Thomas Eizinger
Thomas Eizinger
> > The buffer for control messages is too small, thus Windows fails to fit the `UDP_COALESCED_INFO` message into it and can't tell us that the buffer is coalesced even...
> We see a similar issue on Windows on ARM, see summary in [last comment on this thread](https://bugzilla.mozilla.org/show_bug.cgi?id=1916558#c39). Interesting, according to https://bugzilla.mozilla.org/show_bug.cgi?id=1916558#c43, the `UDP_COALESCED_INFO` never gets received. We only get...
In https://learn.microsoft.com/en-us/windows-hardware/drivers/network/udp-rsc-offload, it is mentioned that this only works on Windows 11, version 24H2 and later. I'll have to investigate what version our customer is using.
> Could the receiver interface be a dual stack (IPv4 and IPv6) one ? I had no chance to test that case when writing the code, maybe we receive ipv4...
@mxinden Do you know by any chance if the affected user is using a dock? We have reports that this only happens when the user is connected via a Dell...
More theories: - Do we have a bug in `cmsg::Iter` and fail to decode the last control message? Perhaps printing the raw bytes sitting in the cmesg buffer might be...
I managed to set up a Windows 11 VM that uses USB passthrough to a dedicated network card and I can confirm that GRO works there. Any ideas on what...
@mxinden Have you checked what the return value of https://github.com/quinn-rs/quinn/blob/f8b8c5032e0db9d7dbc7c3452f09c7d1e2a4295d/quinn-udp/src/windows.rs#L110-L118 is on the problematic machine? Maybe this actually fails but leaves the driver in a kind of limbo state and...
Here is my workaround: - Make `divan` an optional dependency - Put your benchmarks within a `benches` module that uses `cfg(features = "divan")` - Declare the feature in your `Cargo.toml`:...
> @thomaseizinger I believe [rust-lang/cargo#1982](https://github.com/rust-lang/cargo/issues/1982) would remove the need to pass an explicit feature flag The feature flag is required because `cargo` will always compile the library part of the...