trouble with amdgpu and latest tumbleweed kernel

golson765 · December 7, 2018, 5:37pm

background: When I upgraded tumbleweed last week to the 4.19 kernel, my
system locked up. I could only get in by setting nomodeset, and in that
I could not get to the graphical target.

So now, several hours of troubleshooting later, I ended up deleting all
my old kernels and installing the 4.12 kernel from the Leap 15
repository (just to get an older kernel) on my tumbleweed partition.

I can boot now, but it fails to launch the graphical target. I tried to
run the command ‘startx’, but it would not launch, said something about
‘no screens found’.

Here is my hardware information


# lspci | grep VGA
00:01.0 VGA compatible controller: Advanced Micro Devices, Inc.
[AMD/ATI] Carrizo (rev cc)

Also, here is my Xorg.0.log:


    31.876] (--) Log file renamed from "/var/log/Xorg.pid-3846.log" to
"/var/log/Xorg.0.log"
    31.877]
X.Org X Server 1.20.3
X Protocol Version 11, Revision 0
    31.877] Build Operating System: openSUSE SUSE LINUX
    31.877] Current Operating System: Linux tribetrekDellbig
4.12.14-lp150.12.25-default #1 SMP Thu Nov 1 06:14:23 UTC 2018 (3fcf457)
x86_64
    31.877] Kernel command line:
BOOT_IMAGE=/boot/vmlinuz-4.12.14-lp150.12.25-default
root=UUID=04b9a67d-4291-4135-a0da-d227b8d08fa9
resume=/dev/disk/by-uuid/78e12fe1-1805-475f-bd43-ed40c991fc6e
splash=silent quiet showopts
    31.877] Build Date: 22 November 2018  12:00:00PM
    31.877]
    31.877] Current version of pixman: 0.34.0
    31.877] 	Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
    31.877] Markers: (--) probed, (**) from config file, (==) default
setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
    31.877] (==) Log file: "/var/log/Xorg.0.log", Time: Fri Dec  7
10:24:26 2018
    31.877] (==) Using config directory: "/etc/X11/xorg.conf.d"
    31.877] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
    31.879] Parse error on line 1 of section Device in file
/etc/X11/xorg.conf.d/50-elotouch.conf
"Section" is not a valid keyword in this section.
    31.879] (EE) Problem parsing the config file
    31.879] (EE) Error parsing the config file
    31.879] (EE)
Fatal server error:
    31.879] (EE) no screens found(EE)
    31.879] (EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
    31.879] (EE) Please also check the log file at
"/var/log/Xorg.0.log" for additional information.
    31.879] (EE)
    31.880] (EE) Server terminated with error (1). Closing log file.

Oh yes, also, I am running the amdgpu driver (not amdgpu-pro). I tried
to run my system with the radeon driver when the 4.19 kernel was the
problem, but it still would not hit the graphical target. I haven’t
tried using the radeon driver with the 4.12 kernel.

Any help would be really appreciated!!

malcolmlewis · December 7, 2018, 7:43pm

Hi
So this is a dual graphics setup? Is it an AMD APU or discrete cards?


lscpu

/sbin/lspci -nnk | egrep -A3 "VGA|Display|3D"

My laptop has Carrizo and Topaz XT gpu’s, I’ve not suffered any lockups, but did have a bit of screen tearing at 60Hz on the laptop, not when connected over hdmi to external monitor. To resolve that I added amdgpu.dc=0 to the boot options.

The other thing I did was add switcheroo-control (and run the service to add the dgpu option) else running with DRI_PRIME=1 also works fine. Also vulkan and openCL work.

For disk I also changed the i/o scheduler with the kernel boot option scsi_mod.use_blk_mq=1

Not sure about the older kernel, as long as the radeon module is blacklisted and you run mkinitrd (that leap kernel may still have it in initrd).

