* [Intel-gfx] [bisected] suspend broken by DRM fbdev name change on i915 IVB
@ 2021-10-07 7:25 ` Johannes Stezenbach
0 siblings, 0 replies; 5+ messages in thread
From: Johannes Stezenbach @ 2021-10-07 7:25 UTC (permalink / raw)
To: Javier Martinez Canillas; +Cc: intel-gfx, dri-devel
Hi,
yesterday I updated the kernel from 5.13.7 to 5.14.9 and found
it broke suspend-to-RAM. The machine displays a few messages on
text console after resume but hangs when switching to X11.
The hardware: Asus P8H77-V with Intel Core i5-3550 CPU,
display connected via DP.
DP1 connected primary 1920x1200+0+0 (normal left inverted right x axis y axis) 520mm x 320mm
1920x1200 59.95*+
I bisected it and the offending commit is the totally unlikely
and innocent looking
commit b3484d2b03e4c940a9598aa841a52d69729c582a
Author: Javier Martinez Canillas <javierm@redhat.com>
Date: Tue May 25 17:13:13 2021 +0200
drm/fb-helper: improve DRM fbdev emulation device names
Now I'm running 5.14.9 with this commit reverted and
suspend works.
In /var/log/kern.log I found this on suspend:
[ 34.002252][ T3455] ------------[ cut here ]------------
[ 34.002256][ T3455] i915 0000:00:02.0: drm_WARN_ON((intel_de_read(dev_priv, intel_dp->output_reg) & (1 << 31)) == 0)
[ 34.002274][ T3455] WARNING: CPU: 0 PID: 3455 at drivers/gpu/drm/i915/display/g4x_dp.c:431 intel_dp_link_down.isra.0+0x2e7/0x390
[ 34.002285][ T3455] Modules linked in: kvm_intel kvm irqbypass ehci_pci xhci_pci ehci_hcd xhci_hcd
[ 34.002304][ T3455] CPU: 0 PID: 3455 Comm: kworker/u8:27 Not tainted 5.14.9 #29
[ 34.002309][ T3455] Hardware name: System manufacturer System Product Name/P8H77-V, BIOS 1905 10/27/2014
[ 34.002312][ T3455] Workqueue: events_unbound async_run_entry_fn
[ 34.002320][ T3455] RIP: 0010:intel_dp_link_down.isra.0+0x2e7/0x390
[ 34.002326][ T3455] Code: 4c 8b 67 50 4d 85 e4 75 03 4c 8b 27 e8 d2 19 05 00 48 c7 c1 e8 53 8c 88 4c 89 e2 48 c7 c7 7b 83 89 88 48 89 c6 e8 d1 d0 52 00 <0f> 0b 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e 41 5f c3 49 8b 7d 08 4c
[ 34.002330][ T3455] RSP: 0018:ffffaa1dc1be3a88 EFLAGS: 00010282
[ 34.002335][ T3455] RAX: 0000000000000000 RBX: ffffa15a45f28000 RCX: 0000000000000000
[ 34.002338][ T3455] RDX: 0000000000000001 RSI: ffffffff8714af2f RDI: ffffffff8714af2f
[ 34.002341][ T3455] RBP: 0000000080180344 R08: 00000007eab15173 R09: 0000000000000001
[ 34.002344][ T3455] R10: 0000000000080000 R11: 0000000000000000 R12: ffffa15a40f87180
[ 34.002347][ T3455] R13: ffffa15a464e0000 R14: ffffa15a593c6000 R15: 0000000000000001
[ 34.002350][ T3455] FS: 0000000000000000(0000) GS:ffffa15d4f800000(0000) knlGS:0000000000000000
[ 34.002354][ T3455] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 34.002357][ T3455] CR2: 000056000d53cf60 CR3: 000000015fa2a006 CR4: 00000000001706f0
[ 34.002361][ T3455] Call Trace:
[ 34.002373][ T3455] g4x_post_disable_dp+0x2e/0x110
[ 34.002380][ T3455] intel_encoders_post_disable+0x73/0x80
[ 34.002391][ T3455] ilk_crtc_disable+0x96/0x3a0
[ 34.002402][ T3455] intel_old_crtc_state_disables+0x5c/0x110
[ 34.002412][ T3455] intel_atomic_commit_tail+0xdcc/0x1410
[ 34.002434][ T3455] intel_atomic_commit+0x332/0x3b0
[ 34.002443][ T3455] drm_atomic_helper_disable_all+0x175/0x190
