WARNING at dmesg?

deepforest · Yesterday 22:04:30

what this mean?

[  256.582629] ------------[ cut here ]------------
[  256.582641] WARNING: CPU: 2 PID: 2479 at drivers/gpu/drm/ttm/ttm_bo.c:252 ttm_bo_release+0x299/0x2f0 [ttm]
[  256.582671] Modules linked in: qrtr rfkill lp amdgpu amdxcp drm_exec gpu_sched drm_buddy ppdev radeon coretemp kvm_intel drm_suballoc_helper drm_display_helper kvm snd_hda_codec_realtek cec rc_core snd_hda_codec_generic snd_hda_codec_hdmi drm_ttm_helper snd_hda_scodec_component ttm snd_hda_intel drm_kms_helper snd_intel_dspcfg snd_intel_sdw_acpi snd_hda_codec irqbypass joydev drm snd_hda_core snd_hwdep sha512_ssse3 sha256_ssse3 snd_pcm i2c_algo_bit sha1_ssse3 video snd_timer vga16fb wmi sg vgastate snd evdev parport_pc pcspkr soundcore parport acpi_cpufreq button ext4 crc16 mbcache jbd2 crc32c_generic hid_generic usbhid hid sd_mod ata_generic r8169 uhci_hcd iTCO_wdt ata_piix intel_pmc_bxt libata iTCO_vendor_support ehci_pci realtek ehci_hcd watchdog scsi_mod i2c_i801 mdio_devres usbcore serio_raw i2c_smbus libphy scsi_common lpc_ich usb_common
[  256.582746] CPU: 2 UID: 1000 PID: 2479 Comm: xfwm4:rcs0 Tainted: G          I        6.12.90+deb13.1-amd64 #1  Debian 6.12.90-2
[  256.582750] Tainted: [I]=FIRMWARE_WORKAROUND
[  256.582751] Hardware name: Gigabyte Technology Co., Ltd. P31-DS3L/P31-DS3L, BIOS F10A 07/30/2008
[  256.582753] RIP: 0010:ttm_bo_release+0x299/0x2f0 [ttm]
[  256.582761] Code: 24 40 08 00 00 41 8b bc 24 e0 00 00 00 48 83 c4 38 48 8d 53 30 5b 5d 41 5c 41 5d 41 5e e9 6f 4c e2 f0 4c 89 e7 e9 52 fe ff ff <0f> 0b 48 83 7b 20 00 0f 84 99 fd ff ff 0f 0b e9 92 fd ff ff c7 43
[  256.582764] RSP: 0018:ffffcf054050fa70 EFLAGS: 00010202
[  256.582767] RAX: 0000000000000001 RBX: ffff8e50b482ade0 RCX: 0000000000000000
[  256.582768] RDX: 0000000000000001 RSI: ffff8e51805dfc70 RDI: ffff8e50b482ade0
[  256.582770] RBP: ffff8e50b482ac68 R08: 0000000000000064 R09: ffff8e51805dfb40
[  256.582771] R10: 000000000015000c R11: ffff8e50b4895cc0 R12: ffff8e51804c5740
[  256.582773] R13: ffff8e5180201220 R14: ffff8e51804c56c0 R15: 0000000000000000
[  256.582774] FS:  00007f1529fff6c0(0000) GS:ffff8e519bd00000(0000) knlGS:0000000000000000
[  256.582776] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  256.582778] CR2: 00007fe74c6ba41c CR3: 00000000012ca000 CR4: 00000000000026f0
[  256.582780] Call Trace:
[  256.582784]  <TASK>
[  256.582788]  ? fsnotify_destroy_marks+0x2a/0x1a0
[  256.582794]  ? __call_rcu_common+0xcb/0x6d0
[  256.582799]  drm_gem_dmabuf_release+0x37/0x50 [drm]
[  256.582911]  dma_buf_release+0x3b/0x90
[  256.582917]  __dentry_kill+0x90/0x190
[  256.582920]  dput+0xeb/0x1b0
[  256.582923]  __fput+0x12d/0x2a0
[  256.582927]  task_work_run+0x59/0x90
[  256.582931]  do_exit+0x330/0xad0
[  256.582936]  ? __pfx_futex_wake_mark+0x10/0x10
[  256.582942]  do_group_exit+0x30/0x80
[  256.582944]  get_signal+0x87c/0x880
[  256.582949]  arch_do_signal_or_restart+0x3f/0x250
[  256.582953]  syscall_exit_to_user_mode+0x139/0x1b0
[  256.582958]  do_syscall_64+0x93/0x1b0
[  256.582963]  ? do_futex+0x125/0x190
[  256.582965]  ? __x64_sys_futex+0x127/0x1e0
[  256.582968]  ? arch_exit_to_user_mode_prepare.isra.0+0x16/0xa0
[  256.582973]  ? syscall_exit_to_user_mode+0x37/0x1b0
[  256.582975]  ? do_syscall_64+0x93/0x1b0
[  256.582978]  ? __rseq_handle_notify_resume+0xa2/0x490
[  256.582983]  ? do_futex+0xcb/0x190
[  256.582985]  ? __x64_sys_futex+0x127/0x1e0
[  256.582988]  ? switch_fpu_return+0x4e/0xd0
[  256.582993]  ? arch_exit_to_user_mode_prepare.isra.0+0x75/0xa0
[  256.582996]  ? syscall_exit_to_user_mode+0x37/0x1b0
[  256.582998]  ? do_syscall_64+0x93/0x1b0
[  256.583001]  ? syscall_exit_to_user_mode+0x37/0x1b0
[  256.583003]  ? do_syscall_64+0x93/0x1b0
[  256.583006]  ? arch_exit_to_user_mode_prepare.isra.0+0x16/0xa0
[  256.583009]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  256.583014] RIP: 0033:0x7f153a69e9ee
[  256.583016] Code: Unable to access opcode bytes at 0x7f153a69e9c4.
[  256.583018] RSP: 002b:00007f1529ffe9b8 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca
[  256.583020] RAX: fffffffffffffe00 RBX: 00007f1529fff6c0 RCX: 00007f153a69e9ee
[  256.583022] RDX: 0000000000000363 RSI: 0000000000000189 RDI: 0000558c589a4a2c
[  256.583023] RBP: 0000000000000000 R08: 0000000000000000 R09: 00000000ffffffff
[  256.583025] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
[  256.583026] R13: 0000558c589a49d8 R14: 0000000000000365 R15: 00000000000006cb
[  256.583028]  </TASK>
[  256.583029] ---[ end trace 0000000000000000 ]---
root@devuan:/home/freeartist-devuan#

