problems with linux kernel 4-19 on HP notebook 255 G6

If none of the more specific forums is the right place to ask

problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-12 14:55

Hi, I have a HP notebook 255 G6 with debian buster (Linux horsepower 4.19.0-1-amd64 #1 SMP Debian 4.19.12-1 (2018-12-22) x86_64 GNU/Linux).
Since 20 December (when I rebooted) I'm having varius problem, and one of the most painfull is that:
if screen goes black due to inactivity, after that the windows are no more updated (I can move the mouse and use the gui but to see the effects I have to do ctrl-alt-F2 and back to ctrl-alt-F1 with updated gui); also problem with sleep where I'm unable to awake the computer and need to force reboot pressing power button.

Here some additional info:
Code: Select all
sudo lspci
...
00:01.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Stoney [Radeon R2/R3/R4/R5 Graphics] (rev d4)
...
[\code]
and  some lines from xorg.0.log:
[code]
[   225.689] (II) modesetting: Driver for Modesetting Kernel Drivers: kms
[   225.689] (II) FBDEV: driver for framebuffer: fbdev
[   225.689] (II) VESA: driver for VESA chipsets: vesa
[   225.689] (EE) open /dev/dri/card0: No such file or directory
[   225.689] (WW) Falling back to old probe method for modesetting
[   225.689] (EE) open /dev/dri/card0: No such file or directory
[   225.689] (II) Loading sub module "fbdevhw"
[   225.689] (II) LoadModule: "fbdevhw"
[   225.689] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[   225.689] (II) Module fbdevhw: vendor="X.Org Foundation"
[   225.689]    compiled for 1.20.3, module version = 0.0.2
[   225.689]    ABI class: X.Org Video Driver, version 24.0
[   225.689] (EE) Unable to find a valid framebuffer device
[   225.689] (WW) Falling back to old probe method for fbdev
[   225.689] (II) Loading sub module "fbdevhw"
[   225.689] (II) LoadModule: "fbdevhw"
[   225.690] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so
[   225.690] (II) Module fbdevhw: vendor="X.Org Foundation"
[   225.690]    compiled for 1.20.3, module version = 0.0.2
[   225.690]    ABI class: X.Org Video Driver, version 24.0
[   225.690] (II) FBDEV(2): using default device
[   225.690] (EE) Screen 0 deleted because of no matching config section.
[   225.690] (II) UnloadModule: "modesetting"
[   225.690] (EE) Screen 0 deleted because of no matching config section.
[   225.690] (II) UnloadModule: "fbdev"
[   225.690] (II) UnloadSubModule: "fbdevhw"

With older linux kernel 4.18.0-3-amd64 #1 (SMP Debian 4.18.20-2 (2018-11-23) x86_64 GNU/Linux) no problem with sleep awake and screen blanking. Anyone has ideas of what's going on?
Thankyou for attention and best regards.
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby Head_on_a_Stick » 2019-01-12 15:19

alexanderbeerhoff wrote:Since 20 December (when I rebooted)

There has been a version bump since then in buster (2018-12-22), try updating your system.

This sounds like a regression but it may have been fixed already.
User avatar
Head_on_a_Stick
 
Posts: 8900
Joined: 2014-06-01 17:46
Location: /dev/chair

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-12 19:42

running apt update/upgrade daily: rebooted again with linux kernel 4.19 but the problems still there.
Lukily with 4.18 I don't have above problems
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52


Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-19 11:21

Log of job done:
Code: Select all
git bisect start
# good: [94710cac0ef4ee177a63b5227664b38c95bbf703] Linux 4.18
git bisect good 94710cac0ef4ee177a63b5227664b38c95bbf703
# bad: [84df9525b0c27f3ebc2ebb1864fa62a97fdedb7d] Linux 4.19
git bisect bad 84df9525b0c27f3ebc2ebb1864fa62a97fdedb7d
# bad: [db06f826ec12bf0701ea7fc0a3c0aa00b84417c8] Merge tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
git bisect bad db06f826ec12bf0701ea7fc0a3c0aa00b84417c8
# good: [0a957467c5fd46142bc9c52758ffc552d4c5e2f7] x86: i8259: Add missing include file
git bisect good 0a957467c5fd46142bc9c52758ffc552d4c5e2f7
# good: [9a76aba02a37718242d7cdc294f0a3901928aa57] Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
git bisect good 9a76aba02a37718242d7cdc294f0a3901928aa57
# bad: [ef8e0ff97ae8168ffe1558a5726a8b348c8228a3] Merge tag 'drm-intel-next-2018-07-19' of git://anongit.freedesktop.org/drm/drm-intel into drm-next
git bisect bad ef8e0ff97ae8168ffe1558a5726a8b348c8228a3
# bad: [e0d018119ae82cbde32c1d4f8e9b8d8f43a3c88a] drm/v3d: Remove unnecessary dma_fence_ops.
git bisect bad e0d018119ae82cbde32c1d4f8e9b8d8f43a3c88a
# good: [e1cacec9d50d7299893eeab2d895189f3db625da] drm/i915: Update DRIVER_DATE to 20180620
git bisect good e1cacec9d50d7299893eeab2d895189f3db625da
# good: [57e23de02f4878061818fd118129a6b0e1516b11] drm/sun4i: DW HDMI: Expand algorithm for possible crtcs
git bisect good 57e23de02f4878061818fd118129a6b0e1516b11
# bad: [6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a] drm/amd/display: Write TEST_EDID_CHECKSUM_WRITE for EDID tests
git bisect bad 6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a

I had some troubles building linux-4.17.0-rc5-01400-gdaf0678c2036, cloning clean tree from torvalds repo, and couple of builds
Code: Select all
git bisect start
# good: [57e23de02f4878061818fd118129a6b0e1516b11] drm/sun4i: DW HDMI: Expand algorithm for possible crtcs
git bisect good 57e23de02f4878061818fd118129a6b0e1516b11
# bad: [6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a] drm/amd/display: Write TEST_EDID_CHECKSUM_WRITE for EDID tests
git bisect bad 6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a
# good: [daf0678c2036c918f01e4aa6035629d2debc2f30] Merge branch 'drm-next-4.18' of git://people.freedesktop.org/~agd5f/linux into drm-next
git bisect good daf0678c2036c918f01e4aa6035629d2debc2f30

then again error 2 while building (too few argument building something related to amdgpu: if this error persist I'll save output to logfile) so trying with clean tree again.
Any suggestion to avoid this build error?
Latest git bisect log
Code: Select all
git bisect start
# good: [57e23de02f4878061818fd118129a6b0e1516b11] drm/sun4i: DW HDMI: Expand algorithm for possible crtcs
git bisect good 57e23de02f4878061818fd118129a6b0e1516b11
# bad: [6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a] drm/amd/display: Write TEST_EDID_CHECKSUM_WRITE for EDID tests
git bisect bad 6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a
# good: [daf0678c2036c918f01e4aa6035629d2debc2f30] Merge branch 'drm-next-4.18' of git://people.freedesktop.org/~agd5f/linux into drm-next
git bisect good daf0678c2036c918f01e4aa6035629d2debc2f30