[ 34.002453][ T3455] drm_atomic_helper_suspend+0xa6/0x200
[ 34.002474][ T3455] intel_display_suspend+0x23/0x50
[ 34.002480][ T3455] i915_drm_suspend+0x42/0xe0
[ 34.002488][ T3455] pci_pm_suspend+0x74/0x160
[ 34.002496][ T3455] ? pci_pm_freeze+0xb0/0xb0
[ 34.002500][ T3455] dpm_run_callback+0x6f/0x170
[ 34.002512][ T3455] __device_suspend+0x110/0x4b0
[ 34.002521][ T3455] async_suspend+0x1b/0x90
[ 34.002529][ T3455] async_run_entry_fn+0x2e/0x110
[ 34.002535][ T3455] process_one_work+0x2c9/0x550
[ 34.002551][ T3455] worker_thread+0x4f/0x3e0
[ 34.002556][ T3455] ? rescuer_thread+0x340/0x340
[ 34.002563][ T3455] kthread+0x14a/0x170
[ 34.002569][ T3455] ? set_kthread_struct+0x40/0x40
[ 34.002578][ T3455] ret_from_fork+0x22/0x30
[ 34.002600][ T3455] ---[ end trace 2058ff589e8cbd78 ]---
and after resume:
[ 44.513409][ C0] i915 0000:00:02.0: [drm] *ERROR* uncleared pch fifo underrun on pch transcoder A
[ 44.514557][ C0] i915 0000:00:02.0: [drm] *ERROR* PCH transcoder A FIFO underrun
[ 54.997256][ T3520] [drm:drm_atomic_helper_wait_for_flip_done] *ERROR* [CRTC:45:pipe A] flip_done timed out
Scratching my head about it I found one hint in
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler:
using_kms() { grep -q -E '(nouveau|drm)fb' /proc/fb; }
So the ABI change in /proc/fb causes the pm-utils scripts to
skip the --quirk-no-chvt and apply other quirks,
/var/log/pm-suspend.log says:
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend:
No quirk database entry for this system, using default.
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend: success.
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend:
Saving last known working quirks: --quirk-vbe-post --quirk-dpms-on
--quirk-dpms-suspend --quirk-vbestate-restore
--quirk-vbemode-restore --quirk-vga-mode-3
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend: success.
whereas in the normal case it's
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend:
Kernel modesetting video driver detected, not using quirks.
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend: success.
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend:
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend: success.
Johannes
^ permalink raw reply [flat|nested] 5+ messages in thread
* [bisected] suspend broken by DRM fbdev name change on i915 IVB
@ 2021-10-07 7:25 ` Johannes Stezenbach
0 siblings, 0 replies; 5+ messages in thread
From: Johannes Stezenbach @ 2021-10-07 7:25 UTC (permalink / raw)
To: Javier Martinez Canillas; +Cc: intel-gfx, dri-devel
Hi,
yesterday I updated the kernel from 5.13.7 to 5.14.9 and found
it broke suspend-to-RAM. The machine displays a few messages on
text console after resume but hangs when switching to X11.
The hardware: Asus P8H77-V with Intel Core i5-3550 CPU,
display connected via DP.
DP1 connected primary 1920x1200+0+0 (normal left inverted right x axis y axis) 520mm x 320mm
1920x1200 59.95*+
I bisected it and the offending commit is the totally unlikely
and innocent looking
commit b3484d2b03e4c940a9598aa841a52d69729c582a
Author: Javier Martinez Canillas <javierm@redhat.com>
Date: Tue May 25 17:13:13 2021 +0200
drm/fb-helper: improve DRM fbdev emulation device names
Now I'm running 5.14.9 with this commit reverted and
suspend works.
