Martin Dickopp
Martin Dickopp
When I looked at the source code (to see if I can implement a fix), I found another case that is incorrect: #### Case 5 (fixed by #220) | Command...
There's another case: #### Case 6 | Command | Expected Encoding | Observed Encoding | | --- | --- | --- | | `curl -I -H 'accept-encoding: *;q=0.8, gzip;q=0.5' http://127.0.0.1:3000/`...
I will work on a fix for this issue.
Sure. Cases 4 and 6 are currently unresolved.
@Nehliin Sorry, I'm a bit confused now. Could you clarify what behavior you expect when the client specifies multiple encodings with same preference explicitly (e.g. `Accept-Encoding: gzip. deflate, br`), and...
This fixes both #3380 and #4579 for me. Thanks! (Since I do not use Wayland, I cannot test #3606).
Please note that I can reproduce both #3380 and #4579 on an Intel device, so I do not think they are specific to AMD devices. ``` 2022-09-13T17:14:48.220831Z INFO winit::platform_impl::platform::x11::window: Guessed...
@nicopap > Hmm, can't reproduce on my Whiskey Lake intel iGPU. Looks like you are using a Broadwell, which is very common. I've a Broadwell CPU somewhere, but, at the...
After some digging I found out that the Vulkan backend of wgpu calls [vkAcquireNextImageKHR](https://registry.khronos.org/vulkan/specs/1.3-extensions/html/vkspec.html#vkAcquireNextImageKHR) with a timeout of 1 second. A comment in (an older version of) the source code...
Setting the timeout in wgpu to 2 seconds does not fix ~the issues~ issue #3380 for me. (EDIT: See more detailed explanation below.)