7.0-rc4 dcn31_program_compbuf_size trace

7.0-rc4 dcn31_program_compbuf_size trace

[Julian Wollrath]

Dear maintainers,

when turning the display of via 'xset dpms force off' I observe the
following warning and trace under 7.0-rc4 with a AMD Ryzen 7 PRO 8840U
w/ Radeon 780M Graphics card. This I did not observe under 6.19.7.

kernel: amdgpu 0000:c4:00.0: [drm] REG_WAIT timeout 1us * 100 tries - dcn31_program_compbuf_size line:142
kernel: ------------[ cut here ]------------
kernel: WARNING: drivers/gpu/drm/amd/amdgpu/../display/dc/hubbub/dcn31/dcn31_hubbub.c:151 at dcn31_program_compbuf_size+0xd2/0x230 [amdgpu], CPU#2: Xorg/1207
kernel: Modules linked in: r8153_ecm hid_apple apple_mfi_fastcharge usbhid r8152 rfcomm snd_seq_dummy snd_hrtimer snd_seq snd_seq_device michael_mic 8021q garp stp mrp llc uhid algif_hash algif_skcipher af_alg qrtr_mhi bnep binfmt_misc nls_iso8859_1 nls_cp437 vfat fat edac_mce_amd edac_core ext4 mbcache jbd2 ucsi_acpi typec_ucsi roles amd_atl typec spd5118 kvm_amd snd_soc_ps_mach snd_soc_dmic snd_ps_pdm_dma regmap_i2c qrtr uvcvideo videobuf2_vmalloc kvm videobuf2_memops ath11k_pci uvc mhi irqbypass videobuf2_v4l2 ghash_clmulni_intel think_lmi snd_ctl_led snd_sof_amd_acp63 ath11k rapl cdc_mbim btusb snd_hda_codec_alc269 snd_sof_amd_acp qmi_helpers videodev cdc_wdm btintel snd_sof_xtensa_dsp snd_hda_scodec_component amdgpu snd_sof_pci firmware_attributes_class wmi_bmof cdc_ncm videobuf2_common snd_hda_codec_realtek_lib bluetooth mc cdc_ether mac80211 snd_hda_codec_generic snd_sof usbnet mii snd_sof_utils i2c_algo_bit libarc4 drm_ttm_helper snd_pci_ps snd_hda_codec_atihdmi k10temp ttm snd_soc_acpi_amd_match
kernel:  snd_hda_codec_hdmi agpgart snd_acp_config cfg80211 drm_exec snd_amd_sdw_acpi thunderbolt r8169 drm_suballoc_helper soundwire_amd mfd_core ac soundwire_generic_allocation drm_buddy snd_soc_acpi snd_hda_intel drm_panel_backlight_quirks snd_intel_dspcfg soundwire_bus amd_pmf gpu_sched snd_hda_codec amdtee amdxcp sp5100_tco snd_soc_core snd_hda_core amd_sfh watchdog drm_display_helper snd_compress snd_hwdep tpm_crb tee button cec snd_pcm crc16 i2c_piix4 mousedev snd_timer amd_pmc i2c_smbus nft_ct nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nft_limit sch_fq_codel parport_pc msr parport nf_tables fuse efi_pstore configfs nfnetlink efivarfs tpm libaescfb ecdh_generic autofs4 xfs dm_crypt dm_mod hid_multitouch hid_generic i2c_hid_acpi i2c_hid thinkpad_acpi input_leds nvme psmouse xhci_pci nvram sparse_keymap atkbd nvme_core xhci_hcd i2c_designware_platform platform_profile vivaldi_fmap evdev aesni_intel serio_raw snd nvme_keyring usbcore soundcore battery usb_common rfkill nvme_auth ccp hkdf rng_core fan thermal
kernel:  i2c_designware_core simpledrm drm_client_lib drm_shmem_helper drm_sysfb_helper drm_kms_helper drm
kernel: CPU: 2 UID: 0 PID: 1207 Comm: Xorg Not tainted 7.0.0-rc4 #1 PREEMPT
kernel: Hardware name: LENOVO 21MCS03F00/21MCS03F00, BIOS R2LET37W (1.18 ) 11/25/2025
kernel: RIP: 0010:dcn31_program_compbuf_size+0xd2/0x230 [amdgpu]
