>>>>>        sudo sh -c 'echo -n "elantech">
> >>>>>> /sys/bus/serio/devices/serio1/protocol'
> >>>>>>    in order to use the touchpad. However, on a GUI Interface and 
> >>>>>> without
> >>>>>>    an external mouse, it's impossible to apply this workaround
> >>>>>>   (switching to the terminal <CTRL>-<ALT>F1, login, and run the command
> >>>>>> above might work)
> >>>>>>
> >>>>>>    I expect to be able to use the touchpad just out of the box, not 
> >>>>>> needing
> >>>>>>    to run the above workaround
> >>>>>>
> >>>>> Could you :
> >>>>>
> >>>>> - Test with the last kernel uploaded to unstable (4.19.0-4:4.19.28) and 
> >>>>> confirm or
> >>>>>   not is the problem still exists ?
> >>>> Dear Romain
> >>>>
> >>>>
> >>>> I upgraded the kernel and rebooted:
> >>>>
> >>>> schloegl@debian10:~$ uname -a
> >>>> Linux debian10 4.19.0-4-amd64 #1 SMP Debian 4.19.28-2 (2019-03-15)
> >>>> x86_64 GNU/Linux
> >>>>
> >>>>
> >>>> With this kernel the trackpoint is working, the trackpad is still not
> >>>> usable.
> >>>>
> >>>> (This improves the situation because now at least one pointer device is
> >>>> available).
> >>>>
> >>>>
> >>> Good, we did some progress :)
> >>>
> >>>>> - According to the bug on launchpad and to the fix pushed upstream, the
> >>>>>   fix seems to be an hardware quirks, could you give me the output of 
> >>>>> the
> >>>>>   following command :
> >>>>>   $ /sys/bus/serio/devices/serio1/firmware_id
> >>>> root@debian10:~# cat /sys/bus/serio/devices/serio1/firmware_id
> >>>> PNP: LEN2036 PNP0f13
> >>>>
> >>> Could you test the patch attached to this reply ?
> >>> (if you don't know how to do this, I can provide support)
> >>>
> >>> Regards,
> >>> Romain
> >>
> >>
> >> I tried to followed these instructions:
> >>
> >> https://kernel-team.pages.debian.net/kernel-handbook/ch-comm
> >>
> >> 4.5. Building a custom kernel from Debian kernel source
> >>
> >> Specifically using the patched the sources,
> >>
> >> *scripts/config --disable MODULE_SIG*
> >> **scripts/config --disable DEBUG_INFO**
> >> ||*|make clean|* ||*|make deb-pkg
> >>
> >> |*
> >>
> >> and ended up with a kernel that does not boot (missing HD audio firmware),
> >>
> >>
> >> Which procedure do you recommend to build and install a modified kernel ?
> >>
> >>
> > Hi,
> >
> > Section 4.2 from
> > https://kernel-team.pages.debian.net/kernel-handbook/ch-common-tasks.html#s-common-official
> > , until test-patches should work. For the test-patches script, use the 
> > flavour and a
> > featureset as argument, when you invoke it, like this :
> >
> > # debian/bin/test-patches -f amd64 -s none 
> > /path/to/0001-Input-elantech-disable-elan-i2c-for-L480.patch
> >
> > This will apply the patch on the fly, configure the kernel for amd64
> > and build a version with a special changelog entry and a special suffix
> > version dedicated to the test version you generate.
> >
> >
> > In case of troubles, I can provide another way, from git with few
> > commands.
> >
> >
> > Hope this helps,
> > Regards,
> > Romain
> 
> 
> Dear Romain,
> 
> 
> your instructions to build the kernel worked fine, when trying to
> install the kernel,
> 
>    sudo dpkg -i linux-headers-4.19.0-5-amd64_4.19.37-3a~test_amd64.deb 
> linux-image-4.19.0-5-amd64-unsigned_4.19.37-3a~test_amd64.deb
> 
> I run into problem, getting this warning. 
> 
> 
>  │ You are running a kernel (version 4.19.0-5-amd64) and attempting to
> remove the same
> version.                                                                      
>                      
> │
>  │                                                                            
>                                                                               
>                               
> │
>  │ This can make the system unbootable as it will remove
> /boot/vmlinuz-4.19.0-5-amd64 and all modules under the directory
> /lib/modules/4.19.0-5-amd64. This can only be fixed with a copy  │
>  │ of the kernel image and the corresponding
> modules.                                                                      
>                                                                
> │
>  │                                                                            
>                                                                               
>                               
> │
>  │ It is highly recommended to abort the kernel removal unless you are
> prepared to fix the system after
> removal.                                                                      
>     
> │
>  │                                                                            
>                                                                               
>                               
> │
>  │ Abort kernel removal?   
> 
> 
> I'm not sure if I'm "prepared to fix the system". Can you recommend a
> reasonable save way to go forward ?
> 
> 
> Cheers,
> 
>    Alois

Hello,

Well, this is something I have tested here myself, from the linux
git repository (on salsa.debian.org). I have built a 4.19.37-4a~test
with the patch , then I have forced the install with the same question
than you. And he did the trick !

So what you can do is:

 - When the dialog interface (the blue one) asks you to abort or continue the 
install,
   press "no" to don't abort and continue the install
 - Once done, you can reboot
 - Check that the boot is working fine and you're running the intended
   kernel:  4.19.37-3a~test  (via uname -a)
 - Check if your problem is fixed

 - Once you want to re-use the debian kernel, you can :

   1. $ sudo apt remove linux-image-4.19.0-5-amd64-unsigned 
linux-headers-4.19.0-5-amd64

   ==> you will get a conflict , that's normal, you will be adviced to use
   "apt --fix-broken install", so please use it.

   2. It should reinstall the official debian kernel package. Then you
   can reboot to the official kernel image and it should do the trick


Everything has been tested here from git, the only difference is that I
built a kernel 4.19.37-4a~test  but the steps should be the same.

Hopes this help,
Regards,
Romain

Attachment: signature.asc
Description: PGP signature

Reply via email to