ralph.ronnquist · Yesterday 23:09:06

Perhaps it says that the ttm_bo_release module has an OOB dma access? At a guess it's a puts call with a non-terminated string...

deepforest · Yesterday 23:46:17

Perhaps it says that the ttm_bo_release module as an OOB dma access? At a guess it's a puts call with a non-terminated string...

idk
its not related to this free radeon driver bug?
https://gitlab.freedesktop.org/xorg/xse … _items/956

steve_v · Today 08:09:58

its not related to this free radeon driver bug?

Why would it be? That link is a bug report filed against xorg-server, talking about spurious modelines with the modesetting driver. It has nothing to do with radeon, free or otherwise, and it makes no mention of xfwm crashing or a kernel oops releasing DRM translation tables.

I assume this is an AMD card of some kind, as both radeon and amdgpu are loaded. Which drivers (kernel module and xorg/DDX) are you actually using?

My bet is on a driver bug, especially if you are using amdgpu with ancient hardware (feat. motherboard from 2007).
OTOH, if your GPU is GCN 1.0 or newer and you're using radeon, kernel 6.12.x should have support for those in amdgpu as well, it just didn't become the default until 6.19. Can't hurt to try either way.

Ed. Re. your other thread, I assume this is about the HD5770. Forget about the above, that card is old enough to have a retroweb page and is only supported by the 'radeon' driver.

Last edited by steve_v (Today 09:24:36)

deepforest · Today 14:36:16

Why would it be? That link is a bug report filed against xorg-server, talking about spurious modelines with the modesetting driver. It has nothing to do with radeon, free or otherwise, and it makes no mention of xfwm crashing or a kernel oops releasing DRM translation tables.

i take it from here, where man ask about refresh rates with AMD card
https://bbs.archlinux.org/viewtopic.php … 5#p1895205
read good and you will see comment

So, what this a strange "warning"?

Last edited by deepforest (Today 14:38:21)

steve_v · Today 15:33:03

So, what this a strange "warning"?

It's a kernel warning (correcting my earlier comment: similar to, but lower severity than a true "oops") from the DRM translation table manager.
xfwm4 was terminated (by signal, but we don't know from where, could be e.g an xserver crash or a manual 'kill'), the kernel tried to release shared graphics memory buffers, then it hit an invalid opcode - likely because those buffers were in an inconsistent state.
The operation was aborted safely, thus it's a warning rather than an error.
If you want more detail, check the rest of your logs or go ask someone intimately familiar with ttm_bo.c.

That message still has nothing to do with the arch forum thread or the git issue you linked, or refresh rates for that matter... Unless of course it was your screwing about with modelines and unsupported refresh rates that caused the problem in the first place, which given your other thread is entirely possible.

The officially official Devuan Forum!

#1 Yesterday 22:04:30

WARNING at dmesg?

#2 Yesterday 23:09:06

Re: WARNING at dmesg?

#3 Yesterday 23:46:17

Re: WARNING at dmesg?

#4 Today 08:09:58

Re: WARNING at dmesg?

#5 Today 14:36:16

Re: WARNING at dmesg?

#6 Today 15:33:03

Re: WARNING at dmesg?

Board footer