Re: [Intel-gfx] [PATCH v3 6/7] efifb: Set info->fbcon_rotate_hint based on drm_get_panel_orientation

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 30-10-17 10:53, Daniel Vetter wrote:
> On Mon, Oct 23, 2017 at 09:14:24AM +0200, Hans de Goede wrote:
>> On some hardware the LCD panel is not mounted upright in the casing,
>> but rotated by 90 degrees. In this case we want the console to
>> automatically be rotated to compensate.
>>
>> The drm subsys has a quirk table for this, use the
>> drm_get_panel_orientation_quirk function to get the panel orientation
>> and set info->fbcon_rotate_hint based on this, so that the fbcon console
>> on top of efifb gets automatically rotated to compensate for the panel
>> orientation.
>>
>> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
>> ---
>>   drivers/video/fbdev/Kconfig |  1 +
>>   drivers/video/fbdev/efifb.c | 21 ++++++++++++++++++++-
>>   2 files changed, 21 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig
>> index 5e58f5ec0a28..c4a90c497839 100644
>> --- a/drivers/video/fbdev/Kconfig
>> +++ b/drivers/video/fbdev/Kconfig
>> @@ -772,6 +772,7 @@ config FB_VESA
>>   config FB_EFI
>>   	bool "EFI-based Framebuffer Support"
>>   	depends on (FB = y) && !IA64 && EFI
>> +	select DRM_PANEL_ORIENTATION_QUIRKS
>>   	select FB_CFB_FILLRECT
>>   	select FB_CFB_COPYAREA
>>   	select FB_CFB_IMAGEBLIT
>> diff --git a/drivers/video/fbdev/efifb.c b/drivers/video/fbdev/efifb.c
>> index 3a010641f630..8c7f6aeee205 100644
>> --- a/drivers/video/fbdev/efifb.c
>> +++ b/drivers/video/fbdev/efifb.c
>> @@ -15,6 +15,8 @@
>>   #include <linux/screen_info.h>
>>   #include <video/vga.h>
>>   #include <asm/efi.h>
>> +#include <drm/drm_utils.h> /* For drm_get_panel_orientation_quirk */
>> +#include <drm/drm_mode.h>  /* For DRM_MODE_PANEL_ORIENTATION_* */
>>   
>>   static bool request_mem_succeeded = false;
>>   static bool nowc = false;
>> @@ -156,7 +158,7 @@ static u64 bar_offset;
>>   static int efifb_probe(struct platform_device *dev)
>>   {
>>   	struct fb_info *info;
>> -	int err;
>> +	int err, orientation;
>>   	unsigned int size_vmode;
>>   	unsigned int size_remap;
>>   	unsigned int size_total;
>> @@ -328,6 +330,23 @@ static int efifb_probe(struct platform_device *dev)
>>   	info->fix = efifb_fix;
>>   	info->flags = FBINFO_FLAG_DEFAULT | FBINFO_MISC_FIRMWARE;
>>   
>> +	orientation = drm_get_panel_orientation_quirk(efifb_defined.xres,
>> +						      efifb_defined.yres);
> 
> Oh right, that's the reason for the separate function. Still ugh.
> 
> Maybe add a comment in the kernel-doc for why it is what it is ...

Actually the drm_get_panel_orientation_quirk() function was missing
a kernel-doc comment documenting it altogether. For v5 of this
patch-set I've added a kernel-doc comment and I've added this bit
to that kernel-doc comment to explain why drm_get_panel_orientation_quirk()
lives in a separate .ko target:

  * Note this function is also used outside of the drm-subsys, by for example
  * the efifb code. Because of this this function gets compiled into its own
  * kernel-module when built as a module.

Regards,

Hans



> -Daniel
> 
>> +	switch (orientation) {
>> +	default:
>> +		info->fbcon_rotate_hint = FB_ROTATE_UR;
>> +		break;
>> +	case DRM_MODE_PANEL_ORIENTATION_BOTTOM_UP:
>> +		info->fbcon_rotate_hint = FB_ROTATE_UD;
>> +		break;
>> +	case DRM_MODE_PANEL_ORIENTATION_LEFT_UP:
>> +		info->fbcon_rotate_hint = FB_ROTATE_CCW;
>> +		break;
>> +	case DRM_MODE_PANEL_ORIENTATION_RIGHT_UP:
>> +		info->fbcon_rotate_hint = FB_ROTATE_CW;
>> +		break;
>> +	}
>> +
>>   	err = sysfs_create_groups(&dev->dev.kobj, efifb_groups);
>>   	if (err) {
>>   		pr_err("efifb: cannot add sysfs attrs\n");
>> -- 
>> 2.14.2
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
>