amd issueshttps://gitlab.freedesktop.org/drm/amd/-/issues2019-11-19T20:27:00Zhttps://gitlab.freedesktop.org/drm/amd/-/issues/145RX 480 sclk clock speed lowers when under load2019-11-19T20:27:00ZBugzilla Migration UserRX 480 sclk clock speed lowers when under load## Submitted by Christoph Haag `@ChristophHaag`
Assigned to **Default DRI bug account**
**[Link to original bug (#99967)](https://bugs.freedesktop.org/show_bug.cgi?id=99967)**
## Description
Created attachment 129917<br>
screensho...## Submitted by Christoph Haag `@ChristophHaag`
Assigned to **Default DRI bug account**
**[Link to original bug (#99967)](https://bugs.freedesktop.org/show_bug.cgi?id=99967)**
## Description
Created attachment 129917<br>
screenshot with echo manual > power_dpm_force_performance_level<br>
<br>
My GPU is the factory overclocked XFX Radeon RX 480 XXX OC, this might have something to do with it.<br>
<br>
I noticed this on agd5d/drm-next-4.12-wip and found similar behavior on 4.10, 4.9 and 4.8.<br>
<br>
It's even affected even with echo manual > /sys/class/drm/card0/device/power_dpm_force_performance_level<br>
<br>
I've made a screenshot of radeon-profile on linux 4.8.14 and annotated it.<br>
<br>
Steps I took:<br>
echo manual > /sys/class/drm/card0/device/power_dpm_force_performance_level<br>
echo 7 > /sys/class/drm/card0/device/pp_dpm_sclk<br>
then started unigine-heaven, closed it, waited a couple of seconds, started furmark, closed it, waited a couple of seconds and then ran another echo 7 > /sys/class/drm/card0/device/pp_dpm_sclk<br>
<br>
When the GPU is "relatively" idle (just displaying the desktop with a browser, etc.) this doesn't happen. I can echo any number in /sys/class/drm/card0/device/pp_dpm_sclk and it sticks.<br>
But when I start applications that put some stress on the GPU, the sclk clock lowers - and stays on a lower level even when the load is gone, until another number is echo'ed into /sys/class/drm/card0/device/pp_dpm_sclk<br>
<br>
**Attachment 129917**, "screenshot with echo manual > power_dpm_force_performance_level": <br>
![manual-annotated](/uploads/b09fcdcfad000939172b8d561dca29c6/manual-annotated.png)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/142linux-firmware 2017-02-17 update causes varying breaks in AMDGPU for recent c...2023-05-25T05:44:40ZBugzilla Migration Userlinux-firmware 2017-02-17 update causes varying breaks in AMDGPU for recent cards## Submitted by sau..@..ht.edu
Assigned to **Default DRI bug account**
**[Link to original bug (#99907)](https://bugs.freedesktop.org/show_bug.cgi?id=99907)**
## Description
Currently on Arch Linux after they shipped an update to ...## Submitted by sau..@..ht.edu
Assigned to **Default DRI bug account**
**[Link to original bug (#99907)](https://bugs.freedesktop.org/show_bug.cgi?id=99907)**
## Description
Currently on Arch Linux after they shipped an update to the linux-firmware set (20170217.12987ca-1), there's been reports of various issues ranging from power management failing to in my case (AMD Radeon RX 460) Xorg failing to work at all (it either blinks and goes back to a frozen VT as the GPU hangs, or the GPU hangs on a full-screen corruption of some kind.) This is broken on both kernel 4.9.11 and 4.10 in my testing, on Xorg 1.19.1. The system still responds to SSH connections, but fails to shutdown properly if attempted over that. Tracker link: https://bugs.archlinux.org/task/53042<br>
<br>
I've traced it back to a specific commit to linux-firmware, 7a110b85a46d7f884f4ac712ff52e02ed57234bd, https://git.kernel.org/cgit/linux/kernel/git/firmware/linux-firmware.git/commit/amdgpu/carrizo_ce.bin?id=7a110b85a46d7f884f4ac712ff52e02ed57234bd, pushed to the git repo on 2-17-17, which updates a large subset of the firmware images used by AMDGPU. Seeing as how this is a binary file set, I'm really not sure how to proceed from here in testing it to give any more useful information here. <br>
<br>
Apologies if this is the wrong place to report a firmware issue, but I was unsure where to file it otherwise.https://gitlab.freedesktop.org/drm/amd/-/issues/141Power management problems & kernel hangs with Cap Verde2023-05-25T05:44:42ZBugzilla Migration UserPower management problems & kernel hangs with Cap Verde## Submitted by Klaus Kusche
Assigned to **Default DRI bug account**
**[Link to original bug (#99815)](https://bugs.freedesktop.org/show_bug.cgi?id=99815)**
## Description
I recently switched from the radeon kernel driver to the a...## Submitted by Klaus Kusche
Assigned to **Default DRI bug account**
**[Link to original bug (#99815)](https://bugs.freedesktop.org/show_bug.cgi?id=99815)**
## Description
I recently switched from the radeon kernel driver to the amdgpu kernel driver.<br>
As a consequence, battery lifetime of my laptop was more than halved<br>
(before I could give *two* 90 minutes lectures in a row running on battery, <br>
now the battery is empty after around 70 minutes),<br>
i.e. with amdgpu, the graphic card consumes a multiple of the power it needed<br>
with radeon.<br>
<br>
GPU is AMDGPU(0): Chipset: "VERDE" (ChipID = 0x6825)<br>
Laptop is Dell Precision M6700<br>
OS is linux 4.9, userland is running latest official releases (Gentoo)<br>
