The A23's display pipeline is similar to the A33. Differences include:
- Display backend supports larger layers, 8192x8192 instead of 2048x2048
- TCON has DMA input
- There is no SAT module packed in the display backend
Add support for the display pipeline and its components.
As the MIPI
We might want to use the backend pointer from DRM callbacks that get
called within drm_universal_plane_init(), such as the
.format_mod_supported callback.
Move the assignment of the layer's backend pointer to right after the
structure is allocated.
Signed-off-by: Chen-Yu Tsai
---
drivers/gpu/dr
The display backend does not support BGRX. There is also no trace
of this in the original list of supported formats before the commit
b636d3f97d04 ("drm/sun4i: frontend: Add support for the BGRX input
format"). Nor do the backend configuration helpers handle this format.
Remove BGRX fr
The Q8 design for A23/A33 tablets have an 18-bit RGB LCD panel connected
to the LCD interface on the SoC, the DC1SW output on the PMIC providing
power for the LCD, and PH7 toggling the reset pin for the panel.
This patch adds a device node for the panel, describing the above, and
enables the displ
The Q8 tablets follow the A23/A33 tablet reference design, and normally
use a "generic" 800x480 LCD panel. The actual panel may vary between
production runs, and there are no visible markings denoting its model.
This patch uses a panel that has the same dimensions and timings that
are close to what
The display pipeline has the same structure, resources and connections
on both the A23 and A33. The differences include:
- compatible strings
- extra clock, reset control, and IO region for SAT in the backend
only found on the A33
- missing ch1 clock for the TCON
However, while the A23
Hi everyone,
This series enables the display pipeline on the Allwinner A23 SoC.
A few fixes are included for corner cases when the frontend isn't
enabled.
The A23 display pipeline is very much the same as the A33, except
that the A23 does not have the SAT IP block embedded within the
display back
In some cases, such as running a new kernel with an old device tree that
has the frontend disabled, the backend's matching frontend might be
unavailable.
When this happens, the layers should only declare support for formats
that the backend support. This partially reverts commit 1c29d263f624
("drm
The NAND controller device node was inserted into the wrong position,
probably due to a rebase or merge, as the file's structure does not
provide enough context for git to accurately match the previous device
node block.
Fixes: d7b843df13ea ("ARM: dts: sun8i: add NAND controller node for A23/A33")
https://bugs.freedesktop.org/show_bug.cgi?id=109445
Mike Mestnik changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
On Thu, Jan 24, 2019 at 05:58:24PM +0100, Daniel Vetter wrote:
> This should not result in any changes.
I'd love to merge https://patchwork.freedesktop.org/series/53951/
instead (which will -- among other things -- switch qxl over to the
generic fbdev emulation and remove the code you are patching
https://bugs.freedesktop.org/show_bug.cgi?id=105733
--- Comment #61 from Zheng Luo ---
I experienced similar problems, but mine is much worse. I can't recover from
black screen after reboot/hard reset unless I drain the builtin battery.
However this problem disappears in 5.0rc3 (in contrast to th
201 - 212 of 212 matches
Mail list logo
