Avoid ppoll timeout if there is prefix data already ready to send - #242
Conversation
If there is prefix data to send (and a space for it), do not wait for other trigger, but go directly to sending it. Otherwise, it may end up waiting 10s timeout if no other trigger happens. In practice, most of the time (but not every time), vchan trigger would happen early on connection setup anyway. Fixes QubesOS/qubes-issues#9951
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #242 +/- ##
==========================================
+ Coverage 81.12% 81.16% +0.04%
==========================================
Files 58 58
Lines 10946 10981 +35
==========================================
+ Hits 8880 8913 +33
- Misses 2066 2068 +2 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
OpenQA test summaryComplete test suite and dependencies: https://openqa.qubes-os.org/tests/overview?distri=qubesos&version=4.3&build=2026080503-devel&flavor=pull-requests Test run included the following:
New failures, excluding unstableCompared to: https://openqa.qubes-os.org/tests/overview?distri=qubesos&version=4.3&build=2026050504-devel&flavor=update
Failed tests15 failures
Fixed failuresCompared to: https://openqa.qubes-os.org/tests/176874#dependencies 33 fixed
Unstable testsDetails
Performance TestsPerformance degradation:23 performance degradations
Remaining performance tests:88 tests
|
|
Nice catch @marmarek! I believe that was my mistake from when I added prefix data support. |
3 participants
If there is prefix data to send (and a space for it), do not wait for
other trigger, but go directly to sending it. Otherwise, it may end up
waiting 10s timeout if no other trigger happens. In practice, most of
the time (but not every time), vchan trigger would happen early on
connection setup anyway.
Fixes QubesOS/qubes-issues#9951