Typical load is two screens (internal + VGA-Beamer, both 1024x768 mirrored)<br>
with very light load (no 3D & no video at all, just static pdf slides).<br>
<br>
Power settings in battery mode with the old radeon driver were<br>
* no dpm <br>
(always was very unstable and consumed by far more power than "classic" pm)<br>
* "classic" pm with power_method = profile and power_profile = low<br>
<br>
Analysis of the problems with amdgpu showed:<br>
<br>
1.) There is no "classic" power management: If dpm is turned off with amdgpu,<br>
there is no power management at all, and the card permanently runs<br>
with high power. Why was "classic" power management dropped completely?<br>
<br>
Basically, I neither want nor need any dynamic adjustment:<br>
When on battery, the gpu should permanently run on minimal power,<br>
independent of its load.<br>
<br>
2.) The default dpm mode is "balanced" and "auto".<br>
"balanced" does not change to "battery" automatically when AC is removed,<br>
and it is impossible to set the mode to "battery" manually:<br>
Any attempt to set it via sysfs either kills the process doing so<br>
or hangs it in a kill-9-immune state (which prevents system shutdown).<br>
<br>
I then code-changed the default initialization in si_dpm.c from "balanced"<br>
to "battery". This hangs the whole kernel hard during early boot:<br>
No display, no reaction to Alt-Sysrq, ... only hard power off helps.<br>
<br>
3.) It is possible to set the dpm level from "auto" to "low",<br>
but that does not seem to result in any measurable effect or power savings?https://gitlab.freedesktop.org/drm/amd/-/issues/134Rx480 doesn't output properly onto z27q at 5120x28802023-05-25T05:45:10ZBugzilla Migration UserRx480 doesn't output properly onto z27q at 5120x2880## Submitted by Matthew Treinish
Assigned to **Default DRI bug account**
**[Link to original bug (#99801)](https://bugs.freedesktop.org/show_bug.cgi?id=99801)**
## Description
Created attachment 129569<br>
Example output when X fi...## Submitted by Matthew Treinish
Assigned to **Default DRI bug account**
**[Link to original bug (#99801)](https://bugs.freedesktop.org/show_bug.cgi?id=99801)**
## Description
Created attachment 129569<br>
Example output when X first starts<br>
<br>
I have a a z27q monitor and an rx480 gpu I've just purchased. When I startx on the machine trying to run the monitor at it's native 5120x2880 across the 2 display ports doesn't work. The output is divided with the image mirrored across both halves of the screen. Then one half is itself split in 2 with a negative image that is scrolling horizontally. I took a picture with my cellphone, since a screenshot showed the proper image.<br>
<br>
I'm currently running the monitor over the single display port at a lower resolution of 3840x2160 (albeit with heavy flickering, which I think is a separate if not related issue)<br>
<br>
The current output of xrandr is:<br>
<br>
Screen 0: minimum 320 x 200, current 5040 x 2160, maximum 16384 x 16384<br>
DisplayPort-0 connected primary 3840x2160+1200+0 (normal left inverted right x axis y axis) 597mm x 336mm<br>
2560x2880 59.98 +<br>
2560x1440 59.95 +<br>
3840x2160 60.00* <br>
1920x1080 60.00 <br>
1600x900 60.00 <br>
1366x768 59.99 <br>
1280x720 60.00 59.94 <br>
1024x768 60.00 <br>
800x600 60.32 <br>
640x480 60.00 59.94 <br>
DisplayPort-1 connected (normal left inverted right x axis y axis)<br>
2560x2880 59.98 +<br>
DisplayPort-2 disconnected (normal left inverted right x axis y axis)<br>
HDMI-A-0 connected 1200x1920+0+0 left (normal left inverted right x axis y axis) 474mm x 296mm<br>
1920x1200 59.95*+<br>
1920x1080 60.00 <br>
1600x1200 60.00 <br>
1680x1050 59.88 <br>
1400x1050 59.95 <br>
1280x1024 75.02 60.02 <br>
1440x900 74.98 59.90 <br>
1024x768 75.03 70.07 60.00 <br>
800x600 72.19 75.00 60.32<br>
<br>
This might just be a configuration issue as I'm not used to the KMS way of doing things. (I previously had an nvidia gpu using their binary blob on this display without issue) But, from what I can tell things are being configured to do the right thing automatically and it is not rendered properly over the wire. (especially because when I tried to take a screenshot of the issue it looked like it was supposed to)<br>
<br>
I apologize that bug report lacks concrete details I'm willing to provide whatever information is necessary to debug this. I'm just not sure where to begin.<br>
<br>
**Attachment 129569**, "Example output when X first starts": <br>
![IMG_20170213_214640](/uploads/b7c15672593b89ac3a43aa020f41b710/IMG_20170213_214640.jpg)<br>
### See also
* https://bugs.freedesktop.org/show_bug.cgi?id=105651https://gitlab.freedesktop.org/drm/amd/-/issues/132Lenovo Thinkpad E450 (Kaveri A6-7000 Radeon R4) Blank Screen after suspend (d...2023-05-25T05:45:11ZBugzilla Migration UserLenovo Thinkpad E450 (Kaveri A6-7000 Radeon R4) Blank Screen after suspend (displayport link issues)## Submitted by Kyle Brooks
Assigned to **Default DRI bug account**
**[Link to original bug (#99779)](https://bugs.freedesktop.org/show_bug.cgi?id=99779)**
## Description
Created attachment 129527<br>
dmesg 4.9.8<br>
<br>
