| Summary: | Support for VK_KHR_mir_surface should be removed or made internal | ||
|---|---|---|---|
| Product: | SDL | Reporter: | Jason Ekstrand <jason> |
| Component: | *don't know* | Assignee: | Ryan C. Gordon <icculus> |
| Status: | RESOLVED FIXED | QA Contact: | Sam Lantinga <slouken> |
| Severity: | normal | ||
| Priority: | P2 | CC: | brandontschaefer, programmerjake |
| Version: | don't know | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
|
Description
Jason Ekstrand
2018-10-01 15:41:35 UTC
We don't actually use this extension, it's simply part of the Khronos header we've pulled. Please let us know when it's been removed from Khronos headers and we'll grab new ones. Thanks! (In reply to Sam Lantinga from comment #1) > We don't actually use this extension, it's simply part of the Khronos header > we've pulled. > > Please let us know when it's been removed from Khronos headers and we'll > grab new ones. > > Thanks! That's incorrect, see https://github.com/spurious/SDL-mirror/blob/d2fcd08e539f879ba1cfe864a5fd310cf8deb72b/src/video/mir/SDL_mirvulkan.c Oh sorry, you're correct. I think we're also planning to remove the Mir code from SDL. Ryan, do you know the status of that? Any update? We'd like to move the extension removal process along. If it's going to take a long time to rip out Mir support, you can carry a version of the headers internally that still contains VK_KHR_mir_surface to keep your code building. (In reply to Sam Lantinga from comment #3) > Oh sorry, you're correct. I think we're also planning to remove the Mir code > from SDL. > > Ryan, do you know the status of that? Last we spoke, the opinion of the Mir developers was: we're moving away from this to turn Mir into a Wayland client, so hold onto this code a little longer. But honestly? Has anyone outside of the Mir project ever used this, and will anyone _ever_ use this code in SDL? I suspect the answer is no and no. I say we dump it now, before 2.0.9. I'll yank it today if you agree, Sam. --ryan. Let's dump it for 2.0.10. Thanks! (In reply to Sam Lantinga from comment #6) > Let's dump it for 2.0.10. > > Thanks! Now dumped in https://hg.libsdl.org/SDL/rev/ca6e41ade79a --ryan.
> Now dumped in https://hg.libsdl.org/SDL/rev/ca6e41ade79a
Oh, to be clear, this removes the Mir backend from SDL entirely, but the Vulkan headers still have it. But we aren't using this part of the Vulkan headers now, if someone wants to update those.
--ryan.
+1 from me as well, and thanks for removing! FYI, it was removed from the headers in the latest Vulkan spec update. The SDL Vulkan headers have been updated: https://hg.libsdl.org/SDL/rev/77433beaf97a |