[TRACKING] Graphical corruption in Fedora 39 (AMD 3.03 BIOS)

gxnni · November 14, 2023, 5:36pm

It seems that whenever a graphical component of the OS is adjusted in scale, the is issue occurs. In my experience, Gnome’s experimental fractional scaling will trigger it the most. Here is more information on that: How to Enable Fractional Scaling on Fedora Linux - OMG! Linux

For me, i have my AMD FW 13 (Ryzen 5 CPU with 32 GB of RAM) set to 125%, with amdgpu.sg value set to 0, and the flashing squares have not appear for about a couple of days now. Not sure if this is exactly because of the Kernel setting, or if an update possibly resolved the issue altogether. I have dnf-automatic enabled, so it could have been that.

Alternatively, you can try adjusting the Scaling Factor in Gnome tweaks to see if that floats our boat. For me, I’m not a fan of using it too much, it at all since small UI with big text looks odd.

This has definitely been one of the bigger hurdles to getting acquainted with my new Framework. Unfortunately I think it is a matter of tuning until it works better for you.

jwp · November 14, 2023, 6:15pm

I’ve got my first instance of this bug on fedora rawhide (6.7 kernel, with xxmitsu/mesa-git) - It looks like for me it’s preceded by steam runtime coredumping/iommu issues coredumping. My guess is it is corrupting/not freeing some memory region accessed by amdgpu during a power state change.

Steps to reproduce:

a) Launch steam - run something intensive for a while (I used civ6 with proton-ge23)
b) observe steam runtime coredumps and/or iommu errrors in journal
c) change power state from being AC powered to Battery
d) watch as machine goes into sleep/idle
e) Quickly resume from sleep.
f) Get graphical corruption on resume (in my case a whited out screen where sddm/plasma unlock should be) - my cursor was still responsive/active but vtty switching was broken.
g) Hold power to hard reset
h) Look at previous 10m journal

Attached is output from my journal

Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc0000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc1000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc2000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc3000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc4000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc5000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc6000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc7000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dc8000 flags=0x0000]
Nov 15 06:58:00 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff90dd4000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amd_iommu_report_page_fault: 87654 callbacks suppressed
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83400000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83401000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83402000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83403000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83404000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83405000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83406000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83407000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83408000 flags=0x0000]
Nov 15 06:58:05 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83414000 flags=0x0000]
Nov 15 06:58:08 emiemi kernel: i2c_hid_acpi i2c-FRMW0005:00: i2c_hid_get_input: incomplete report (7/65535)
Nov 15 06:58:10 emiemi kernel: amd_iommu_report_page_fault: 87201 callbacks suppressed
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83640000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83641000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83642000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83643000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83644000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83645000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83646000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83647000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83648000 flags=0x0000]
Nov 15 06:58:10 emiemi kernel: amdgpu 0000:c1:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xfff83654000 flags=0x0000]
Nov 15 06:58:15 emiemi kernel: amd_iommu_report_page_fault: 86848 callbacks suppressed
                                                #26 0x00007fb1552efdda n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #6  0x00007fb152d2c19c __clone3 (libc.so.6 + 0x10f19c)
                                                
                                                Stack trace of thread 36523:
                                                #0  0x00007fb152ca8799 __futex_abstimed_wait_common (libc.so.6 + 0x8b799)
                                                #1  0x00007fb152cab139 pthread_cond_wait@@GLIBC_2.3.2 (libc.so.6 + 0x8e139)
                                                #2  0x00007fb13e76e64d cnd_wait (radeonsi_dri.so + 0x16e64d)
                                                #3  0x00007fb13e74d4bb util_queue_thread_func (radeonsi_dri.so + 0x14d4bb)
                                                #4  0x00007fb13e76e57c impl_thrd_routine (radeonsi_dri.so + 0x16e57c)
                                                #5  0x00007fb152cabec7 start_thread (libc.so.6 + 0x8eec7)
                                                #6  0x00007fb152d2c19c __clone3 (libc.so.6 + 0x10f19c)
                                                
