[new release] mirage-clock-freestanding, mirage-clock and mirage-clock-unix (3.1.0)

[dinosaure]

Paravirtual implementation of the MirageOS Clock interface

• Project page: https://github.com/mirage/mirage-clock

CHANGES:

• Use implicit executable dependency for discover.exe
  into the dune file (Use implicit executable dependency for discover.exe mirage/mirage-clock#46 @TheLortex)
• Fix cross-compilation on Android (fix discover on Android mirage/mirage-clock#45 @EduardoRFS)

[camelus]

Commit: 9d85a11

A pull request by opam-seasoned @dinosaure.

☀️ All lint checks passed 9d85a11

• These packages passed lint tests: mirage-clock-freestanding.3.1.0, mirage-clock-unix.3.1.0, mirage-clock.3.1.0

---

☀️ Installability check (+3)

• new installable packages (3): mirage-clock.3.1.0 mirage-clock-freestanding.3.1.0 mirage-clock-unix.3.1.0

[dinosaure]

I think the error is unrelated about mirage-clock (compilation error on tcpip.4.0.0). I should good enough to merge.

[mseri]

It seems to me that tcpip compiles fine but it fails to run some tests with no output, isn’t that strange?

In the failed tests, the following tests are missing from the output but no failure is signalled

-   [OK]          socket               0   two sockets connect via TCP.
-   [SKIP]        socket               1   icmp echo-requests are sent.
-   [OK]          connect              0   connect two stacks, basic test.
-   [SKIP]        connect              1   connect two stacks, uniform packet loss of packets with no payload x 100.
-   [OK]          connect              2   connect two stacks, with trailing bytes.
-   [SKIP]        deadlock             0   test tcp deadlock with slow receiver.
-   [OK]          iperf                0   iperf with two stacks, basic tests.
-   [OK]          iperf                1   iperf with two stacks, over an MTU-enforcing backend.
-   [OK]          iperf                2   iperf with two stacks, testing trailing_bytes.
-   [OK]          iperf                3   iperf with two stacks and uniform packet loss.
-   [OK]          iperf                4   iperf with two stacks and uniform packet loss of packets with no payload.
-   [SKIP]        iperf                5   iperf with two stacks and uniform packet loss of packets with no payload, longer.
-   [SKIP]        iperf                6   iperf with two stacks, basic tests, longer.
-   [SKIP]        iperf                7   iperf with two stacks and uniform packet loss, longer.
-   [OK]          iperf                8   iperf with two stacks drop 1 sec after 1 mb.
-   [OK]          keepalive            0   correct number of keepalives.
-   [OK]          keepalive            1   we don't try to send old keepalives.
-   [SKIP]        keepalive            2   check we close if we miss all probes.
-   [SKIP]        keepalive            3   check that TCP keepalives detect a network failure.

[kit-ty-kate]

It seems to have been transient failures somewhere in tcpip's (4.0.0, 5.0.0 & 5.0.1) tests. Some of them didn't even show any skips.

Merging this since the changes in this release don't seem to be related. Thanks!