kernel: Code: 00 48 8b 43 28 8b 88 d8 01 00 00 48 8b 43 20 0f b6 50 76 48 8b 43 18 8b b0 14 01 00 00 e8 b6 d3 0b 00 85 c0 0f 85 32 01 00 00 <0f> 0b 48 8b 44 24 08 65 48 2b 05 6f 34 9f ce 0f 85 35 01 00 00 48
kernel: RSP: 0018:ffffd27ac33376e8 EFLAGS: 00010202
kernel: RAX: 0000000000000001 RBX: ffff8f3fa057c800 RCX: 0000000080041619
kernel: RDX: ffffd27ac510e5e8 RSI: 000000000000397a RDI: ffff8f3eda180000
kernel: RBP: 0000000000000004 R08: 0000000080041619 R09: 000000000000397a
kernel: R10: 0000000000000064 R11: ffffd27ac33373f8 R12: ffff8f3ec5c00000
kernel: R13: ffff8f3f61800000 R14: ffff8f3fa057c800 R15: 0000000000000004
kernel: FS:  00007f505be98b80(0000) GS:ffff8f466e70c000(0000) knlGS:0000000000000000
kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
kernel: CR2: 00007faaf46f5f40 CR3: 0000000106478000 CR4: 0000000000750ef0
kernel: PKRU: 55555554
kernel: Call Trace:
kernel:  <TASK>
kernel:  dcn20_optimize_bandwidth+0xe6/0x220 [amdgpu]
kernel:  dc_commit_state_no_check+0xc1e/0xe90 [amdgpu]
kernel:  dc_commit_streams+0x2f9/0x4f0 [amdgpu]
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? dm_read_reg_func+0x5c/0x100 [amdgpu]
kernel:  amdgpu_dm_atomic_commit_tail+0x51f/0x34c0 [amdgpu]
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? dcn30_internal_validate_bw+0x890/0x960 [amdgpu]
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? dcn314_validate_bandwidth+0xf7/0x2e0 [amdgpu]
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? dma_resv_get_fences+0xb2/0x290
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? dma_resv_get_singleton+0x43/0x130
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? wait_for_completion_timeout+0xf7/0x130
kernel:  commit_tail+0x9d/0x130 [drm_kms_helper]
kernel:  drm_atomic_helper_commit+0x13c/0x180 [drm_kms_helper]
kernel:  drm_atomic_commit+0xb1/0xe0 [drm]
kernel:  ? drm_plane_create_color_pipeline_property.cold+0x29/0x29 [drm]
kernel:  drm_atomic_connector_commit_dpms+0xfa/0x110 [drm]
kernel:  drm_mode_obj_set_property_ioctl+0x1b4/0x3b0 [drm]
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? drm_connector_set_obj_prop+0xa0/0xa0 [drm]
kernel:  drm_connector_property_set_ioctl+0x3c/0x60 [drm]
kernel:  drm_ioctl_kernel+0xae/0x100 [drm]
kernel:  drm_ioctl+0x29b/0x540 [drm]
kernel:  ? drm_connector_set_obj_prop+0xa0/0xa0 [drm]
kernel:  amdgpu_drm_ioctl+0x4a/0x80 [amdgpu]
kernel:  __x64_sys_ioctl+0x86/0xd0
kernel:  do_syscall_64+0xec/0x940
kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
kernel:  ? common_interrupt+0x3f/0xa0
kernel:  entry_SYSCALL_64_after_hwframe+0x55/0x5d
kernel: RIP: 0033:0x7f505c21dd3b
kernel: Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
kernel: RSP: 002b:00007ffda60126e0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
kernel: RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f505c21dd3b
kernel: RDX: 00007ffda6012770 RSI: 00000000c01064ab RDI: 000000000000000f
kernel: RBP: 00007ffda6012770 R08: 0000000000154eb5 R09: 7fffffffffffffff
kernel: R10: 00007f505bab9cb0 R11: 0000000000000246 R12: 00000000c01064ab
kernel: R13: 000000000000000f R14: 000055ba642c5c60 R15: 0000000000000000
kernel:  </TASK>
kernel: ---[ end trace 0000000000000000 ]---