                                                Stack trace of thread 36520:
                                                #0  0x00007fb152ca8799 __futex_abstimed_wait_common (libc.so.6 + 0x8b799)
                                                #1  0x00007fb152cab139 pthread_cond_wait@@GLIBC_2.3.2 (libc.so.6 + 0x8e139)
                                                #2  0x00007fb13e76e64d cnd_wait (radeonsi_dri.so + 0x16e64d)
                                                #3  0x00007fb13e74d4bb util_queue_thread_func (radeonsi_dri.so + 0x14d4bb)
                                                #4  0x00007fb13e76e57c impl_thrd_routine (radeonsi_dri.so + 0x16e57c)
                                                #5  0x00007fb152cabec7 start_thread (libc.so.6 + 0x8eec7)
                                                #6  0x00007fb152d2c19c __clone3 (libc.so.6 + 0x10f19c)
                                                
                                                Stack trace of thread 36529:
                                                #0  0x00007fb152ca8799 __futex_abstimed_wait_common (libc.so.6 + 0x8b799)
                                                #1  0x00007fb152cab139 pthread_cond_wait@@GLIBC_2.3.2 (libc.so.6 + 0x8e139)
                                                #2  0x00007fb13e76e64d cnd_wait (radeonsi_dri.so + 0x16e64d)
                                                #3  0x00007fb13e74d4bb util_queue_thread_func (radeonsi_dri.so + 0x14d4bb)
                                                #4  0x00007fb13e76e57c impl_thrd_routine (radeonsi_dri.so + 0x16e57c)
                                                #5  0x00007fb152cabec7 start_thread (libc.so.6 + 0x8eec7)
                                                #6  0x00007fb152d2c19c __clone3 (libc.so.6 + 0x10f19c)
                                                
                                                Stack trace of thread 36521:
                                                #0  0x00007fb152ca8799 __futex_abstimed_wait_common (libc.so.6 + 0x8b799)
                                                #1  0x00007fb152cab139 pthread_cond_wait@@GLIBC_2.3.2 (libc.so.6 + 0x8e139)
                                                #2  0x00007fb13e76e64d cnd_wait (radeonsi_dri.so + 0x16e64d)
                                                #3  0x00007fb13e74d4bb util_queue_thread_func (radeonsi_dri.so + 0x14d4bb)
                                                #4  0x00007fb13e76e57c impl_thrd_routine (radeonsi_dri.so + 0x16e57c)
                                                #5  0x00007fb152cabec7 start_thread (libc.so.6 + 0x8eec7)
                                                #6  0x00007fb152d2c19c __clone3 (libc.so.6 + 0x10f19c)
                                                
                                                Stack trace of thread 36509:
                                                #0  0x00007fb152d1e61d __poll (libc.so.6 + 0x10161d)
                                                #1  0x00007fb15297d0ba _xcb_conn_wait.part.0 (libxcb.so.1 + 0xe0ba)
                                                #2  0x00007fb15297f1ac xcb_wait_for_special_event (libxcb.so.1 + 0x101ac)
                                                #3  0x00007fb141c6ade1 dri3_wait_for_event_locked (libGLX_mesa.so.0 + 0x51de1)
                                                #4  0x00007fb141c6c35b loader_dri3_wait_for_msc (libGLX_mesa.so.0 + 0x5335b)
                                                #5  0x00007fb141c5da33 dri3_drawable_get_msc (libGLX_mesa.so.0 + 0x44a33)
                                                #6  0x00007fb14103dc86 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libGLESv2.so +>
                                                #7  0x00007fb14103f887 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libGLESv2.so +>
                                                #8  0x00007fb140f2c429 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libGLESv2.so +>
                                                #9  0x00007fb140e5295b n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libGLESv2.so +>
                                                #10 0x00007fb158f02a2b n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #11 0x00007fb158ef1e24 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #12 0x00007fb158eed2a3 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #13 0x00007fb157c16176 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #14 0x00007fb157c26bbc n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #15 0x00007fb157bdfa6a n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #16 0x00007fb157c27284 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #17 0x00007fb157bfee3e n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #18 0x00007fb15c830712 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #19 0x00007fb157a69c2a n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #20 0x00007fb157a6add2 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #21 0x00007fb15a3e012f n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #22 0x00007fb15a3e067e n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #23 0x00007fb157a69101 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #24 0x00007fb157afb93c n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #25 0x00007fb157ad4e5d n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #26 0x00007fb1552efdda n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/libcef.so + 0x>
                                                #27 0x00000000005c48b0 n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/steamwebhelper>
                                                #28 0x00000000005871ab n/a (/home/aenertia/.local/share/Steam/ubuntu12_64/steamwebhelper>
                                                ELF object binary architecture: AMD x86-64
Nov 15 06:57:23 emiemi systemd[1]: systemd-coredump@7-54213-0.service: Deactivated successfully.
Nov 15 06:57:23 emiemi audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=syst>
Nov 15 06:57:23 emiemi systemd[1]: systemd-coredump@7-54213-0.service: Consumed 2.057s CPU time.
Nov 15 06:57:23 emiemi audit: BPF prog-id=175 op=UNLOAD
Nov 15 06:57:23 emiemi audit: BPF prog-id=174 op=UNLOAD