If I shu...## Submitted by Kyle Brooks
Assigned to **Default DRI bug account**
**[Link to original bug (#99779)](https://bugs.freedesktop.org/show_bug.cgi?id=99779)**
## Description
Created attachment 129527<br>
dmesg 4.9.8<br>
<br>
If I shut my laptop lit it suspends, then when opening the lid and resuming I get a blank screen and am unable to get the display to work again unless I reboot. I can ssh into computer and I get displayport link errors in the attached dmesg log.<br>
<br>
This issue does not occur if I use the radeon kernel module instead of amdgpu.<br>
<br>
This seems very similar to [bug 90320](https://bugs.freedesktop.org/show_bug.cgi?id=90320) but I filed this one because it is amdgpu not radeon:<br>
https://bugs.freedesktop.org/show_bug.cgi?id=90320<br>
<br>
I've also attached the Xorg.0.log but it does not have any errors indicated. It indicates successful exit because I stopped the server over ssh and tried to restart gdm a few times which is why there are multiple displayport link status failed messages in dmesg.<br>
<br>
**Attachment 129527**, "dmesg 4.9.8": <br>
[dmesg-4.9.8-1-ARCH.log](/uploads/a750bf717feee19158e804c8f2c78be9/dmesg-4.9.8-1-ARCH.log)<br>
### See also
* https://bugs.freedesktop.org/show_bug.cgi?id=90320https://gitlab.freedesktop.org/drm/amd/-/issues/129VM/GPU fault on 4.10-rc6 (Kaveri + Topaz)2023-05-25T05:45:12ZBugzilla Migration UserVM/GPU fault on 4.10-rc6 (Kaveri + Topaz)## Submitted by SET
Assigned to **Default DRI bug account**
**[Link to original bug (#99680)](https://bugs.freedesktop.org/show_bug.cgi?id=99680)**
## Description
Created attachment 129338<br>
VM and GPU faults<br>
<br>
dmesg repo...## Submitted by SET
Assigned to **Default DRI bug account**
**[Link to original bug (#99680)](https://bugs.freedesktop.org/show_bug.cgi?id=99680)**
## Description
Created attachment 129338<br>
VM and GPU faults<br>
<br>
dmesg reports VM and GPU faults with 4.10-rc6, as in the attachement. The X session remains usable. On very rare occasions, the laptop won't resume from suspend. Reported for any usefulness.<br>
<br>
**Attachment 129338**, "VM and GPU faults": <br>
[vm_fault.txt](/uploads/3e498eafbd1997356e9a7f4d8a15411e/vm_fault.txt)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/128Performace drop on AMD TONGA(380x) on kernels 4.9 and 4.10RC32023-05-25T05:45:14ZBugzilla Migration UserPerformace drop on AMD TONGA(380x) on kernels 4.9 and 4.10RC3## Submitted by Miguel Granda Martínez
Assigned to **Default DRI bug account**
**[Link to original bug (#99459)](https://bugs.freedesktop.org/show_bug.cgi?id=99459)**
## Description
Created attachment 129046<br>
Performance drop<b...## Submitted by Miguel Granda Martínez
Assigned to **Default DRI bug account**
**[Link to original bug (#99459)](https://bugs.freedesktop.org/show_bug.cgi?id=99459)**
## Description
Created attachment 129046<br>
Performance drop<br>
<br>
Performance drop beyond kernel 4.8 on games like alien isolation(heavy fps drop), pillars of eternity or tomb raider due to higher cpu usage. GPU AMD TONGA, latest mesa-git, xorg 1.19.1.<br>
<br>
**Attachment 129046**, "Performance drop": <br>
![PerformaceIssues](/uploads/2429e4a0f350b04b142be0c38f1c854a/PerformaceIssues.png)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/125A few amdgpu errors with kernel 4.10-rc2 (Kaveri + Topaz)2023-05-25T05:46:11ZBugzilla Migration UserA few amdgpu errors with kernel 4.10-rc2 (Kaveri + Topaz)## Submitted by SET
Assigned to **Default DRI bug account**
**[Link to original bug (#99313)](https://bugs.freedesktop.org/show_bug.cgi?id=99313)**
## Description
Created attachment 128806<br>
dmesg for amdgpu on kernel 4.10-rc2<b...## Submitted by SET
Assigned to **Default DRI bug account**
**[Link to original bug (#99313)](https://bugs.freedesktop.org/show_bug.cgi?id=99313)**
## Description
Created attachment 128806<br>
dmesg for amdgpu on kernel 4.10-rc2<br>
<br>
This is just to report a few errors with amdgpu in kernel 4.10-rc2 on my laptop, featuring :<br>
<br>
APU : AMD A10-7300 Radeon R6, 10 Compute Cores 4C+6G (from /proc/cpuinfo).<br>
iGPU : 00:01.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Kaveri HDMI/DP Audio Controller<br>
dGPU : 01:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Topaz XT [Radeon R7 M260/M265 / M340/M360 / M440/M445]<br>
Xorg : 1.18.4, on Arch Linux<br>
<br>
$(dmesg |grep -i amdgpu) is in the attachement k4.10-rc2_amdgpu_dmesg.txt.<br>
In brief, there are errors related to UVD and IB with Kaveri iGPU.<br>
<br>
I do have a normal X session, except that the laptop never suspends, and things can then go as bad as corrupting the file system beyond recognition by Grub.<br>
<br>
I am posting this information in the hope it can be useful to developers.<br>
<br>
Regards.<br>
<br>
**Attachment 128806**, "dmesg for amdgpu on kernel 4.10-rc2": <br>
[k4.10-rc2_amdgpu_dmesg.txt](/uploads/366892d1428141ddd3f89cebbd786007/k4.10-rc2_amdgpu_dmesg.txt)<br>
### See also
* https://bugzilla.kernel.org/show_bug.cgi?id=192161https://gitlab.freedesktop.org/drm/amd/-/issues/119Kernel 4.9: amdgpu regression; gui flickers; amd radeon rx 4602023-05-25T05:46:12ZBugzilla Migration UserKernel 4.9: amdgpu regression; gui flickers; amd radeon rx 460## Submitted by Reimar Imhof
