On 1/12/21 10:20 PM, Daniel Vetter wrote:

On Tue, Jan 12, 2021 at 06:46:44PM +0800, ZhiJie.Zhang wrote:

...

From: zhangzhijie zhangzhijie@loongson.cn

codeview the implementation of few Drivers.

I'm not really understanding what you're trying to say here.

I exc "make htmls" and read the gpu's html via browser. then founded that drm_fb_helper_hotplug_changed not a tag in kernel. then I touch this callback been called via drm_kms_helper_hotplug_event().

Sorry， I will delete this words at next push

...

this callback was used by drm_kms_helper_hotplug_event()

Signed-off-by: zhangzhijie zhangzhijie@loongson.cn

include/drm/drm_mode_config.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h index ab424ddd7665..e01c4d0f07d1 100644 --- a/include/drm/drm_mode_config.h +++ b/include/drm/drm_mode_config.h @@ -104,7 +104,7 @@ struct drm_mode_config_funcs { * changes. * * Drivers implementing fbdev emulation with the helpers can call

• * drm_fb_helper_hotplug_changed from this hook to inform the fbdev
  

• * drm_kms_helper_hotplug_event() from this hook to inform the fbdev
  

  • helper of output changes.

I think since we touch this, maybe better to revamp it complete. The best way to handle all this is by registering a struct drm_client, since that provides the &drm_client_funcs.hotplug callback. Also for fbdev support drivers shouldn't even use that, but instead use the drm_fbdev_generic_setup() function, which takes care of everything.

I think we can also remove the FIXME below, since with the drm_client infrastructure and the generic fbdev emulation we've resolved this all very neatly now.

Can you please respin with my suggestions taking into account somehow?

Thanks, Daniel

Could i copy this words to comment of output_poll_changed? I want better reading the gpu's chapter at kernel doc

Thanks，ZhiJie

...

*
* FIXME:

-- 2.29.2