Skip to content

Commit

Permalink
Refine comment of async_io test
Browse files Browse the repository at this point in the history
This change refines the comment on why we've cfged out a test on windows with a link to the issue tracker for those curious as to why this is the case.

Co-authored-by: Aaron Turon <aturon@fastly.com>
  • Loading branch information
mgattozzi and aturon authored Nov 16, 2022
1 parent 3c9e62b commit 12ec538
Showing 1 changed file with 6 additions and 4 deletions.
10 changes: 6 additions & 4 deletions cli/tests/integration/async_io.rs
Original file line number Diff line number Diff line change
Expand Up @@ -9,10 +9,12 @@ use std::sync::atomic::Ordering;
use std::sync::Arc;
use tokio::sync::Barrier;

// For some reason on windows this test fails with the write body portion due to
// some difference we believe between unix systems and windows when it comes to
// hyper. We don't believe this means the implementation is wrong. As such we've
// disabled this test only on windwos.
// On Windows, streaming body backpressure doesn't seem to work as expected, either
// due to the Hyper client or server too eagerly clearing the chunk buffer. This issue does
// not appear related to async I/O hostcalls; the behavior is seen within the streaming body
// implementation in general. For the time being, this test is unix-only.
//
// https://github.com/fastly/Viceroy/issues/207 tracks the broader issue.
#[cfg(target_family = "unix")]
#[tokio::test(flavor = "multi_thread")]
async fn async_io_methods() -> TestResult {
Expand Down

0 comments on commit 12ec538

Please sign in to comment.