Assigned to **Default DRI bug account**
**[Link to original bug (#99275)](https://bugs.freedesktop.org/show_bug.cgi?id=99275)**
## Description
Created attachment 128754<br>
hwinfo<br>
<br>
System: open...## Submitted by Reimar Imhof
Assigned to **Default DRI bug account**
**[Link to original bug (#99275)](https://bugs.freedesktop.org/show_bug.cgi?id=99275)**
## Description
Created attachment 128754<br>
hwinfo<br>
<br>
System: openSuse Leap 42.2,<br>
kernel from download.opensuse.org/repositories/Kernel:/stable/standard/<br>
4.9.0-4.g1af4b0f<br>
Mesa from download.opensuse.org/repositories/X11:/XOrg/openSUSE_Leap_42.2/<br>
<br>
graphics adapter: amd radeon rx 460<br>
cpu: i5-6402p<br>
<br>
desktop: kde5 (from 42.2-oss/update)<br>
<br>
Problem: Address-drop-down in firefox flickers.<br>
<br>
Steps to reproduce:<br>
<br>
start firefox<br>
enter address (example: www.heise.de)<br>
enter an other address<br>
address drop down starts to flicker.<br>
<br>
Perhaps problem shows up only when virtual desktop has been changed before firefox was started. (I use cube animation for virtual desktop change.)<br>
<br>
Problem also occurs with tumbleweed live system.<br>
<br>
There is no problem with kernel 4.8.x (4.8.14)<br>
<br>
I've also reported this bug at https://bugzilla.suse.com/show_bug.cgi?id=1017938<br>
<br>
**Attachment 128754**, "hwinfo": <br>
[hwinfo.txt](/uploads/cf9561687fe600950d385739ae2f5534/hwinfo.txt)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/118Deterministic crash on RX460 "NULL pointer dereference"2023-05-25T05:46:13ZBugzilla Migration UserDeterministic crash on RX460 "NULL pointer dereference"## Submitted by Daniël Mantione
Assigned to **Default DRI bug account**
**[Link to original bug (#99264)](https://bugs.freedesktop.org/show_bug.cgi?id=99264)**
## Description
Created attachment 128734<br>
Complete syslog data from...## Submitted by Daniël Mantione
Assigned to **Default DRI bug account**
**[Link to original bug (#99264)](https://bugs.freedesktop.org/show_bug.cgi?id=99264)**
## Description
Created attachment 128734<br>
Complete syslog data from boot to crash<br>
<br>
Hello,<br>
<br>
I am in the process of migrating from a Radeon HD6670 to an RX 460 for quite a few months now. I regularily fit the RX460, but keep running into issues, crashes and others, that force me to install the HD6670 again if I need my computer for serious work or even a more stable gaming situation. However, I am making progress identifying issues, and it looks like there are 3 different causes for crashes. One of them I can now reproduce very easy, and smells like a real driver bug, so therefore I would like to report it.<br>
<br>
My hardware is as follows:<br>
Xeon E5-2650v2 CPU (once it was an Opteron, but you stopped making new ones :( )<br>
Supermicro X9SRE-3F mainboard<br>
32GB RAM<br>
HIS Radeon RX 460 2GB<br>
3 * HP LP2065 1600x1200 monitor<br>
- One connected via active DP to DVI converter<br>
- One connected via DVI<br>
- One connected via HDMI to DVI cable<br>
<br>
My software configuration is as follows:<br>
OpenSuSE 13.1 with the following modifications:<br>
- Amd-staging-4.7 kernel as of 21 december 2016 (compiled it myself)<br>
(DAL is needed to use all my 3 monitors)<br>
- Xorg upgraded to 7.6 (via http://download.opensuse.org/repositories/X11:/XOrg/openSUSE_13.1/ )<br>
- Mesa upgraded to 13.0.1 (via http://download.opensuse.org/repositories/X11:/XOrg/openSUSE_13.1/ )<br>
<br>
How to reproduce?<br>
<br>
Using the game The Great Whale Road, I have captured the OpenGL command stream with Apitrace:<br>
<br>
http://apitrace.github.io/<br>
<br>
I have uploaded it here, be warned that this is a 770MB download:<br>
<br>
http://www.freepascal.org/~daniel/greatwhaleroad.trace.bz2<br>
<br>
Then:<br>
<br>
bunzip2 greatwhaleroad.trace.bz2<br>
apitrace replay greatwhaleroad.trace<br>
<br>
At the end of the replay, all monitors lose signal and go black. Because my mainboard has a small Aspeed onboard VGA controller, I can switch my monitor input to that VGA controller, login to the Linux VGA text console and recover some information. In dmesg the stack trace below is visible. You can also see that the X server and game processes are still running, but hanging inside the kernel, so they cannot be killed.<br>
<br>
Best regards,<br>
<br>
Daniël Mantione<br>
<br>
[ 1631.286172] BUG: unable to handle kernel NULL pointer dereference at 0000000000000030<br>
[ 1631.333419] IP: [`<ffffffffa08e3e7a>`] amdgpu_gtt_mgr_alloc+0x2a/0x150 [amdgpu]<br>
[ 1631.367823] PGD b9adf067 PUD b8ef5067 PMD 0<br>
[ 1631.402734] Oops: 0000 [#1] SMP<br>
[ 1631.436707] Modules linked in: ppdev parport zram lz4_compress lz4_decompress fuse af_packet k8temp hwmon_vid sr_mod cdrom amdkfd amd_iommu_v2 amdgpu x86_pkg_temp_thermal<br>
intel_powerclamp coretemp snd_seq_dummy snd_seq_oss snd_emu10k1_synth snd_emux_synth snd_seq_virmidi snd_seq_midi_emul snd_seq_midi snd_seq_midi_event kvm_intel<br>
snd_hda_codec_hdmi snd_emu10k1 snd_hda_intel kvm snd_hda_codec snd_rawmidi snd_hda_core snd_ac97_codec ipmi_ssif ac97_bus snd_pcm_oss snd_pcm irqbypass crct10dif_pclmul isci<br>