Back to building errors: the first error seems to be that
Code: Select all
CC      net/core/flow_dissector.o
lib/.tmp_nlattr.o: warning: objtool: validate_nla()+0xce: sibling call from callable instruction with modified stack frame
lib/.tmp_nlattr.o: warning: objtool: nla_parse()+0x99: sibling call from callable instruction with modified stack frame
  CC [M]  fs/nfs/nfs2super.o
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c: In function ‘amdgpu_ctx_mgr_entity_fini’:
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c:466:5: error: too few arguments to function ‘drm_sched_entity_do_release’
     drm_sched_entity_do_release(&ctx->adev->rings[i]->sched,
     ^~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from drivers/gpu/drm/amd/amdgpu/amdgpu.h:48,
                 from drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c:27:
./include/drm/gpu_scheduler.h:287:6: note: declared here
 long drm_sched_entity_do_release(struct drm_gpu_scheduler *sched,
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~
make[7]: *** [scripts/Makefile.build:312: drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.o] Error 1
make[6]: *** [scripts/Makefile.build:559: drivers/gpu/drm/amd/amdgpu] Error 2
make[5]: *** [scripts/Makefile.build:559: drivers/gpu/drm] Error 2
make[4]: *** [scripts/Makefile.build:559: drivers/gpu] Error 2
make[4]: *** Attente des tâches non terminées....
  CC [M]  drivers/hid/hid-ntrig.o

seems happens only after one/two successfull bisection.
The commands I'm using are:
Code: Select all
cd src/linux
cp /boot/config-$(uname -r) .config
make oldconfig
sudo pico .config #(commenting CONFIG_SYSTEM_TRUSTED_KEYS for which anyway I'm asked at next step)
sudo make -j4 bindeb-pkg

Suggestion to shorten build time from approx 6hrs are the welcome?
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby Wheelerof4te » 2019-01-19 12:11

alexanderbeerhoff wrote:Suggestion to shorten build time from approx 6hrs are the welcome?

Plug in your phone, usb and camera and while in the build directory, run
Code: Select all
make localmodconfig

This will make a .config file only for your local configuration and shorten build time significantly.
You can also try other kernels, like the ones from kernel.org.
You can also wait for new kernel version from Sid. Linux 4.19.16 is already there.
User avatar
Wheelerof4te
 
Posts: 1164
Joined: 2015-08-30 20:14

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-19 13:25

That would have been the command! But I'm using .config file from my boot folder. Probably there is lots of stuff like driver that I can remove since seems not related to my problem's debug.
And from what git bisect says I'm only 5 steps from my goal so keep building.
I've also found this patch seems to resolve building error: https://patchwork.kernel.org/patch/10446791/
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-21 12:26

I think to have found the problem
Code: Select all
git bisect start
# bad: [6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a] drm/amd/display: Write TEST_EDID_CHECKSUM_WRITE for EDID tests
git bisect bad 6e0ef9d85b99baeeea4b9c4a9777809cb0c6040a
# good: [daf0678c2036c918f01e4aa6035629d2debc2f30] Merge branch 'drm-next-4.18' of git://people.freedesktop.org/~agd5f/linux into drm-next
git bisect good daf0678c2036c918f01e4aa6035629d2debc2f30
# bad: [741f01e636b72ff3f81204fd595ac1078907671b] drm/scheduler: Avoid using wait_event_killable for dying process (V4)
git bisect bad 741f01e636b72ff3f81204fd595ac1078907671b
# good: [4ea209073d229a0cb65f9978e83fa7c044026416] drm/amd/display: Dynamic HDR metadata mem buffer
git bisect good 4ea209073d229a0cb65f9978e83fa7c044026416
# good: [b646c1dc835b6b73884a88643c2534f1a4a1928f] drm/amdgpu: add kernel doc for memory domains.
git bisect good b646c1dc835b6b73884a88643c2534f1a4a1928f
# bad: [99fa7ce3a82bf8047b6104682da6ae276beb22d8] drm/doc: Add initial amdgpu driver documentation
git bisect bad 99fa7ce3a82bf8047b6104682da6ae276beb22d8
# bad: [d1a9146b3a98a139c79b79c7e1426f1128e2bbdf] drm/amdgpu/display: enable CONFIG_DRM_AMD_DC_DCN1_0 by default
git bisect bad d1a9146b3a98a139c79b79c7e1426f1128e2bbdf
# bad: [5099114ba3b2e5ae9fb487aeb3ae0434fe38a7da] drm/amdgpu/display: drop DRM_AMD_DC_FBC kconfig option
git bisect bad 5099114ba3b2e5ae9fb487aeb3ae0434fe38a7da
# first bad commit: [5099114ba3b2e5ae9fb487aeb3ae0434fe38a7da] drm/amdgpu/display: drop DRM_AMD_DC_FBC kconfig option
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby Wheelerof4te » 2019-01-21 17:34

