On Fri Dec 5 11:34:58 2025 +0000, Rémi Bernon wrote:
The same question can apply to the winebus side, but CRC32 doesn't seem like a very good hashing method? Could we use something a bit more robust or does it not matter for some reason? Ideally of course we would do the same hashing as Windows but it's probably tricky to figure it out by trial and error. According to the [docs](https://learn.microsoft.com/en-us/windows-hardware/drivers/install/avoiding-...), which lines up with what I've seen:
- The operating system concatenates the USB device serial number, vendor ID, product ID, and revision number to generate a string. - The string that results is hashed into a GUID by using the UUID Version 5 (SHA-1) hash algorithm under a USB-specific namespace. The generated container ID will be unique, provided the independent hardware vendor (IHV) provides a unique serial number on each device. We probably don't need to find the exact same string layout, but I think the general idea sounds easy to implement. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9687#note_124931