crc32_pclmul crc32c_intel ttm ghash_clmulni_intel snd_util_mem snd_hwdep snd_seq drbg iTCO_wdt iTCO_vendor_support igb ansi_cprng drm_kms_helper libsas aesni_intel<br>
snd_seq_device snd_timer ablk_helper cryptd lrw snd_mixer_oss gf128mul mei_me ptp glue_helper drm snd emu10k1_gp usb_storage mei scsi_transport_sas<br>
[ 1631.592832] aes_x86_64 pps_core joydev md_mod gameport ioatdma backlight soundcore fb_sys_fops pcspkr serio_raw shpchp sysimgblt i2c_i801 sysfillrect syscopyarea lpc_ich<br>
dca i2c_algo_bit mfd_core wmi ipmi_si ipmi_msghandler button binfmt_misc sg dm_mod autofs4 ext4 mbcache jbd2 crc16 hid_generic usbhid ehci_pci ehci_hcd usbcore sd_mod<br>
usb_common xenbus_probe_frontend reiserfs fan thermal ahci libahci libata scsi_mod [last unloaded: parport_pc]<br>
[ 1631.721604] CPU: 0 PID: 3948 Comm: glretrace Not tainted 4.7.0-2-default+ #1<br>
[ 1631.764334] Hardware name: Supermicro X9SRE/X9SRE-3F/X9SRi/X9SRi-3F/X9SRE/X9SRE-3F/X9SRi/X9SRi-3F, BIOS 3.2a 08/31/2015<br>
[ 1631.785241] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring sdma0 timeout, last signaled seq=10998, last emitted seq=10998<br>
[ 1631.785245] [drm] IP block:tonga_ih is hung!<br>
[ 1631.785518] [drm] Atomic commit: RESET. crtc id 0:[ffff880816a06000]<br>
[ 1631.785540] [drm] Atomic commit: RESET. crtc id 1:[ffff880817700000]<br>
[ 1631.785559] [drm] Atomic commit: RESET. crtc id 2:[ffff88081c3e4000]<br>
[ 1631.785578] [drm] dc_commit_targets: 0 targets<br>
[ 1632.074160] task: ffff88080564ccc0 ti: ffff8808057c0000 task.ti: ffff8808057c0000<br>
[ 1632.119037] RIP: 0010:[`<ffffffffa08e3e7a>`] [`<ffffffffa08e3e7a>`] amdgpu_gtt_mgr_alloc+0x2a/0x150 [amdgpu]<br>
[ 1632.164304] RSP: 0018:ffff8808057c3a10 EFLAGS: 00010282<br>
[ 1632.210314] RAX: ffff880808fe1970 RBX: ffff880818f1f890 RCX: 7fffffffffffffff<br>
[ 1632.256563] RDX: 0000000000000000 RSI: ffff880818f1f858 RDI: ffff880808fe1970<br>
[ 1632.302370] RBP: ffff8808057c3a70 R08: 0000000000000001 R09: ffff8806ccb7b928<br>
[ 1632.348805] R10: ffff880811ebe540 R11: 0000000000000287 R12: 0000000000000000<br>
[ 1632.394486] R13: ffff880818f1f890 R14: ffff880818f1f800 R15: ffff8807bfaf4d80<br>
[ 1632.439637] FS: 00007fed53c56700(0000) GS:ffff88081f200000(0000) knlGS:0000000000000000<br>
[ 1632.485411] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033<br>
[ 1632.531172] CR2: 0000000000000030 CR3: 00000000355cf000 CR4: 00000000001406f0<br>
[ 1632.577066] Stack:<br>
[ 1632.622940] ffff8806ccb7b000 0000000000000680 0000000000000246 ffff8808057c3a50<br>
[ 1632.669400] ffffffff810a1799 ffff880808fe8608 ffff880808fe8608 ffff8807de2293c0<br>
[ 1632.719718] ffff8807de229470 ffff880818f1f890 ffff880818f1f800 ffff8807bfaf4d80<br>
[ 1632.769794] Call Trace:<br>
[ 1632.824132] [`<ffffffff810a1799>`] ? __might_sleep+0x49/0x80<br>
[ 1632.879008] [`<ffffffffa08c7afb>`] amdgpu_ttm_bind+0x5b/0x150 [amdgpu]<br>
[ 1632.932452] [`<ffffffffa08df45d>`] amdgpu_vm_update_page_directory+0x7d/0x480 [amdgpu]<br>
[ 1632.978788] [`<ffffffff811b019b>`] ? krealloc+0x2b/0xa0<br>
[ 1633.025524] [`<ffffffffa040ff54>`] ? ttm_eu_reserve_buffers+0x184/0x330 [ttm]<br>
[ 1633.072382] [`<ffffffffa08ce70b>`] amdgpu_gem_va_update_vm+0x13b/0x180 [amdgpu]<br>
[ 1633.119681] [`<ffffffffa0409c99>`] ? ttm_bo_add_to_lru+0x89/0xe0 [ttm]<br>
[ 1633.167081] [`<ffffffffa08cf7af>`] amdgpu_gem_va_ioctl+0x1df/0x2a0 [amdgpu]<br>
[ 1633.215410] [`<ffffffff810a1799>`] ? __might_sleep+0x49/0x80<br>
[ 1633.262823] [`<ffffffffa050062d>`] drm_ioctl+0x25d/0x510 [drm]<br>
[ 1633.310491] [`<ffffffff8122ea93>`] ? touch_atime+0x23/0xa0<br>
[ 1633.358466] [`<ffffffffa08cf5d0>`] ? amdgpu_gem_metadata_ioctl+0x1f0/0x1f0 [amdgpu]<br>
[ 1633.406696] [`<ffffffffa08b504b>`] amdgpu_drm_ioctl+0x4b/0x80 [amdgpu]<br>
[ 1633.454146] [`<ffffffff81224896>`] do_vfs_ioctl+0x96/0x690<br>
[ 1633.501757] [`<ffffffff81003246>`] ? do_audit_syscall_entry+0x66/0x70<br>
[ 1633.549670] [`<ffffffff81003729>`] ? syscall_trace_enter_phase1+0xf9/0x110<br>
[ 1633.597796] [`<ffffffff81224f09>`] SyS_ioctl+0x79/0x90<br>
[ 1633.645903] [`<ffffffff81003a79>`] do_syscall_64+0x69/0x110<br>
[ 1633.694445] [`<ffffffff81600925>`] entry_SYSCALL64_slow_path+0x25/0x25<br>
[ 1633.742299] Code: 00 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41 55 41 54 53 48 89 cb 48 83 ec 38 4c 8b 21 48 b9 ff ff ff ff ff ff ff 7f 4c 8b 57 30 `<49>` 39 4c 24 30 74 11<br>
31 c0 48 83 c4 38 5b 41 5c 41 5d 41 5e 41<br>
[ 1633.842789] RIP [`<ffffffffa08e3e7a>`] amdgpu_gtt_mgr_alloc+0x2a/0x150 [amdgpu]<br>
[ 1633.894212] RSP `<ffff8808057c3a10>`<br>
[ 1633.944739] CR2: 0000000000000030<br>
[ 1633.994974] ---[ end trace d06de6dc7a13ea3e ]---<br>
[ 1634.047545] [drm] dc_link_handle_hpd_rx_irq: Got short pulse HPD on link 0<br>
[ 1634.047549] amdgpu 0000:04:00.0: SRBM_SOFT_RESET=0x00000400<br>
[ 1635.184392] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.184446] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
[ 1635.184501] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.184543] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
[ 1635.184585] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.184623] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