In /var/log/kern.log I found this on suspend:
[ 34.002252][ T3455] ------------[ cut here ]------------
[ 34.002256][ T3455] i915 0000:00:02.0: drm_WARN_ON((intel_de_read(dev_priv, intel_dp->output_reg) & (1 << 31)) == 0)
[ 34.002274][ T3455] WARNING: CPU: 0 PID: 3455 at drivers/gpu/drm/i915/display/g4x_dp.c:431 intel_dp_link_down.isra.0+0x2e7/0x390
[ 34.002285][ T3455] Modules linked in: kvm_intel kvm irqbypass ehci_pci xhci_pci ehci_hcd xhci_hcd
[ 34.002304][ T3455] CPU: 0 PID: 3455 Comm: kworker/u8:27 Not tainted 5.14.9 #29
[ 34.002309][ T3455] Hardware name: System manufacturer System Product Name/P8H77-V, BIOS 1905 10/27/2014
[ 34.002312][ T3455] Workqueue: events_unbound async_run_entry_fn
[ 34.002320][ T3455] RIP: 0010:intel_dp_link_down.isra.0+0x2e7/0x390
[ 34.002326][ T3455] Code: 4c 8b 67 50 4d 85 e4 75 03 4c 8b 27 e8 d2 19 05 00 48 c7 c1 e8 53 8c 88 4c 89 e2 48 c7 c7 7b 83 89 88 48 89 c6 e8 d1 d0 52 00 <0f> 0b 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e 41 5f c3 49 8b 7d 08 4c
[ 34.002330][ T3455] RSP: 0018:ffffaa1dc1be3a88 EFLAGS: 00010282
[ 34.002335][ T3455] RAX: 0000000000000000 RBX: ffffa15a45f28000 RCX: 0000000000000000
[ 34.002338][ T3455] RDX: 0000000000000001 RSI: ffffffff8714af2f RDI: ffffffff8714af2f
[ 34.002341][ T3455] RBP: 0000000080180344 R08: 00000007eab15173 R09: 0000000000000001
[ 34.002344][ T3455] R10: 0000000000080000 R11: 0000000000000000 R12: ffffa15a40f87180
[ 34.002347][ T3455] R13: ffffa15a464e0000 R14: ffffa15a593c6000 R15: 0000000000000001
[ 34.002350][ T3455] FS: 0000000000000000(0000) GS:ffffa15d4f800000(0000) knlGS:0000000000000000
[ 34.002354][ T3455] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 34.002357][ T3455] CR2: 000056000d53cf60 CR3: 000000015fa2a006 CR4: 00000000001706f0
[ 34.002361][ T3455] Call Trace:
[ 34.002373][ T3455] g4x_post_disable_dp+0x2e/0x110
[ 34.002380][ T3455] intel_encoders_post_disable+0x73/0x80
[ 34.002391][ T3455] ilk_crtc_disable+0x96/0x3a0
[ 34.002402][ T3455] intel_old_crtc_state_disables+0x5c/0x110
[ 34.002412][ T3455] intel_atomic_commit_tail+0xdcc/0x1410
[ 34.002434][ T3455] intel_atomic_commit+0x332/0x3b0
[ 34.002443][ T3455] drm_atomic_helper_disable_all+0x175/0x190
[ 34.002453][ T3455] drm_atomic_helper_suspend+0xa6/0x200
[ 34.002474][ T3455] intel_display_suspend+0x23/0x50
[ 34.002480][ T3455] i915_drm_suspend+0x42/0xe0
[ 34.002488][ T3455] pci_pm_suspend+0x74/0x160
[ 34.002496][ T3455] ? pci_pm_freeze+0xb0/0xb0
[ 34.002500][ T3455] dpm_run_callback+0x6f/0x170
[ 34.002512][ T3455] __device_suspend+0x110/0x4b0
[ 34.002521][ T3455] async_suspend+0x1b/0x90
[ 34.002529][ T3455] async_run_entry_fn+0x2e/0x110
[ 34.002535][ T3455] process_one_work+0x2c9/0x550
[ 34.002551][ T3455] worker_thread+0x4f/0x3e0
[ 34.002556][ T3455] ? rescuer_thread+0x340/0x340
[ 34.002563][ T3455] kthread+0x14a/0x170
[ 34.002569][ T3455] ? set_kthread_struct+0x40/0x40
