git: 5d1403a79a3e - main - [rtsock] Enforce netmask/RTF_HOST consistency.

[Alexander V. Chernikov]

> On 27 Apr 2021, at 13:26, Alexander V. Chernikov <melifaro at freebsd.org> wrote:
> 
> Ack. Interesting. Will look later today. Thank you for bisecting it - thought the issue was caused by a different change.
Looks like there is no more failures in https://ci.freebsd.org/job/FreeBSD-main-amd64-test/18039/#showFailuresLink .
In fact, I think https://cgit.freebsd.org/src/commit/?id=439d087d0b55574db81f4a2799a411c1236d95e3 is the reason.

> 
> --
> Sent from Yandex.Mail for mobile
> 
> 27.04.2021, 12:38, "Kristof Provost" <kp at freebsd.org>:
> On 25 Apr 2021, at 1:03, Alexander V. Chernikov wrote:
> 
> The branch main has been updated by melifaro:
> 
> URL: https://cgit.FreeBSD.org/src/commit/?id=5d1403a79a3e56403fb63c062252a23fce81e5f1 <https://cgit.freebsd.org/src/commit/?id=5d1403a79a3e56403fb63c062252a23fce81e5f1>
> 
> commit 5d1403a79a3e56403fb63c062252a23fce81e5f1
> Author: Alexander V. Chernikov <melifaro at FreeBSD.org <mailto:melifaro at FreeBSD.org>>
> AuthorDate: 2021-04-23 21:53:47 +0000
> Commit: Alexander V. Chernikov <melifaro at FreeBSD.org <mailto:melifaro at FreeBSD.org>>
> CommitDate: 2021-04-24 22:41:27 +0000
> 
> [rtsock] Enforce netmask/RTF_HOST consistency.
> 
> Traditionally we had 2 sources of information whether the
> added/delete route request targets network or a host route:
> netmask (RTA_NETMASK) and RTF_HOST flag.
> 
> The former one is tricky: netmask can be empty or can explicitly
> specify the host netmask. Parsing netmask sockaddr requires per-family
> parsing and that's what rtsock code traditionally avoided. As a result,
> consistency was not enforced and it was possible to specify network with
> the RTF_HOST flag and vice versa.
> 
> Continue normalization efforts from D29826 and D29826 and ensure that
> RTF_HOST flag always reflects host/network data from netmask field.
> 
> Differential Revision: https://reviews.freebsd.org/D29958 <https://reviews.freebsd.org/D29958>
> MFC after: 2 days
> 
> I believe this commit triggers a number of regression test failures: https://ci.freebsd.org/job/FreeBSD-main-amd64-test/18011/#showFailuresLink <https://ci.freebsd.org/job/FreeBSD-main-amd64-test/18011/#showFailuresLink>
> The easiest way to reproduce the problem is to kldload ipsec ; cd /usr/tests/sys/netipsec ; kyua test.
> Reverting this patch allows the tests to pass again. (There are a few others that also fail, but the ipsec tests appear to be the most reliable way to demonstrate the problem.)
> 
> Best regards,
> Kristof
>