[ 1635.184673] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :4<br>
[ 1635.184717] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :5<br>
[ 1635.184758] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :6<br>
[ 1635.184799] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :7<br>
[ 1635.184838] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :8<br>
[ 1635.184879] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :9<br>
[ 1635.184917] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :10<br>
[ 1635.184955] [drm:amdgpu_dm_set_crtc_irq_state [amdgpu]] *ERROR* amdgpu_dm_set_crtc_irq_state: crtc is NULL at id :11<br>
[ 1635.184998] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.185038] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
[ 1635.185078] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.185116] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
[ 1635.185158] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_ack: called for non-implemented irq source<br>
[ 1635.185195] [drm:log_to_debug_console [amdgpu]] *ERROR* dal_irq_service_dummy_set: called for non-implemented irq source<br>
<br>
**Attachment 128734**, "Complete syslog data from boot to crash": <br>
[syslog.txt](/uploads/9b70590de1764abac41c9f5e119dbdd1/syslog.txt)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/117r9 390: Hardware cursor invisible after hibernate/resume2019-11-19T16:15:51ZBugzilla Migration Userr9 390: Hardware cursor invisible after hibernate/resume## Submitted by Harald Judt
Assigned to **Default DRI bug account**
**[Link to original bug (#99143)](https://bugs.freedesktop.org/show_bug.cgi?id=99143)**
## Description
I have merged https://cgit.freedesktop.org/~agd5f/linux/log...## Submitted by Harald Judt
Assigned to **Default DRI bug account**
**[Link to original bug (#99143)](https://bugs.freedesktop.org/show_bug.cgi?id=99143)**
## Description
I have merged https://cgit.freedesktop.org/~agd5f/linux/log/?h=drm-next-4.10-wip into 4.9, as this solves the failures to hibernate/resume on my machine with the r9 390. The only problem with hibernation left now is that the HW cursor vanishes after hibernating and resuming. Widgets still get focused, only the cursor is no longer visible. Switching to software cursor as a workaround solves this.<br>
<br>
lspci:<br>
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Hawaii PRO [Radeon R9 290/390] (rev 80)<br>
01:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Hawaii HDMI Audio [Radeon R9 290/290X / 390/390X]<br>
<br>
Using xorg-server-1.18.4, libdrm-git, mesa-git and xf86-video-amdgpu-git.https://gitlab.freedesktop.org/drm/amd/-/issues/116amdgpu doesn't reinitialize after reboot in Xen PV DomU2023-05-25T05:46:16ZBugzilla Migration Useramdgpu doesn't reinitialize after reboot in Xen PV DomU## Submitted by Mateusz Zalega
Assigned to **Default DRI bug account**
**[Link to original bug (#98901)](https://bugs.freedesktop.org/show_bug.cgi?id=98901)**
## Description
Created attachment 128264<br>
Logs from Xen Dom0 and aff...## Submitted by Mateusz Zalega
Assigned to **Default DRI bug account**
**[Link to original bug (#98901)](https://bugs.freedesktop.org/show_bug.cgi?id=98901)**
## Description
Created attachment 128264<br>
Logs from Xen Dom0 and affected DomU<br>
<br>
I'm not sure whether this is a Xen or amdgpu bug.<br>
<br>
amdgpu consistently does not reinitialize after paravirtualized Xen DomU reboots. It's a PCI passthrough setup - the physical machine does not reboot and it can't reset PCI Express devices.<br>
<br>
This setup is working fine until I need to reboot the affected DomU.<br>
<br>
DomU:<br>
Debian testing (29-11-2016)<br>
Linux version 4.7.0-1-amd64<br>
<br>
Dom0:<br>
Debian GNU/Linux 8.6 (jessie)<br>
Linux 3.16.0-4-amd64<br>
<br>
Graphics adapter:<br>
Radeon RX480<br>
<br>
I attached an archive with all the relevant logs I could gather. Please contact me for further information/testing.<br>
<br>
**Attachment 128264**, "Logs from Xen Dom0 and affected DomU": <br>
[bugreport.tar.gz](/uploads/76aa1dba07ac4e5eebbbc5d3fcd464b4/bugreport.tar.gz)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/115On kernel startup, all monitors will go blank and stay blank with newest amd-...2023-05-25T05:47:10ZBugzilla Migration UserOn kernel startup, all monitors will go blank and stay blank with newest amd-staging-4.7## Submitted by Guido Winkelmann
Assigned to **Default DRI bug account**
**[Link to original bug (#98730)](https://bugs.freedesktop.org/show_bug.cgi?id=98730)**
## Description
When booting with the newest kernel from the amd-stagi...## Submitted by Guido Winkelmann
Assigned to **Default DRI bug account**
**[Link to original bug (#98730)](https://bugs.freedesktop.org/show_bug.cgi?id=98730)**
## Description
When booting with the newest kernel from the amd-staging-4.7 branch from https://cgit.freedesktop.org/~agd5f/linux/, all connected monitors will go blank and stay blank indefinitely. The power indicator on the monitors will start to blink as if there was no signal.<br>
<br>