[ 34.002578][ T3455] ret_from_fork+0x22/0x30
[ 34.002600][ T3455] ---[ end trace 2058ff589e8cbd78 ]---
and after resume:
[ 44.513409][ C0] i915 0000:00:02.0: [drm] *ERROR* uncleared pch fifo underrun on pch transcoder A
[ 44.514557][ C0] i915 0000:00:02.0: [drm] *ERROR* PCH transcoder A FIFO underrun
[ 54.997256][ T3520] [drm:drm_atomic_helper_wait_for_flip_done] *ERROR* [CRTC:45:pipe A] flip_done timed out
Scratching my head about it I found one hint in
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler:
using_kms() { grep -q -E '(nouveau|drm)fb' /proc/fb; }
So the ABI change in /proc/fb causes the pm-utils scripts to
skip the --quirk-no-chvt and apply other quirks,
/var/log/pm-suspend.log says:
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend:
No quirk database entry for this system, using default.
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend: success.
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend:
Saving last known working quirks: --quirk-vbe-post --quirk-dpms-on
--quirk-dpms-suspend --quirk-vbestate-restore
--quirk-vbemode-restore --quirk-vga-mode-3
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend: success.
whereas in the normal case it's
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend:
Kernel modesetting video driver detected, not using quirks.
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler suspend suspend: success.
Running hook /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend:
/usr/lib/pm-utils/sleep.d/98video-quirk-db-handler resume suspend: success.
Johannes
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Intel-gfx] [bisected] suspend broken by DRM fbdev name change on i915 IVB
2021-10-07 7:25 ` Johannes Stezenbach
(?)
@ 2021-10-07 12:38 ` Ville Syrjälä
2021-10-07 13:01 ` Javier Martinez Canillas
-1 siblings, 1 reply; 5+ messages in thread
From: Ville Syrjälä @ 2021-10-07 12:38 UTC (permalink / raw)
To: Johannes Stezenbach; +Cc: Javier Martinez Canillas, intel-gfx, dri-devel
On Thu, Oct 07, 2021 at 09:25:58AM +0200, Johannes Stezenbach wrote:
> Hi,
>
> yesterday I updated the kernel from 5.13.7 to 5.14.9 and found
> it broke suspend-to-RAM. The machine displays a few messages on
> text console after resume but hangs when switching to X11.
>
> The hardware: Asus P8H77-V with Intel Core i5-3550 CPU,
> display connected via DP.
>
> DP1 connected primary 1920x1200+0+0 (normal left inverted right x axis y axis) 520mm x 320mm
> 1920x1200 59.95*+
>
> I bisected it and the offending commit is the totally unlikely
> and innocent looking
>
> commit b3484d2b03e4c940a9598aa841a52d69729c582a
> Author: Javier Martinez Canillas <javierm@redhat.com>
> Date: Tue May 25 17:13:13 2021 +0200
>
> drm/fb-helper: improve DRM fbdev emulation device names
>
> Now I'm running 5.14.9 with this commit reverted and
> suspend works.