Best regards,
Julian

-- 
 ()  ascii ribbon campaign - against html e-mail 
 /\                        - against proprietary attachments

Re: 7.0-rc4 dcn31_program_compbuf_size trace

[Eric Naim]

On 3/16/26 4:42 PM, Julian Wollrath wrote:
> Dear maintainers,
> 
> when turning the display of via 'xset dpms force off' I observe the
> following warning and trace under 7.0-rc4 with a AMD Ryzen 7 PRO 8840U
> w/ Radeon 780M Graphics card. This I did not observe under 6.19.7.
> 
> kernel: amdgpu 0000:c4:00.0: [drm] REG_WAIT timeout 1us * 100 tries - dcn31_program_compbuf_size line:142
> kernel: ------------[ cut here ]------------
> kernel: WARNING: drivers/gpu/drm/amd/amdgpu/../display/dc/hubbub/dcn31/dcn31_hubbub.c:151 at dcn31_program_compbuf_size+0xd2/0x230 [amdgpu], CPU#2: Xorg/1207
> kernel: Modules linked in: r8153_ecm hid_apple apple_mfi_fastcharge usbhid r8152 rfcomm snd_seq_dummy snd_hrtimer snd_seq snd_seq_device michael_mic 8021q garp stp mrp llc uhid algif_hash algif_skcipher af_alg qrtr_mhi bnep binfmt_misc nls_iso8859_1 nls_cp437 vfat fat edac_mce_amd edac_core ext4 mbcache jbd2 ucsi_acpi typec_ucsi roles amd_atl typec spd5118 kvm_amd snd_soc_ps_mach snd_soc_dmic snd_ps_pdm_dma regmap_i2c qrtr uvcvideo videobuf2_vmalloc kvm videobuf2_memops ath11k_pci uvc mhi irqbypass videobuf2_v4l2 ghash_clmulni_intel think_lmi snd_ctl_led snd_sof_amd_acp63 ath11k rapl cdc_mbim btusb snd_hda_codec_alc269 snd_sof_amd_acp qmi_helpers videodev cdc_wdm btintel snd_sof_xtensa_dsp snd_hda_scodec_component amdgpu snd_sof_pci firmware_attributes_class wmi_bmof cdc_ncm videobuf2_common snd_hda_codec_realtek_lib bluetooth mc cdc_ether mac80211 snd_hda_codec_generic snd_sof usbnet mii snd_sof_utils i2c_algo_bit libarc4 drm_ttm_helper snd_pci_ps snd_hda_codec_atihdmi k10temp ttm snd_soc_acpi_amd_match
> kernel:  snd_hda_codec_hdmi agpgart snd_acp_config cfg80211 drm_exec snd_amd_sdw_acpi thunderbolt r8169 drm_suballoc_helper soundwire_amd mfd_core ac soundwire_generic_allocation drm_buddy snd_soc_acpi snd_hda_intel drm_panel_backlight_quirks snd_intel_dspcfg soundwire_bus amd_pmf gpu_sched snd_hda_codec amdtee amdxcp sp5100_tco snd_soc_core snd_hda_core amd_sfh watchdog drm_display_helper snd_compress snd_hwdep tpm_crb tee button cec snd_pcm crc16 i2c_piix4 mousedev snd_timer amd_pmc i2c_smbus nft_ct nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nft_limit sch_fq_codel parport_pc msr parport nf_tables fuse efi_pstore configfs nfnetlink efivarfs tpm libaescfb ecdh_generic autofs4 xfs dm_crypt dm_mod hid_multitouch hid_generic i2c_hid_acpi i2c_hid thinkpad_acpi input_leds nvme psmouse xhci_pci nvram sparse_keymap atkbd nvme_core xhci_hcd i2c_designware_platform platform_profile vivaldi_fmap evdev aesni_intel serio_raw snd nvme_keyring usbcore soundcore battery usb_common rfkill nvme_auth ccp hkdf rng_core fan thermal
> kernel:  i2c_designware_core simpledrm drm_client_lib drm_shmem_helper drm_sysfb_helper drm_kms_helper drm
> kernel: CPU: 2 UID: 0 PID: 1207 Comm: Xorg Not tainted 7.0.0-rc4 #1 PREEMPT
> kernel: Hardware name: LENOVO 21MCS03F00/21MCS03F00, BIOS R2LET37W (1.18 ) 11/25/2025
> kernel: RIP: 0010:dcn31_program_compbuf_size+0xd2/0x230 [amdgpu]