Matt_Hartley · November 15, 2023, 1:46am

This is correct. The bug is filed, AMD is actively working on it and is also active here as well.

OEM C 22.04.3 is what we support. I’d test against this.

Matt_Hartley · November 15, 2023, 1:50am

jwp:

I’ve got my first instance of this bug on fedora rawhide (6.7 kernel, with xxmitsu/mesa-git) - It looks like for me it’s preceded by steam runtime coredumping/iommu issues coredumping. My guess is it is corrupting/not freeing some memory region accessed by amdgpu during a power state change.

Steps to reproduce:

a) Launch steam - run something intensive for a while (I used civ6 with proton-ge23)
b) observe steam runtime coredumps and/or iommu errrors in journal
c) change power state from being AC powered to Battery
d) watch as machine goes into sleep/idle
e) Quickly resume from sleep.
f) Get graphical corruption on resume (in my case a whited out screen where sddm/plasma unlock should be) - my cursor was still responsive/active but vtty switching was broken.
g) Hold power to hard reset
h) Look at previous 10m journal

Attached is output from my journal

Appreciate this @jwp, if you feel this differs at all from the other mentions of odd AMD blocky/graphical corruption, please file a bug report with them. If you feel this is in line, did enabling the kernel parameter or UMA_Game_Optimized in BIOS helped at all.

elehayyme · November 15, 2023, 2:26pm

I’ve run into this issue and would like to try and add in that amdgpu.sg_display=0 parameter that was alluded to above. However, I’m not too familiar in amending the kernel to make the change (how to edit/where to put in the line). I’m using pop OS (yes, I know that it isn’t officially supported so I wanted to put in the question here as I couldn’t reach out to support).

If someone could provide guidance on this, that’d be much appreciated. I only started using the new laptop about a day ago and am a little concerned about screwing something up.

Jonathan_Haas · November 15, 2023, 3:05pm

I don’t use Pop OS, but from a quick google and look at the documentation, it seems you should run

kernelstub -a "amdgpu.sg_display=0"

(if necessary with sudo).

However I haven’t test this, maybe look up kernelstub yourself and see if it seems right to you.

elehayyme · November 15, 2023, 5:58pm

@Jonathan_Haas Thank you for this! I’ll look into this a bit more as you suggested.

jwp · November 15, 2023, 6:50pm

@Matt_Hartley ; The UMA 4GB VRAM flag should be the default as shipped i’m not sure why it isn’t. 512MB reserved is far too small for the Display Size IMNSHO.

But yes the UMA bios flag not being set out of the box was causing other issues with slow performance in a few things hitting the GPU - so I had it turned on quite early in testing.

Matt_Valites · November 15, 2023, 6:55pm

Yes, sorry I meant 22.04.3 with the 6.1.x OEM C kernel (I can’t seem to figure out how to edit the post).