>
> In /var/log/kern.log I found this on suspend:
>
> [ 34.002252][ T3455] ------------[ cut here ]------------
> [ 34.002256][ T3455] i915 0000:00:02.0: drm_WARN_ON((intel_de_read(dev_priv, intel_dp->output_reg) & (1 << 31)) == 0)
> [ 34.002274][ T3455] WARNING: CPU: 0 PID: 3455 at drivers/gpu/drm/i915/display/g4x_dp.c:431 intel_dp_link_down.isra.0+0x2e7/0x390
> [ 34.002285][ T3455] Modules linked in: kvm_intel kvm irqbypass ehci_pci xhci_pci ehci_hcd xhci_hcd
> [ 34.002304][ T3455] CPU: 0 PID: 3455 Comm: kworker/u8:27 Not tainted 5.14.9 #29
> [ 34.002309][ T3455] Hardware name: System manufacturer System Product Name/P8H77-V, BIOS 1905 10/27/2014
> [ 34.002312][ T3455] Workqueue: events_unbound async_run_entry_fn
> [ 34.002320][ T3455] RIP: 0010:intel_dp_link_down.isra.0+0x2e7/0x390
> [ 34.002326][ T3455] Code: 4c 8b 67 50 4d 85 e4 75 03 4c 8b 27 e8 d2 19 05 00 48 c7 c1 e8 53 8c 88 4c 89 e2 48 c7 c7 7b 83 89 88 48 89 c6 e8 d1 d0 52 00 <0f> 0b 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e 41 5f c3 49 8b 7d 08 4c
> [ 34.002330][ T3455] RSP: 0018:ffffaa1dc1be3a88 EFLAGS: 00010282
> [ 34.002335][ T3455] RAX: 0000000000000000 RBX: ffffa15a45f28000 RCX: 0000000000000000
> [ 34.002338][ T3455] RDX: 0000000000000001 RSI: ffffffff8714af2f RDI: ffffffff8714af2f
> [ 34.002341][ T3455] RBP: 0000000080180344 R08: 00000007eab15173 R09: 0000000000000001
> [ 34.002344][ T3455] R10: 0000000000080000 R11: 0000000000000000 R12: ffffa15a40f87180
> [ 34.002347][ T3455] R13: ffffa15a464e0000 R14: ffffa15a593c6000 R15: 0000000000000001
> [ 34.002350][ T3455] FS: 0000000000000000(0000) GS:ffffa15d4f800000(0000) knlGS:0000000000000000
> [ 34.002354][ T3455] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 34.002357][ T3455] CR2: 000056000d53cf60 CR3: 000000015fa2a006 CR4: 00000000001706f0
> [ 34.002361][ T3455] Call Trace:
> [ 34.002373][ T3455] g4x_post_disable_dp+0x2e/0x110
> [ 34.002380][ T3455] intel_encoders_post_disable+0x73/0x80
> [ 34.002391][ T3455] ilk_crtc_disable+0x96/0x3a0
> [ 34.002402][ T3455] intel_old_crtc_state_disables+0x5c/0x110
> [ 34.002412][ T3455] intel_atomic_commit_tail+0xdcc/0x1410
> [ 34.002434][ T3455] intel_atomic_commit+0x332/0x3b0
> [ 34.002443][ T3455] drm_atomic_helper_disable_all+0x175/0x190
> [ 34.002453][ T3455] drm_atomic_helper_suspend+0xa6/0x200
> [ 34.002474][ T3455] intel_display_suspend+0x23/0x50
> [ 34.002480][ T3455] i915_drm_suspend+0x42/0xe0
> [ 34.002488][ T3455] pci_pm_suspend+0x74/0x160
> [ 34.002496][ T3455] ? pci_pm_freeze+0xb0/0xb0
> [ 34.002500][ T3455] dpm_run_callback+0x6f/0x170
> [ 34.002512][ T3455] __device_suspend+0x110/0x4b0
> [ 34.002521][ T3455] async_suspend+0x1b/0x90
> [ 34.002529][ T3455] async_run_entry_fn+0x2e/0x110
> [ 34.002535][ T3455] process_one_work+0x2c9/0x550
> [ 34.002551][ T3455] worker_thread+0x4f/0x3e0
> [ 34.002556][ T3455] ? rescuer_thread+0x340/0x340
> [ 34.002563][ T3455] kthread+0x14a/0x170
> [ 34.002569][ T3455] ? set_kthread_struct+0x40/0x40
> [ 34.002578][ T3455] ret_from_fork+0x22/0x30
> [ 34.002600][ T3455] ---[ end trace 2058ff589e8cbd78 ]---
>
> and after resume:
>
> [ 44.513409][ C0] i915 0000:00:02.0: [drm] *ERROR* uncleared pch fifo underrun on pch transcoder A
> [ 44.514557][ C0] i915 0000:00:02.0: [drm] *ERROR* PCH transcoder A FIFO underrun
> [ 54.997256][ T3520] [drm:drm_atomic_helper_wait_for_flip_done] *ERROR* [CRTC:45:pipe A] flip_done timed out
>
>
> Scratching my head about it I found one hint in
> /usr/lib/pm-utils/sleep.d/98video-quirk-db-handler:
>
> using_kms() { grep -q -E '(nouveau|drm)fb' /proc/fb; }
>
> So the ABI change in /proc/fb causes the pm-utils scripts to
> skip the --quirk-no-chvt and apply other quirks,
> /var/log/pm-suspend.log says:
Nasty. This pm-utils quirk stuff really has no business existing IMO,
and so I recommend nuking pm-utils from your system as soon as possible.
