[PATCH v3 0/7] drm/bridge_connector: perform HPD enablement automatically

[Dmitry Baryshkov]

On 10/01/2023 08:57, Laurentiu Palcu wrote:
> Hi,
> 
> On Mon, Jan 09, 2023 at 10:26:28PM +0200, Dmitry Baryshkov wrote:
>> Hi,
>>
>> On 09/01/2023 18:21, Laurentiu Palcu wrote:
>>> Hi Dmitry,
>>>
>>> It looks like there are some issues with this patchset... :/ I just
>>> fetched the drm-tip and, with these patches included, the "Hot plug
>>> detection already enabled" warning is back for i.MX DCSS.
>>
>> Could you please provide a backtrace?
> 
> Sure, see below:

I wondered, why didn't I see this on msm, my main target nowadays. The 
msm driver is calling msm_kms_helper_poll_init() after initializing 
fbdev, so all previous kms_helper_poll_enable() calls return early.

I think I have the fix ready. Let me test it locally before posting.

> 
> ------------[ cut here ]------------
> Hot plug detection already enabled
> WARNING: CPU: 2 PID: 9 at drivers/gpu/drm/drm_bridge.c:1257 drm_bridge_hpd_enable+0x94/0x9c [drm]
> Modules linked in: videobuf2_memops snd_soc_simple_card snd_soc_simple_card_utils fsl_imx8_ddr_perf videobuf2_common snd_soc_imx_spdif adv7511 etnaviv imx8m_ddrc imx_dcss mc cec nwl_dsi gov
> CPU: 2 PID: 9 Comm: kworker/u8:0 Not tainted 6.2.0-rc2-15208-g25b283acd578 #6
> Hardware name: NXP i.MX8MQ EVK (DT)
> Workqueue: events_unbound deferred_probe_work_func
> pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> pc : drm_bridge_hpd_enable+0x94/0x9c [drm]
> lr : drm_bridge_hpd_enable+0x94/0x9c [drm]
> sp : ffff800009ef3740
> x29: ffff800009ef3740 x28: ffff000009331f00 x27: 0000000000001000
> x26: 0000000000000020 x25: ffff800001148ed8 x24: ffff00000a8fe000
> x23: 00000000fffffffd x22: ffff000005086348 x21: ffff800001133ee0
> x20: ffff00000550d800 x19: ffff000005086288 x18: 0000000000000006
> x17: 0000000000000000 x16: ffff8000096ef008 x15: 97ffff2891004260
> x14: 2a1403e194000000 x13: 97ffff2891004260 x12: 2a1403e194000000
> x11: 7100385f29400801 x10: 0000000000000aa0 x9 : ffff800008112744
> x8 : ffff000000250b00 x7 : 0000000000000003 x6 : 0000000000000011
> x5 : 0000000000000000 x4 : ffff0000bd986a48 x3 : 0000000000000001
> x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff000000250000
> Call trace:
>   drm_bridge_hpd_enable+0x94/0x9c [drm]
>   drm_bridge_connector_enable_hpd+0x2c/0x3c [drm_kms_helper]
>   drm_kms_helper_poll_enable+0x94/0x10c [drm_kms_helper]
>   drm_helper_probe_single_connector_modes+0x1a8/0x510 [drm_kms_helper]
>   drm_client_modeset_probe+0x204/0x1190 [drm]
>   __drm_fb_helper_initial_config_and_unlock+0x5c/0x4a4 [drm_kms_helper]
>   drm_fb_helper_initial_config+0x54/0x6c [drm_kms_helper]
>   drm_fbdev_client_hotplug+0xd0/0x140 [drm_kms_helper]
>   drm_fbdev_generic_setup+0x90/0x154 [drm_kms_helper]
>   dcss_kms_attach+0x1c8/0x254 [imx_dcss]
>   dcss_drv_platform_probe+0x90/0xfc [imx_dcss]
>   platform_probe+0x70/0xcc
>   really_probe+0xc4/0x2e0
>   __driver_probe_device+0x80/0xf0
>   driver_probe_device+0xe0/0x164
>   __device_attach_driver+0xc0/0x13c
>   bus_for_each_drv+0x84/0xe0
>   __device_attach+0xa4/0x1a0
>   device_initial_probe+0x1c/0x30
>   bus_probe_device+0xa4/0xb0
>   deferred_probe_work_func+0x90/0xd0