I would set the system to runlevel 3;


systemctl set-default multi-user.target

When it comes up to the console, switch to root user;


systemctl isolate graphical.target

See how that goes to get to a graphical target, then take it from there.

golson765 · December 9, 2018, 12:43am

On 12/7/18 12:46 PM, malcolmlewis wrote:
>
> Hi
> So this is a dual graphics setup? Is it an AMD APU or discrete cards?
>
>
> Code:
> --------------------
>
> lscpu
>
> /sbin/lspci -nnk | egrep -A3 “VGA|Display|3D”
>
> --------------------

Thanks for your help. Here are my outputs:


# lscpu
Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              4
On-line CPU(s) list: 0-3
Thread(s) per core:  2
Core(s) per socket:  2
Socket(s):           1
NUMA node(s):        1
Vendor ID:           AuthenticAMD
CPU family:          21
Model:               101
Model name:          AMD FX-9830P RADEON R7, 12 COMPUTE CORES 4C+8G
Stepping:            1
CPU MHz:             1800.000
CPU max MHz:         3000.0000
CPU min MHz:         1400.0000
BogoMIPS:            5989.24
Virtualization:      AMD-V
L1d cache:           32K
L1i cache:           96K
L2 cache:            1024K
NUMA node0 CPU(s):   0-3
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr
pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext
fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good acc_power nopl
nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma
cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm
cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch
osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core
perfctr_nb bpext ptsc mwaitx cpb hw_pstate ssbd vmmcall fsgsbase bmi1
avx2 smep bmi2 xsaveopt arat npt lbrv svm_lock nrip_save tsc_scale
vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic
v_vmsave_vmload vgif overflow_recov

# lspci -nnk | egrep -A3 "VGA|Display|3D"
00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc.
[AMD/ATI] Carrizo [1002:9874] (rev cc)
Subsystem: Dell Device [1028:07e2]
Kernel driver in use: amdgpu
Kernel modules: amdgpu
--
03:00.0 Display controller [0380]: Advanced Micro Devices, Inc.
[AMD/ATI] Baffin [Radeon RX 460/560D / Pro 450/455/460/560] [1002:67ef]
(rev ff)
Kernel driver in use: amdgpu
Kernel modules: amdgpu

So I added the boot options amdgpu.dc=0 and scsi_mod.use_blk_mq=1 to grub.

But I still cannot boot using kernel 4.19 without nomodeset, even if I
only boot to runlevel 3. It locks up. I also get an error there that
says something like this before it locks up:


[drm:amdgpu_device_ip_late_init_func_handler[amdgpu]] *ERROR* ib ring
test failed (-110).

There are a few other lines indicating a similar error.

I haven't tried installing switcheroo yet. I have been assuming I need
to at least be able to get to a functional graphical target with
networking before that, but if there is a way to apply changes from that
package as it is booting up, please let me know.

--
Tribe
Box:       42.3 | KDE Plasma 5 | AMD Phenom IIX4 | 64 | 32GB
Laptop #1: TW   | Plasma 5     | AMD FX 7TH GEN  | 64 | 32GB
Laptop #2: TW   | Plasma 5     | Core i5         | 64 | 8GB

golson765 · December 9, 2018, 12:45am

On 12/7/18 12:46 PM, malcolmlewis wrote:
>
> Hi
> So this is a dual graphics setup? Is it an AMD APU or discrete cards?
>
>
I believe it is a dual graphics setup with an AMD APU. It is on my
laptop, laptop #1 below. A Dell Inspiron 15 5000 gaming laptop.

mrmazda · December 9, 2018, 12:54am

Were all those old kernels 4.19.x? Were you able to run X with 4.18.x? If yes to the latter, and 64bit, reinstall it: http://fm.no-ip.com/Tmp/SUSE/Factory/kernel-default-4.18.15-1.2.x86_64.rpm If 32bit let me know and I’ll put i586 default up.

malcolmlewis · December 9, 2018, 1:47am

On Sat 08 Dec 2018 11:43:42 PM CST, SChinaSeaTribe wrote:

On 12/7/18 12:46 PM, malcolmlewis wrote:
>
> Hi
> So this is a dual graphics setup? Is it an AMD APU or discrete cards?
>
>
> Code:
> --------------------
>
> lscpu
>
> /sbin/lspci -nnk | egrep -A3 “VGA|Display|3D”
>
> --------------------

Thanks for your help. Here are my outputs:

Model name: AMD FX-9830P RADEON R7, 12 COMPUTE CORES 4C+8G Stepping: 1

lspci -nnk | egrep -A3 “VGA|Display|3D”

00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc.
[AMD/ATI] Carrizo [1002:9874] (rev cc)
Subsystem: Dell Device [1028:07e2]
Kernel driver in use: amdgpu
Kernel modules: amdgpu

03:00.0 Display controller [0380]: Advanced Micro Devices, Inc.
[AMD/ATI] Baffin [Radeon RX 460/560D / Pro 450/455/460/560] [1002:67ef]
(rev ff)
Kernel driver in use: amdgpu
Kernel modules: amdgpu

So I added the boot options amdgpu.dc=0 and scsi_mod.use_blk_mq=1 to
grub.

