[CRTC:24] vblank wait timed out

[Philipp Zabel]

Hi Martyn,

On Tue, 2017-03-21 at 09:50 +0000, Martyn Welch wrote:
> I have an i.MX6 platform with 2 display port interfaces, one driven by the
> HDMI interface, the other by LVDS, both via bridges. We are currently
> experiencing the following error when we boot with the monitor connected
> to the LVDS backed interface and then connect a monitor to the HDMI backed
> interface after boot:
> 
> Mar 20 18:15:23 GE00409729044C kernel: ------------[ cut here ]------------
> Mar 20 18:15:23 GE00409729044C kernel: WARNING: CPU: 1 PID: 85 at /home/martyn/build-helix/tmp/work-shared/csmon/kernel-source/drivers/gpu/drm/drm_atomic_helper.c:1121 drm_atomic_helper_wait_for_vblanks+0x264/0x274
> Mar 20 18:15:23 GE00409729044C kernel: [CRTC:24] vblank wait timed out
> Mar 20 18:15:23 GE00409729044C kernel: Modules linked in: bonding snd_usb_audio snd_hwdep snd_usbmidi_lib snd_rawmidi cp210x usbserial atmel_mxt_ts
> Mar 20 18:15:23 GE00409729044C kernel: CPU: 1 PID: 85 Comm: kworker/u4:1 Not tainted 4.8.0 #4
> Mar 20 18:15:23 GE00409729044C kernel: Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)
> Mar 20 18:15:23 GE00409729044C kernel: Workqueue: events_unbound commit_work
> Mar 20 18:15:23 GE00409729044C kernel: Backtrace:
> Mar 20 18:15:23 GE00409729044C kernel: [<8010c968>] (dump_backtrace) from [<8010cbb0>] (show_stack+0x20/0x24)
> Mar 20 18:15:23 GE00409729044C kernel:  r7:00000000 r6:80d2bf98 r5:600b0013 r4:00000000
> Mar 20 18:15:23 GE00409729044C kernel: [<8010cb90>] (show_stack) from [<803c0e68>] (dump_stack+0x98/0xb4)
> Mar 20 18:15:23 GE00409729044C kernel: [<803c0dd0>] (dump_stack) from [<80122abc>] (__warn+0xe4/0x110)
> Mar 20 18:15:23 GE00409729044C kernel:  r7:00000009 r6:80a8d490 r5:00000000 r4:ee173e10
> Mar 20 18:15:23 GE00409729044C kernel: [<801229d8>] (__warn) from [<80122b2c>] (warn_slowpath_fmt+0x44/0x4c)
> Mar 20 18:15:23 GE00409729044C kernel:  r9:ee1e5418 r8:00000000 r7:00000000 r6:00000000 r5:ecc04f00 r4:80a8d5ec
> Mar 20 18:15:23 GE00409729044C kernel: [<80122aec>] (warn_slowpath_fmt) from [<80486ce0>] (drm_atomic_helper_wait_for_vblanks+0x264/0x274)
> Mar 20 18:15:23 GE00409729044C kernel:  r3:00000018 r2:80a8d5ec
> Mar 20 18:15:23 GE00409729044C kernel:  r4:edaa8200
> Mar 20 18:15:23 GE00409729044C kernel: [<80486a7c>] (drm_atomic_helper_wait_for_vblanks) from [<804b3990>] (imx_drm_atomic_commit_tail+0x1b4/0x1e0)
> Mar 20 18:15:23 GE00409729044C kernel:  r10:0ee80680 r9:80d76580 r8:00000000 r7:ee1e5000 r6:ecc04f00 r5:00000000
> Mar 20 18:15:23 GE00409729044C kernel:  r4:00000004
> Mar 20 18:15:23 GE00409729044C kernel: [<804b37dc>] (imx_drm_atomic_commit_tail) from [<80487498>] (commit_tail+0x50/0x6c)
> Mar 20 18:15:23 GE00409729044C kernel:  r7:ee806800 r6:ee82b000 r5:80d3a5fc r4:ecc04f00
> Mar 20 18:15:23 GE00409729044C kernel: [<80487448>] (commit_tail) from [<804874d0>] (commit_work+0x1c/0x20)
> Mar 20 18:15:23 GE00409729044C kernel:  r5:eeb97280 r4:ecc04f1c
> Mar 20 18:15:23 GE00409729044C kernel: [<804874b4>] (commit_work) from [<8013b638>] (process_one_work+0x154/0x510)
> Mar 20 18:15:23 GE00409729044C kernel: [<8013b4e4>] (process_one_work) from [<8013ba30>] (worker_thread+0x3c/0x5cc)
> Mar 20 18:15:23 GE00409729044C kernel:  r10:eeb97280 r9:ee82b000 r8:80d02100 r7:ee82b018 r6:00000088 r5:eeb97298
> Mar 20 18:15:23 GE00409729044C kernel:  r4:ee82b000
> Mar 20 18:15:23 GE00409729044C kernel: [<8013b9f4>] (worker_thread) from [<80141670>] (kthread+0xe4/0x100)
> Mar 20 18:15:23 GE00409729044C kernel:  r10:00000000 r9:00000000 r8:00000000 r7:8013b9f4 r6:eeb97280 r5:eebae640
> Mar 20 18:15:23 GE00409729044C kernel:  r4:00000000
> Mar 20 18:15:23 GE00409729044C kernel: [<8014158c>] (kthread) from [<80108278>] (ret_from_fork+0x14/0x3c)
> Mar 20 18:15:23 GE00409729044C kernel:  r7:00000000 r6:00000000 r5:8014158c r4:eebae640
> Mar 20 18:15:23 GE00409729044C kernel: ---[ end trace ba005811962ba6f2 ]---
> 
> We believe this may be due to the vblank interrupt for the LVDS interface
> being affected when the vblank interface for the HDMI backed interface
> gets enabled. Any pointers regarding how to proceed narrowing down/fixing
> this would be appreciated.
> 
> We are currently running 4.8 kernel with 1.11 Weston compositor.
> 
> Martyn

Could you move to v4.9 or v4.10 and check if the four patches in
https://git.pengutronix.de/cgit/pza/linux/tag/?id=v4.9-ipu-dp-plane-fix
or
https://git.pengutronix.de/cgit/pza/linux/tag/?id=v4.10-ipu-dp-plane-fix-2
help?

regards
Philipp