>   process_one_work+0x200/0x474
>   worker_thread+0x74/0x43c
>   kthread+0xfc/0x110
>   ret_from_fork+0x10/0x20
> ---[ end trace 0000000000000000 ]---
> 
> Cheers,
> Laurentiu
> 
>>
>>>
>>> After a short investigation, it seems that we end up calling
>>> drm_bridge_hpd_enable() from both drm_kms_helper_poll_init() and
>>> drm_fbdev_generic_setup(), hence the warning.
>>>
>>> There are drivers using the drm_bridge_connector API that also call
>>> drm_kms_helper_poll_init() followed by drm_fbdev_generic_setup(). So,
>>> they might experience the same behavior, unless I'm missing something...
>>> :/
>>>
>>> Also, even if drm_fbdev_generic_setup() is not called in the driver
>>> initialization, the warning will still appear the first time the
>>> GETCONNECTOR ioctl is called, because that'll call
>>> drm_helper_probe_single_connector_modes() helper which will eventually
>>> call drm_bridge_hpd_enable().
>>>
>>> Any idea?
>>>
>>> Cheers,
>>> Laurentiu
>>>
>>> On Wed, Nov 02, 2022 at 09:06:58PM +0300, Dmitry Baryshkov wrote:
>>>>   From all the drivers using drm_bridge_connector only iMX/dcss and OMAP
>>>> DRM driver do a proper work of calling
>>>> drm_bridge_connector_en/disable_hpd() in right places. Rather than
>>>> teaching each and every driver how to properly handle
>>>> drm_bridge_connector's HPD, make that automatic.
>>>>
>>>> Add two additional drm_connector helper funcs: enable_hpd() and
>>>> disable_hpd(). Make drm_kms_helper_poll_* functions call them (as this
>>>> is the time where the drm_bridge_connector's functions are called by the
>>>> drivers too).
>>>>
>>>> Changes since v2:
>>>>    - Fixed a typo in the commit message of the second patch.
>>>>
>>>> Changes since v1:
>>>>    - Rebased on top of v6.1-rc1
>>>>    - Removed the drm_bridge_connector_enable_hpd() from
>>>>      drm_bridge_connector_init()
>>>>    - Removed extra underscore prefix from
>>>>      drm_bridge_connector_en/disable_hpd() helpers
>>>>
>>>> Dmitry Baryshkov (7):
>>>>     drm/poll-helper: merge drm_kms_helper_poll_disable() and _fini()
>>>>     drm/probe-helper: enable and disable HPD on connectors
>>>>     drm/bridge_connector: rely on drm_kms_helper_poll_* for HPD enablement
>>>>     drm/imx/dcss: stop using drm_bridge_connector_en/disable_hpd()
>>>>     drm/msm/hdmi: stop using drm_bridge_connector_en/disable_hpd()
>>>>     drm/omap: stop using drm_bridge_connector_en/disable_hpd()
>>>>     drm/bridge_connector: drop drm_bridge_connector_en/disable_hpd()
>>>>
>>>>    drivers/gpu/drm/drm_bridge_connector.c   | 27 +++-------------
>>>>    drivers/gpu/drm/drm_probe_helper.c       | 40 ++++++++++++++++++-----
>>>>    drivers/gpu/drm/imx/dcss/dcss-dev.c      |  4 ---
>>>>    drivers/gpu/drm/imx/dcss/dcss-kms.c      |  2 --
>>>>    drivers/gpu/drm/msm/hdmi/hdmi.c          |  2 --
>>>>    drivers/gpu/drm/omapdrm/omap_drv.c       | 41 ------------------------
>>>>    include/drm/drm_bridge_connector.h       |  2 --
>>>>    include/drm/drm_modeset_helper_vtables.h | 22 +++++++++++++
>>>>    8 files changed, 59 insertions(+), 81 deletions(-)
>>>>
>>>> -- 
>>>> 2.35.1
>>>>
>>
>> -- 
>> With best wishes
>> Dmitry
>>

-- 
With best wishes
Dmitry