In my case that has resolved the flicker/artifacts. I have other nits, mostly minor performance or UX related annoyances. But I’ll survive

Thanks,

Mario_Limonciello · November 15, 2023, 7:27pm

Those of you that have hit issues that amdgpu.sg_display=0 or changing VRAM flags helped, can I ask you to revert to the default BIOS settings and see if amdgpu.dcdebugmask=0x10 can also help your issues? This will disable PSR, which I’m wondering if maybe be playing into the underlying issue.

framy · November 15, 2023, 8:48pm

I have done:
gsettings reset org.gnome.shell disable-extension-version-validation
and uninstalled gnome-theme-extra, this seemed firstly to work, but now I get the flickering back. I may try the amdgpu.sg_display=0
I got actually a few bug reports in the program for that (“sorry crashed thing”), I sended one, but may finished with errors and others I can’t even report (get errors then).
But no errors accour when flickering happens (like right now while typing).
Have done now sudo systemctl restart gdm and have now the following via journalctl --since 20 minutes:

21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac60000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac61000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac62000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac63000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac64000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac65000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac66000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac67000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac68000 flags=0x0000]
21:38:40 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac34000 flags=0x0000]
21:38:40 my-laptop rtkit-daemon[1716]: Successfully made thread 11533 of process 11498 (/usr/bin/gnome-shell) owned by '1000' RT at priority 20.
21:38:40 my-laptop goa-daemon[11878]: goa_http_client_check() failed: 0 — (null)
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab5e5150 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab5e5150 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aad663d0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aad663d0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aad668b0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aad668b0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab93d070 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab93d070 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab93d550 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab93d550 Gjs_window-list_gnome-shell-extensions_gcampax_github_com
---
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab60bc30 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab60bc30 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab60c110 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab60c110 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab8697f0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab8697f0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab869cd0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab869cd0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab81d3d0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab81d3d0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab6d45b0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:45 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578ab6d45b0 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
---
21:38:45 my-laptop chronyd[1808]: Forward time jump detected!
---
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop systemd[1]: iscsi.service: Unit cannot be reloaded because it is inactive.
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:45 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
---
21:38:45 my-laptop kernel: amd_iommu_report_page_fault: 6272 callbacks suppressed
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac20000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac21000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac22000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac23000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac24000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac25000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac26000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac27000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac28000 flags=0x0000]
21:38:45 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac34000 flags=0x0000]
---
21:38:50 my-laptop kernel: amd_iommu_report_page_fault: 29488 callbacks suppressed
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac20000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac21000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac22000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac23000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac24000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac25000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac26000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac27000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac28000 flags=0x0000]
21:38:50 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac34000 flags=0x0000]
21:38:55 my-laptop kernel: amd_iommu_report_page_fault: 18991 callbacks suppressed
---
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac20000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac21000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac22000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac23000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac24000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac25000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac26000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac27000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac70000 flags=0x0000]
21:38:55 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac7c000 flags=0x0000]
---
21:38:59 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aba88a50 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:59 my-laptop gnome-shell[11498]: st_widget_get_theme_node called on the widget [0x5578aba88a50 Gjs_window-list_gnome-shell-extensions_gcampax_github_com_workspaceIndica>
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: Meta.Rectangle is deprecated, use Mtk.Rectangle instead
21:38:59 my-laptop gnome-shell[11498]: meta_window_set_stack_position_no_sync: assertion 'window->stack_position >= 0' failed
21:39:00 my-laptop kernel: amd_iommu_report_page_fault: 23489 callbacks suppressed
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac00000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac01000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac02000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac03000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac04000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac05000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac06000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac07000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac08000 flags=0x0000]
21:39:00 my-laptop kernel: amdgpu 0000:c0:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0005 address=0xffffac14000 flags=0x0000]
21:39:05 my-laptop kernel: amd_iommu_report_page_fault: 25428 callbacks suppressed
---

and so on. The amd_iommu_report_page_fault and then the [IO_PAGE_FAULT] multiple times
--- means lines removed, e.g. NetworkManager lines.
Flickering happend after the laptop was multiple hours closed.
If the @Framework Team wants to see the report logs (e.g. the one I sent), write me via DM.

Tim_Bosse · November 16, 2023, 4:31pm

I’ve had one hiccup since following your suggestion and here is what spat out on dmesg.