But I still cannot boot using kernel 4.19 without nomodeset, even if I
only boot to runlevel 3. It locks up. I also get an error there that
says something like this before it locks up:

[drm:amdgpu_device_ip_late_init_func_handler[amdgpu]] ERROR ib ring
test failed (-110).

<snip>

Hi
Can you use amdgpu.dpm=0 instead and see how that goes.

–
Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890)
SLES 15 | GNOME Shell 3.26.2 | 4.12.14-25.25-default
If you find this post helpful and are logged into the web interface,
please show your appreciation and click on the star below… Thanks!

golson765 · December 11, 2018, 4:15am

On 12/8/18 6:47 PM, malcolmlewis wrote:

>

> Hi
> Can you use amdgpu.dpm=0 instead and see how that goes.
>

Awesome, that did it. I have my desktop back. Just got in, so I will see
how the performance is, but everything so far seems to be working, and I
am on the amdgpu driver.

golson765 · December 11, 2018, 8:18pm

On 12/10/18 9:15 PM, SChinaSeaTribe wrote:
> On 12/8/18 6:47 PM, malcolmlewis wrote:
>

>>
>
>> Hi
>> Can you use amdgpu.dpm=0 instead and see how that goes.
>>
>

>
> Awesome, that did it. I have my desktop back. Just got in, so I will see
> how the performance is, but everything so far seems to be working, and I
> am on the amdgpu driver.
>
>

My laptop has Carrizo and Topaz XT gpu’s, I’ve not suffered any lockups,
but did have a bit of screen tearing at 60Hz on the laptop, not when
connected over hdmi to external monitor. To resolve that I added
amdgpu.dc=0 to the boot options.

So, things are working well now, but in reference to your earlier
statement, I also do have some “screen tearing”, if that is what it is.

Basically I have lines flashing, mostly on the right side of my screen,
intermittently.

Any ideas on how to get rid of this?

malcolmlewis · December 11, 2018, 8:28pm

Hi
Yup, same thing I saw, add amdgpu.dc=0 to your boot options I’m connected to an external screen which isn’t affected.

golson765 · December 12, 2018, 6:51pm

On 12/11/18 1:36 PM, malcolmlewis wrote:

>>
>> So, things are working well now, but in reference to your earlier
>> statement, I also do have some “screen tearing”, if that is what it is.
>>
>> Basically I have lines flashing, mostly on the right side of my screen,
>> intermittently.
>>
>> Any ideas on how to get rid of this?
>>
> Hi
> Yup, same thing I saw, add amdgpu.dc=0 to your boot options I’m
> connected to an external screen which isn’t affected.
>

Ok, just want to be sure… so I already have amdgpu.dpm=0 as one of my
boot options in grub. Do I now add amdgpu.dc=0 as well, so that both of
those boot options are in the grub command line?

malcolmlewis · December 12, 2018, 7:10pm

On Wed 12 Dec 2018 05:51:33 PM CST, SChinaSeaTribe wrote:

On 12/11/18 1:36 PM, malcolmlewis wrote:

[QUOTE]
>>
>> So, things are working well now, but in reference to your earlier
>> statement, I also do have some “screen tearing”, if that is what it
>> is.
>>
>> Basically I have lines flashing, mostly on the right side of my
>> screen, intermittently.
>>
>> Any ideas on how to get rid of this?
>>
> Hi
> Yup, same thing I saw, add amdgpu.dc=0 to your boot options I’m
> connected to an external screen which isn’t affected.
>

Ok, just want to be sure… so I already have amdgpu.dpm=0 as one of my
boot options in grub. Do I now add amdgpu.dc=0 as well, so that both of
those boot options are in the grub command line?

[/QUOTE]
Hi
Yes, your correct both options

–
Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890)
SLES 15 | GNOME Shell 3.26.2 | 4.12.14-25.25-default
If you find this post helpful and are logged into the web interface,
please show your appreciation and click on the star below… Thanks!

trouble with amdgpu and latest tumbleweed kernel

lspci -nnk | egrep -A3 “VGA|Display|3D”

00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Carrizo [1002:9874] (rev cc) Subsystem: Dell Device [1028:07e2] Kernel driver in use: amdgpu Kernel modules: amdgpu

00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc.
[AMD/ATI] Carrizo [1002:9874] (rev cc)
Subsystem: Dell Device [1028:07e2]
Kernel driver in use: amdgpu
Kernel modules: amdgpu