This is with commit 680da8633ae17cdea8fddd1ede87253325c8ca25. With an older commit, d330e1e9c232895dbd0c183a647d41d54ecc0884, this branch still works.<br>
<br>
DAL is enabled in the config in both cases.<br>
<br>
The graphics card is a Sapphire Nitro Radeon R9 380X.<br>
<br>
There are three monitors connected in total, one to the DVI-I port, one to the DVI-D port and one to the DisplayPort.<br>
<br>
I am using this particular branch as a workaround for the bug described in https://bugs.freedesktop.org/show_bug.cgi?id=91202https://gitlab.freedesktop.org/drm/amd/-/issues/114[AMDGPU][CIK] Unable to hibernate functionality is not stable2023-05-25T05:47:12ZBugzilla Migration User[AMDGPU][CIK] Unable to hibernate functionality is not stable## Submitted by Shawn Starr
Assigned to **Default DRI bug account**
**[Link to original bug (#98513)](https://bugs.freedesktop.org/show_bug.cgi?id=98513)**
## Description
Something regressed for CIK on AMDGPU, I am unable to hiber...## Submitted by Shawn Starr
Assigned to **Default DRI bug account**
**[Link to original bug (#98513)](https://bugs.freedesktop.org/show_bug.cgi?id=98513)**
## Description
Something regressed for CIK on AMDGPU, I am unable to hibernate anymore, hangs laptop when it attempts to freeze().https://gitlab.freedesktop.org/drm/amd/-/issues/113Radeon and 5k resolution 5120x2880 = corrupted image2023-05-25T05:47:13ZBugzilla Migration UserRadeon and 5k resolution 5120x2880 = corrupted image## Submitted by ble..@..il.com
Assigned to **Default DRI bug account**
**[Link to original bug (#98461)](https://bugs.freedesktop.org/show_bug.cgi?id=98461)**
## Description
Created attachment 127570<br>
Photo of display<br>
<br>
...## Submitted by ble..@..il.com
Assigned to **Default DRI bug account**
**[Link to original bug (#98461)](https://bugs.freedesktop.org/show_bug.cgi?id=98461)**
## Description
Created attachment 127570<br>
Photo of display<br>
<br>
I try to set 2560x2880 resolution on DisplayPort-0 and the same resolution on DisplayPort-1, but screen is wobbly and distorted.<br>
xrandr --output DisplayPort-0 --mode 2560x2880 --right-of DisplayPort-1 --output DisplayPort-1 --mode 2560x2880<br>
<br>
xrandr <br>
Screen 0: minimum 320 x 200, current 2560 x 1440, maximum 16384 x 16384<br>
DisplayPort-0 connected primary 2560x1440+0+0 (normal left inverted right x axis y axis) 597mm x 336mm<br>
3840x2160 60.00 + 60.00 <br>
2560x1440 59.95*+<br>
2560x2880 59.98 29.99 <br>
1920x1200 59.88 <br>
1920x1080 60.00 <br>
1600x1200 60.00 <br>
1680x1050 59.95 <br>
1280x1024 60.02 <br>
1280x800 59.81 <br>
1024x768 60.00 <br>
800x600 60.32 <br>
640x480 60.00 59.94 <br>
DisplayPort-1 connected (normal left inverted right x axis y axis)<br>
848x480 59.74 +<br>
2560x2880 59.98 29.99 <br>
<br>
<br>
Monitor: Dell UP2715K<br>
VGA: AMD FirePro W2100<br>
<br>
It is related with Bug id 97244 (5k resolution but on Intel card).<br>
<br>
**Attachment 127570**, "Photo of display": <br>
![display](/uploads/9bfe949f5521f134c9193c76a379069e/display.jpg)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/112TTM broken on 4.9-rc22023-05-25T05:47:14ZBugzilla Migration UserTTM broken on 4.9-rc2## Submitted by Armin K. `@krejzi`
Assigned to **Default DRI bug account**
**[Link to original bug (#98417)](https://bugs.freedesktop.org/show_bug.cgi?id=98417)**
## Description
Created attachment 127520<br>
dmesg<br>
<br>
As the ...## Submitted by Armin K. `@krejzi`
Assigned to **Default DRI bug account**
**[Link to original bug (#98417)](https://bugs.freedesktop.org/show_bug.cgi?id=98417)**
## Description
Created attachment 127520<br>
dmesg<br>
<br>
As the title says.<br>
<br>
Relevant dmesg output:<br>
<br>
[ 12.409144] ttm: no symbol version for memcpy<br>
[ 12.409146] ttm: Unknown symbol memcpy (err -22)<br>
[ 12.409167] ttm: no symbol version for clear_page<br>
[ 12.409167] ttm: Unknown symbol clear_page (err -22)<br>
[ 12.409178] ttm: no symbol version for copy_page<br>
[ 12.409179] ttm: Unknown symbol copy_page (err -22)<br>
[ 12.409204] ttm: no symbol version for phys_base<br>
[ 12.409205] ttm: Unknown symbol phys_base (err -22)<br>
[ 12.409216] ttm: no symbol version for memset<br>
[ 12.409216] ttm: Unknown symbol memset (err -22)<br>
[ 12.409219] ttm: no symbol version for ___preempt_schedule<br>
[ 12.409219] ttm: Unknown symbol ___preempt_schedule (err -22)<br>
<br>
Full dmesg output and kernel config are attached.<br>
<br>
**Attachment 127520**, "dmesg": <br>
[dmesg.txt](/uploads/e579443183b8ba28f6d0891188863f01/dmesg.txt)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/1094K @ 60Hz over HDMI 2.0 not working with AMDGPU-DAL2023-05-25T05:47:16ZBugzilla Migration User4K @ 60Hz over HDMI 2.0 not working with AMDGPU-DAL## Submitted by Dillon Dixon
Assigned to **Default DRI bug account**
**[Link to original bug (#98333)](https://bugs.freedesktop.org/show_bug.cgi?id=98333)**
## Description
So this is regarding AMDGPU DAL as it currently exists in ...## Submitted by Dillon Dixon
Assigned to **Default DRI bug account**
**[Link to original bug (#98333)](https://bugs.freedesktop.org/show_bug.cgi?id=98333)**
## Description
So this is regarding AMDGPU DAL as it currently exists in https://cgit.freedesktop.org/~agd5f/linux/log/?h=amd-staging-4.7, which from my understanding contains the most up-to-date public-facing iteration of the DAL patch.<br>