[64294.541861] amdgpu 0000:c1:00.0: amdgpu: [mmhub] page fault (src_id:0 ring:24 vmid:2 pasid:32780, for process RDD Process pid 14453 thread firefox:cs0 pid 14471)
[64294.541871] amdgpu 0000:c1:00.0: amdgpu:   in page starting at address 0x0000800100bda000 from client 18
[64294.541875] amdgpu 0000:c1:00.0: amdgpu: MMVM_L2_PROTECTION_FAULT_STATUS:0x00203A30
[64294.541878] amdgpu 0000:c1:00.0: amdgpu:      Faulty UTCL2 client ID: unknown (0x1d)
[64294.541880] amdgpu 0000:c1:00.0: amdgpu:      MORE_FAULTS: 0x0
[64294.541883] amdgpu 0000:c1:00.0: amdgpu:      WALKER_ERROR: 0x0
[64294.541885] amdgpu 0000:c1:00.0: amdgpu:      PERMISSION_FAULTS: 0x3
[64294.541887] amdgpu 0000:c1:00.0: amdgpu:      MAPPING_ERROR: 0x0
[64294.541889] amdgpu 0000:c1:00.0: amdgpu:      RW: 0x0

Mario_Limonciello · November 16, 2023, 5:03pm

Thanks, it confirms this is not a PSR sourced issue. The S/G or increased VRAM in BIOS are best workarounds for now.

Could you bring a bug report to AMD Gitlab? This needs to be in front of more people at AMD.

Tim_Bosse · November 16, 2023, 5:20pm

I should probably be clear what I did before we call this the problem. I have been working happily with the following:

sudo grubby --args="amdgpu.sg_display=0 amd_iommu=off" --update-kernel=ALL

When you had me try without amdgpu.sg_display=0,I went and removed amd_iommu=off as well…

I have added back amdgpu.sg_display=0 to the amdgpu.dcdebugmask=0x10 argument and so far (5 minutes trying to make things twitchy), no problems.

jwp · November 16, 2023, 9:44pm

IOMMU is kinda important and shouldn’t be disabled ideally. Disabling the amdgpu Scatter gather alone was enough to fix the bug for me.

I might try and see if disabling IOMMU whilst forcing scatter gather on changes behaviour. But IOMMU is kinda important for a heap of other things like SRIOV/Virt and I would imagine it likely has a relatively important place in combination with the amdgpu APU for shuffling memory around consistently.

J3d · November 18, 2023, 2:22pm

Seems like the amd-gpu-firmware update is out of testing and is now globally available.

Once this is installed, can we remove amdgpu.sg_display=0 from our kernel params?

gxnni · November 19, 2023, 12:31am

Seems like it does not fix the graphical corruption, at least for me. I just had the issue arise for me again after I plugged my system into my USBC TB dock, with the lid open. I set Kernel parameter back to the adjusted variable (0).

Alex_H · November 19, 2023, 10:04am

I have the same issue and unfortunately neither amdgpu.sg_display=0 nor adding amdgpu.dcdebugmask=0x10 to it made any difference. Fedora 39 with BIOS 3.03. Other threads mention reseating the left display connector on the motherboard, which I did. The cable looks healthy and undamaged.

Charging and running heavy loads at the same time makes the laptop unusable due to extreme flickering. So I either have to run only on battery, or stick to light tasks like e-mail and web browsing.

jwp · November 19, 2023, 6:57pm

I had my first failure to resume yesterday even with the SG toggle. Like use after heavy load both plugged and unplugged. I am wondering if it’s thermal related

mikeymop · November 20, 2023, 7:05am

I can trigger this issue much more often if I leave an electron application open during suspend.
I suspect the Javascript VM garbage collection puts more memory pressure on the suspended system triggering the artifacting.

It seems anything with significant memory usage variances can trigger this to happen as even a Rust Gtk application that I set to sync in the background can trigger this issue also.

None of the acpi kernel params for amdgpu alter this behavior, it will always come up if there is memory usage variances in suspend and you wait long enough.

Personally, I don’t believe this is on Framework and am holding out for kernel 6.6 (and it’s bundled firmware) to drop hopefully resolving this as is has a lot of AMD improvements in it.