If you think this is a bug, maybe you should report it.
https://www.debian.org/Bugs/Reporting
User avatar
Wheelerof4te
 
Posts: 1164
Joined: 2015-08-30 20:14

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby stevepusser » 2019-01-22 19:38

Since you are using an AMD card and the problems seem to be related to the driver, maybe some information about what driver is being used on 4.18 versus 4.19 would help (install inxi and give us "inxi -G" results)

Also,what do you have for
Code: Select all
apt policy firmware-amd-graphics
?
The MX Linux repositories: Backports galore! If we don't have something, just ask and we'll try--we like challenges. New packages: Flightgear 2018.2.2, 4.20.7 kernel, wine-staging 4.0.2, Strawberry 0.5.2, Calibre 3.39.1, Pale Moon 28.4.0
User avatar
stevepusser
 
Posts: 10535
Joined: 2009-10-06 05:53

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby alexanderbeerhoff » 2019-01-22 21:56

Here the output of the commands with 4.18 and 4.19 kernel running:
Code: Select all
uname -r
4.18.0-3-amd64
inxi -G
Graphics:
  Device-1: AMD Stoney [Radeon R2/R3/R4/R5 Graphics] driver: amdgpu
  v: kernel
  Display: tty server: X.Org 1.20.3 driver: ati,radeon,vesa
  unloaded: fbdev,modesetting resolution: 1366x768~60Hz
  OpenGL: renderer: AMD STONEY (DRM 3.26.0 4.18.0-3-amd64 LLVM 7.0.1)
  v: 4.5 Mesa 18.2.8
apt policy firmware-amd-graphics
firmware-amd-graphics:
  Installé : 20190114-1
  Candidat : 20190114-1
 Table de version :
 *** 20190114-1 500
        500 http://ftp.it.debian.org/debian testing/non-free amd64 Packages
        100 /var/lib/dpkg/status

uname -r
4.19.0-1-amd64
inxi -G
Graphics:  Device-1: Advanced Micro Devices [AMD/ATI] Stoney [Radeon R2/R3/R4/R5 Graphics] driver: amdgpu v: kernel
           Display: tty server: X.Org 1.20.3 driver: ati,radeon,vesa unloaded: fbdev,modesetting resolution: 1366x768~60Hz
           OpenGL: renderer: AMD STONEY (DRM 3.27.0 4.19.0-1-amd64 LLVM 7.0.1) v: 4.5 Mesa 18.2.8
apt policy firmware-amd-graphics
firmware-amd-graphics:
  Installé : 20190114-1
  Candidat : 20190114-1
 Table de version :
 *** 20190114-1 500
        500 http://ftp.it.debian.org/debian testing/non-free amd64 Packages
        100 /var/lib/dpkg/status
alexanderbeerhoff
 
Posts: 8
Joined: 2017-09-05 11:52

Re: problems with linux kernel 4-19 on HP notebook 255 G6

Postby stevepusser » 2019-01-23 23:38

It's using the same driver with both kernels, so I agree it seems to be a regression in the 4.19 kernel's code.
The MX Linux repositories: Backports galore! If we don't have something, just ask and we'll try--we like challenges. New packages: Flightgear 2018.2.2, 4.20.7 kernel, wine-staging 4.0.2, Strawberry 0.5.2, Calibre 3.39.1, Pale Moon 28.4.0
User avatar
stevepusser
 
Posts: 10535
Joined: 2009-10-06 05:53


Return to General Questions

Who is online

Users browsing this forum: No registered users and 9 guests

fashionable