From 379e5c09a6f046cfd16128886ef3ebf8b2d69914 Mon Sep 17 00:00:00 2001 From: Andreas Reich Date: Mon, 16 Dec 2024 12:27:33 +0100 Subject: [PATCH] link issue --- crates/viewer/re_viewer/src/lib.rs | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/crates/viewer/re_viewer/src/lib.rs b/crates/viewer/re_viewer/src/lib.rs index 60524824134a..3b6bede7f1cd 100644 --- a/crates/viewer/re_viewer/src/lib.rs +++ b/crates/viewer/re_viewer/src/lib.rs @@ -193,13 +193,13 @@ pub(crate) fn wgpu_options(force_wgpu_backend: Option) -> egui_wgpu::Wgp egui_wgpu::SurfaceErrorAction::SkipFrame } }), - // TODO(andreas): It would be great to use `egui_wgpu::WgpuSetup::Existing` and put the + // TODO(#8475): It would be great to use `egui_wgpu::WgpuSetup::Existing` and put the // full control of adapter creation into the hands of `re_renderer`. // However, we generally need to take into account the _surface_ as well: // * this is a strict *requirement* when using WebGL // * on OpenGL & Linux it _helps_ to know the surface because either Vulkan or OpenGL may not be happy with all surfaces // - // TODO(andreas): Next better thing that we should aspire for is to allow rejecting adapters on native in egui. + // Next better thing that we should aspire for is to allow rejecting adapters on native in egui. // I.e. instead of always providing a device descriptor, we should allow it to fail for a given device. // This rejection should happen with reason-message so it's tractable why a given adapter wasn't chosen. // Which is obviously what we want to show when we're rejecting all adapters, but it would