<br>
So I have an HDMI 2.0 4K 60 Hz monitor (Vizio D40u-D1) as my main display connected to an XFX RX 480. I'm using a kernel compiled from the branch listed above.<br>
<br>
Now the issue I'm having is that I'm stuck at 30 Hz. The RX 480 is capable of driving this monitor at 60 Hz @ 4K, which works under Windows. Trying to switch to 60 Hz either results in no signal, or just 30 Hz again.<br>
<br>
I had the same issue in the past with the proprietary Linux Nvidia drivers on a 960 GTX I used to own, and after several months, the devs there were able to figure out what the issue is, describe it in a forum post, and get a fix in. See here: https://devtalk.nvidia.com/default/topic/939971/linux/4k-60hz-works-in-windows-not-in-linux/<br>
<br>
So what I'm hoping is that a similar fix would work in DAL.<br>
<br>
I can provide logs on request (not sure to log right now).<br>
<br>
Thanks!https://gitlab.freedesktop.org/drm/amd/-/issues/108[DC] amd-staging-4.7: problems with unblanking displays when monitors are swi...2023-11-13T03:28:36ZBugzilla Migration User[DC] amd-staging-4.7: problems with unblanking displays when monitors are switched off## Submitted by Darren Salt
Assigned to **Default DRI bug account**
**[Link to original bug (#98324)](https://bugs.freedesktop.org/show_bug.cgi?id=98324)**
## Description
Created attachment 127405<br>
Kernel log extract, showing t...## Submitted by Darren Salt
Assigned to **Default DRI bug account**
**[Link to original bug (#98324)](https://bugs.freedesktop.org/show_bug.cgi?id=98324)**
## Description
Created attachment 127405<br>
Kernel log extract, showing the unblanking failure.<br>
<br>
Setup: RX 470 with two monitors attached – one via DVI-and one via HDMI. (Both are AOC monitors.)<br>
<br>
* Ensure that outputs are blanked (via xset, time-out or similar).<br>
* Switch off the monitors.<br>
* Unblank the outputs.<br>
* Switch on the monitors.<br>
<br>
Invariably, in this situation, the DVI-connected monitor is unblanked and displaying what it should be, but the HDMI-connected monitor is receiving no signal (or at least nothing to display) and is therefore displaying nothing.<br>
<br>
The HDMI-connected monitor can be restored by using xrandr to change mode or by blanking and unblanking – so long as it's switched on while doing so.<br>
<br>
~~**Attachment 127405**~~, "Kernel log extract, showing the unblanking failure.": <br>
[log-extract.txt](/uploads/6838c3455119d32e32334160e20f8c11/log-extract.txt)<br>https://gitlab.freedesktop.org/drm/amd/-/issues/106Kernel Panic on shutdown caused by "drm/amdgpu: always apply pci shutdown cal...2019-11-19T20:40:48ZBugzilla Migration UserKernel Panic on shutdown caused by "drm/amdgpu: always apply pci shutdown callbacks"## Submitted by Johannes Hirte
Assigned to **Default DRI bug account**
**[Link to original bug (#98276)](https://bugs.freedesktop.org/show_bug.cgi?id=98276)**
## Description
Kernel 4.9-rc1 fails on shutdown/reboot with a panic. Bi...## Submitted by Johannes Hirte
Assigned to **Default DRI bug account**
**[Link to original bug (#98276)](https://bugs.freedesktop.org/show_bug.cgi?id=98276)**
## Description
Kernel 4.9-rc1 fails on shutdown/reboot with a panic. Bisect pointed me to <br>
<br>
00ea8cba5ef7b783f11cb1a0b900b7c18d2ce0b6 is the first bad commit<br>
commit 00ea8cba5ef7b783f11cb1a0b900b7c18d2ce0b6<br>
Author: Alex Deucher <alexander.deucher@amd.com><br>
Date: Thu Sep 22 14:40:29 2016 -0400<br>
<br>
drm/amdgpu: always apply pci shutdown callbacks (v2)<br>
<br>
We can't properly detect all hypervisors and we<br>
need this to properly tear down the hardware.<br>
<br>
v2: trivial warning fix<br>
<br>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com><br>
<br>
:040000 040000 868a00ff980f95dac82b908df0737342bf40cdc3 02bf2cf4e8797d37d454301afedf2f97735b60ca M drivers<br>
<br>
After removing this commit, everything works again. System is a Carrizo.https://gitlab.freedesktop.org/drm/amd/-/issues/104[amdgpu SI] "gpu fault detect" in Alien: Isolation2023-05-25T05:48:09ZBugzilla Migration User[amdgpu SI] "gpu fault detect" in Alien: Isolation## Submitted by Arek Ruśniak
Assigned to **Default DRI bug account**
**[Link to original bug (#98183)](https://bugs.freedesktop.org/show_bug.cgi?id=98183)**
## Description
Created attachment 127168<br>
kernel log<br>
<br>
Hi, <br>...## Submitted by Arek Ruśniak
Assigned to **Default DRI bug account**
**[Link to original bug (#98183)](https://bugs.freedesktop.org/show_bug.cgi?id=98183)**
## Description
Created attachment 127168<br>
kernel log<br>
<br>
Hi, <br>
kernel - drm-next-4.9-wip + drm/amdgpu: disable CRTCs before teardown<br>
mesa,libdrm,amdgpu,llvm form git/svn<br>
This is not regression, this happened since dpm is workable. <br>
<br>
GPU/VM faults appears during loading game only, menu/gameplay(20min) looks normal. <br>
Probably this problem doesn't performance impact, frame rate is the same as for radeon.<br>
<br>
**Attachment 127168**, "kernel log": <br>
[alien-gpu_faults.log](/uploads/aa4b76a4dc51beda8af675221bffd327/alien-gpu_faults.log)<br>