Back when I still had it on my machines (due to some silly dependency
I think), I just created empty override files in /etc/pm/ to permanently
disable all the quirks.
But as long people might be using it I guess we need some kind of
revert/fix to put the "drmfb" back into the name. Javier?
--
Ville Syrjälä
Intel
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Intel-gfx] [bisected] suspend broken by DRM fbdev name change on i915 IVB
2021-10-07 12:38 ` [Intel-gfx] " Ville Syrjälä
@ 2021-10-07 13:01 ` Javier Martinez Canillas
2021-10-07 13:04 ` Johannes Stezenbach
0 siblings, 1 reply; 5+ messages in thread
From: Javier Martinez Canillas @ 2021-10-07 13:01 UTC (permalink / raw)
To: Ville Syrjälä, Johannes Stezenbach; +Cc: intel-gfx, dri-devel
Hello,
On 10/7/21 14:38, Ville Syrjälä wrote:
[snip]
>>
>> So the ABI change in /proc/fb causes the pm-utils scripts to
>> skip the --quirk-no-chvt and apply other quirks,
>> /var/log/pm-suspend.log says:
>
> Nasty. This pm-utils quirk stuff really has no business existing IMO,
> and so I recommend nuking pm-utils from your system as soon as possible.
> Back when I still had it on my machines (due to some silly dependency
> I think), I just created empty override files in /etc/pm/ to permanently
> disable all the quirks.
>
> But as long people might be using it I guess we need some kind of
> revert/fix to put the "drmfb" back into the name. Javier?
>
Yes, the change was just cosmetic because we had confusing names such as
"simpledrmdrmfb". When it was proposed, the agreement was that /proc/fb
shouldn't be considered an ABI but now we found that people are using it.
So I agree that would be better to revert this patch. Johannes, will you
post a revert or do you want me to do it ?
Best regards,
--
Javier Martinez Canillas
Linux Engineering
Red Hat
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Intel-gfx] [bisected] suspend broken by DRM fbdev name change on i915 IVB
2021-10-07 13:01 ` Javier Martinez Canillas
@ 2021-10-07 13:04 ` Johannes Stezenbach
0 siblings, 0 replies; 5+ messages in thread
From: Johannes Stezenbach @ 2021-10-07 13:04 UTC (permalink / raw)
To: Javier Martinez Canillas; +Cc: Ville Syrjälä, intel-gfx, dri-devel
Hi Javier,
On Thu, Oct 07, 2021 at 03:01:46PM +0200, Javier Martinez Canillas wrote:
> Yes, the change was just cosmetic because we had confusing names such as
> "simpledrmdrmfb". When it was proposed, the agreement was that /proc/fb
> shouldn't be considered an ABI but now we found that people are using it.
>
> So I agree that would be better to revert this patch. Johannes, will you
> post a revert or do you want me to do it ?
Please do it.
Thanks,
Johannes
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-10-07 13:05 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-07 7:25 [Intel-gfx] [bisected] suspend broken by DRM fbdev name change on i915 IVB Johannes Stezenbach
2021-10-07 7:25 ` Johannes Stezenbach
2021-10-07 12:38 ` [Intel-gfx] " Ville Syrjälä
2021-10-07 13:01 ` Javier Martinez Canillas
2021-10-07 13:04 ` Johannes Stezenbach
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.