> kernel: Code: 00 48 8b 43 28 8b 88 d8 01 00 00 48 8b 43 20 0f b6 50 76 48 8b 43 18 8b b0 14 01 00 00 e8 b6 d3 0b 00 85 c0 0f 85 32 01 00 00 <0f> 0b 48 8b 44 24 08 65 48 2b 05 6f 34 9f ce 0f 85 35 01 00 00 48
> kernel: RSP: 0018:ffffd27ac33376e8 EFLAGS: 00010202
> kernel: RAX: 0000000000000001 RBX: ffff8f3fa057c800 RCX: 0000000080041619
> kernel: RDX: ffffd27ac510e5e8 RSI: 000000000000397a RDI: ffff8f3eda180000
> kernel: RBP: 0000000000000004 R08: 0000000080041619 R09: 000000000000397a
> kernel: R10: 0000000000000064 R11: ffffd27ac33373f8 R12: ffff8f3ec5c00000
> kernel: R13: ffff8f3f61800000 R14: ffff8f3fa057c800 R15: 0000000000000004
> kernel: FS:  00007f505be98b80(0000) GS:ffff8f466e70c000(0000) knlGS:0000000000000000
> kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> kernel: CR2: 00007faaf46f5f40 CR3: 0000000106478000 CR4: 0000000000750ef0
> kernel: PKRU: 55555554
> kernel: Call Trace:
> kernel:  <TASK>
> kernel:  dcn20_optimize_bandwidth+0xe6/0x220 [amdgpu]
> kernel:  dc_commit_state_no_check+0xc1e/0xe90 [amdgpu]
> kernel:  dc_commit_streams+0x2f9/0x4f0 [amdgpu]
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? dm_read_reg_func+0x5c/0x100 [amdgpu]
> kernel:  amdgpu_dm_atomic_commit_tail+0x51f/0x34c0 [amdgpu]
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? dcn30_internal_validate_bw+0x890/0x960 [amdgpu]
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? dcn314_validate_bandwidth+0xf7/0x2e0 [amdgpu]
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? dma_resv_get_fences+0xb2/0x290
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? dma_resv_get_singleton+0x43/0x130
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? wait_for_completion_timeout+0xf7/0x130
> kernel:  commit_tail+0x9d/0x130 [drm_kms_helper]
> kernel:  drm_atomic_helper_commit+0x13c/0x180 [drm_kms_helper]
> kernel:  drm_atomic_commit+0xb1/0xe0 [drm]
> kernel:  ? drm_plane_create_color_pipeline_property.cold+0x29/0x29 [drm]
> kernel:  drm_atomic_connector_commit_dpms+0xfa/0x110 [drm]
> kernel:  drm_mode_obj_set_property_ioctl+0x1b4/0x3b0 [drm]
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? drm_connector_set_obj_prop+0xa0/0xa0 [drm]
> kernel:  drm_connector_property_set_ioctl+0x3c/0x60 [drm]
> kernel:  drm_ioctl_kernel+0xae/0x100 [drm]
> kernel:  drm_ioctl+0x29b/0x540 [drm]
> kernel:  ? drm_connector_set_obj_prop+0xa0/0xa0 [drm]
> kernel:  amdgpu_drm_ioctl+0x4a/0x80 [amdgpu]
> kernel:  __x64_sys_ioctl+0x86/0xd0
> kernel:  do_syscall_64+0xec/0x940
> kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
> kernel:  ? common_interrupt+0x3f/0xa0
> kernel:  entry_SYSCALL_64_after_hwframe+0x55/0x5d
> kernel: RIP: 0033:0x7f505c21dd3b
> kernel: Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
> kernel: RSP: 002b:00007ffda60126e0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
> kernel: RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f505c21dd3b
> kernel: RDX: 00007ffda6012770 RSI: 00000000c01064ab RDI: 000000000000000f
> kernel: RBP: 00007ffda6012770 R08: 0000000000154eb5 R09: 7fffffffffffffff
> kernel: R10: 00007f505bab9cb0 R11: 0000000000000246 R12: 00000000c01064ab
> kernel: R13: 000000000000000f R14: 000055ba642c5c60 R15: 0000000000000000
> kernel:  </TASK>
> kernel: ---[ end trace 0000000000000000 ]---
> 
> 
> 
> Best regards,
> Julian
> 

