video/out/gpu/context: add target_csp callback to ra_swapchain #15279
Conversation
|
/cc @Dudemanguy |
|
Download the artifacts for this pull request: |
|
Also auto being implemented in this fashion would make everything else (like drm) basically go back to behaving like |
This allows to get target display parameters.
The intention for now is that when the |
| struct ra_swapchain_fns { | ||
| // Gets the current framebuffer depth in bits (0 if unknown). Optional. | ||
| int (*color_depth)(struct ra_swapchain *sw); | ||
|
|
||
| // Target device color space. Optional. | ||
| pl_color_space_t (*target_csp)(struct ra_swapchain *sw); |
There was a problem hiding this comment.
Kind of nitpicky but is there a reason to return a colorspace? It seems like you only actually use the display's peak brightness.
There was a problem hiding this comment.
My initial version was using only target peak, but it turns out it is also returned for SDR displays. Like I have one with 270 nits. Correct way is to check the color space returned from the API. Also primaries might be useful in the future.
There is still one thing to fix, because I noticed that lib placebo completely fails to render SDR correctly if the target peak is not 203. I may have time tomorrow to fix that.
There was a problem hiding this comment.
You could only return the MaxLuminance value if the colorspace is actually an HDR one and return 203 otherwise.
No description provided.