• What changed?

Only one change since 1.1.1 to fix a major regression in the network stack to reassemble fragmented packets during heavy load situations.

• How risky is this change?

The change revert to the old code that´s known to work. Apparently the issue was not spotted in normal regression testing as it shows only in very specific environments and network loads. Upstream now has a specific test case to validate the code path.
The code now passes the test suite.

• How important is this change?

Very important, because there is no way to predict when the bug will trigger exactly and there is no workaround.

• How well tested is this package build?

As the described above, upstream has created a specific test for this situation and test suite passes. The user reporting the issue has confirmed that reverting to the old code fixes the problem.

• Is this package in the critical-path list?

No