Hi maintainers,

Per [1], this seems to be caused by 592c5b80110d5e9e50873b5364818cb6f401e26d.

[1] https://github.com/CachyOS/linux-cachyos/issues/810#issuecomment-4275280460


-- 
Regards,
  Eric

Re: 7.0-rc4 dcn31_program_compbuf_size trace

[Minxi Hou]

On Mon, 16 Mar 2026 09:42:32 +0100, Julian Wollrath wrote:
> when turning the display off via 'xset dpms force off' I observe the
> following warning and trace under 7.0-rc4 with a AMD Ryzen 7 PRO 8840U

I can reproduce the same warning on different hardware with a different
trigger.

Machine: HP Z66 G6 14 (ProBook 4 G1ah), AMD Ryzen 7 H 255 (HawkPoint1),
kernel 7.0.8-200.fc44.x86_64.

Trigger: HDMI hotplug with an external 4K monitor. The Z66's HDMI port
uses a DP-to-HDMI protocol converter with FRL support (dmesg shows
"DP-HDMI FRL PCON supported"). Connecting the monitor at boot causes
the first timeout during initialization; subsequent hotplug events
repeat it at runtime.

The call trace matches yours exactly:
  dcn31_program_compbuf_size line:141
  dcn20_optimize_bandwidth
  dc_commit_state_no_check
  dc_commit_streams
  amdgpu_dm_atomic_commit_tail

Not seen with 6.19.x. CachyOS community bisected this to:

  commit 592c5b80110d5e9e50873b5364818cb6f401e26d
  "drm/amd/display: Migrate HUBBUB register access from hwseq to hubbub component"

One thing I noticed on my setup: the amdgpu driver defaults to
max_bpc=16 for the HDMI connector. With a HDMI 2.0 monitor (BenQ
EL2870U, Max TMDS 340 MHz) at 4K@60Hz, the higher color depth
increases bandwidth demand during compbuf reprogramming and makes the
timeout happen more often. Adding amdgpu.deep_color=0 helps but does
not fix the root cause.

Partial workaround:
  amdgpu.runpm=0 amdgpu.gfx_off=0 amdgpu.deep_color=0

A patch proposed by Giuseppe Ranieri in the CachyOS issue tracker [1]
changes the REG_WAIT parameters from 1us*100 to 1000us*30 and resolves
the warning for several users with DCN 3.1 hardware. Discussion there
is ongoing about whether 2-10us*1000 would be a better fit.

[1] https://github.com/CachyOS/linux-cachyos/issues/810

Minxi Hou

Re: 7.0-rc4 dcn31_program_compbuf_size trace

[Thorsten Leemhuis]

[+a few people]

On 5/18/26 10:08, Minxi Hou wrote:
> On Mon, 16 Mar 2026 09:42:32 +0100, Julian Wollrath wrote:
>> when turning the display off via 'xset dpms force off' I observe the
>> following warning and trace under 7.0-rc4 with a AMD Ryzen 7 PRO 8840U
> 
> I can reproduce the same warning on different hardware with a different
> trigger.
> 
> Machine: HP Z66 G6 14 (ProBook 4 G1ah), AMD Ryzen 7 H 255 (HawkPoint1),
> kernel 7.0.8-200.fc44.x86_64.
> 
> Trigger: HDMI hotplug with an external 4K monitor. The Z66's HDMI port
> uses a DP-to-HDMI protocol converter with FRL support (dmesg shows
> "DP-HDMI FRL PCON supported"). Connecting the monitor at boot causes
> the first timeout during initialization; subsequent hotplug events
> repeat it at runtime.
> 
> The call trace matches yours exactly:
>   dcn31_program_compbuf_size line:141
>   dcn20_optimize_bandwidth
>   dc_commit_state_no_check
>   dc_commit_streams
>   amdgpu_dm_atomic_commit_tail
> 
> Not seen with 6.19.x. CachyOS community bisected this to:
> 
>   commit 592c5b80110d5e9e50873b5364818cb6f401e26d
>   "drm/amd/display: Migrate HUBBUB register access from hwseq to hubbub component"

Antonio posted a RFC fix for that comment here with a reference to a
warning that on a *very quick* look looks similar:
https://lore.kernel.org/all/20260519144509.2646680-1-antonio@mandelbit.com/

Wondering if we need a ticket here to get this resolved:
https://gitlab.freedesktop.org/drm/amd/-/work_items/

But from the reply to Antonio's patch is sounds like it's not needed.

Ciao, Thorsten

> One thing I noticed on my setup: the amdgpu driver defaults to
> max_bpc=16 for the HDMI connector. With a HDMI 2.0 monitor (BenQ
> EL2870U, Max TMDS 340 MHz) at 4K@60Hz, the higher color depth
> increases bandwidth demand during compbuf reprogramming and makes the
> timeout happen more often. Adding amdgpu.deep_color=0 helps but does
> not fix the root cause.
> 
> Partial workaround:
>   amdgpu.runpm=0 amdgpu.gfx_off=0 amdgpu.deep_color=0
> 
> A patch proposed by Giuseppe Ranieri in the CachyOS issue tracker [1]
> changes the REG_WAIT parameters from 1us*100 to 1000us*30 and resolves
> the warning for several users with DCN 3.1 hardware. Discussion there
> is ongoing about whether 2-10us*1000 would be a better fit.
> 
> [1] https://github.com/CachyOS/linux-cachyos/issues/810
> 
> Minxi Hou

RE: 7.0-rc4 dcn31_program_compbuf_size trace

[Deucher, Alexander]

AMD General

+ Harry, Leo

> -----Original Message-----
> From: Thorsten Leemhuis <regressions@leemhuis.info>
> Sent: Wednesday, May 20, 2026 4:59 AM
> To: Minxi Hou <houminxi@gmail.com>; jwollrath@web.de
> Cc: amd-gfx@lists.freedesktop.org; Eric Naim <dnaim@cachyos.org>; Pinninti,
> Bhuvana Chandra <BhuvanaChandra.Pinninti@amd.com>; Linux kernel
> regressions list <regressions@lists.linux.dev>; Deucher, Alexander
> <Alexander.Deucher@amd.com>; Antonio Quartulli
> <antonio@mandelbit.com>; Koenig, Christian <Christian.Koenig@amd.com>
> Subject: Re: 7.0-rc4 dcn31_program_compbuf_size trace
>
> [+a few people]
>
> On 5/18/26 10:08, Minxi Hou wrote:
> > On Mon, 16 Mar 2026 09:42:32 +0100, Julian Wollrath wrote:
> >> when turning the display off via 'xset dpms force off' I observe the
> >> following warning and trace under 7.0-rc4 with a AMD Ryzen 7 PRO
> >> 8840U
> >
> > I can reproduce the same warning on different hardware with a
> > different trigger.
> >
> > Machine: HP Z66 G6 14 (ProBook 4 G1ah), AMD Ryzen 7 H 255
> > (HawkPoint1), kernel 7.0.8-200.fc44.x86_64.
> >
> > Trigger: HDMI hotplug with an external 4K monitor. The Z66's HDMI port
> > uses a DP-to-HDMI protocol converter with FRL support (dmesg shows
> > "DP-HDMI FRL PCON supported"). Connecting the monitor at boot causes
> > the first timeout during initialization; subsequent hotplug events
> > repeat it at runtime.
> >
> > The call trace matches yours exactly:
> >   dcn31_program_compbuf_size line:141
> >   dcn20_optimize_bandwidth
> >   dc_commit_state_no_check
> >   dc_commit_streams
> >   amdgpu_dm_atomic_commit_tail
> >
> > Not seen with 6.19.x. CachyOS community bisected this to:
> >
> >   commit 592c5b80110d5e9e50873b5364818cb6f401e26d
> >   "drm/amd/display: Migrate HUBBUB register access from hwseq to hubbub
> component"
>
> Antonio posted a RFC fix for that comment here with a reference to a warning
> that on a *very quick* look looks similar:
> https://lore.kernel.org/all/20260519144509.2646680-1-
> antonio@mandelbit.com/
>
> Wondering if we need a ticket here to get this resolved:
> https://gitlab.freedesktop.org/drm/amd/-/work_items/
>
> But from the reply to Antonio's patch is sounds like it's not needed.
>
> Ciao, Thorsten
>
> > One thing I noticed on my setup: the amdgpu driver defaults to
> > max_bpc=16 for the HDMI connector. With a HDMI 2.0 monitor (BenQ
> > EL2870U, Max TMDS 340 MHz) at 4K@60Hz, the higher color depth
> > increases bandwidth demand during compbuf reprogramming and makes
> the
> > timeout happen more often. Adding amdgpu.deep_color=0 helps but does
> > not fix the root cause.
> >
> > Partial workaround:
> >   amdgpu.runpm=0 amdgpu.gfx_off=0 amdgpu.deep_color=0
> >
> > A patch proposed by Giuseppe Ranieri in the CachyOS issue tracker [1]
> > changes the REG_WAIT parameters from 1us*100 to 1000us*30 and
> resolves
> > the warning for several users with DCN 3.1 hardware. Discussion there
> > is ongoing about whether 2-10us*1000 would be a better fit.
> >
> > [1] https://github.com/CachyOS/linux-cachyos/issues/810
> >
> > Minxi Hou