Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1734333

Re: [PATCH 0/3] fix reuseaddr regression

Show all headers | View raw

On 09/18/2017 12:28 PM, josef@toxicpanda.com wrote:
> I introduced a regression when reworking the fastreuse port stuff that allows
> bind conflicts to occur once a reuseaddr socket successfully opens on an
> existing tb.  The root cause is I reversed an if statement which caused us to
> set the tb as if there were no owners on the socket if there were, which
> obviously is not correct.
> 
> Dave I have follow up patches that will add a selftest for this case and I ran
> the other reuseport related tests as well.  These need to go in pretty quickly
> as it breaks kvm, I've marked them for stable.  Sorry for the regression,
> 

To clarify, it doesn't really break KVM specifically, but it breaks a
port collision detection idiom that libvirt depends on to successfully
launch qemu/xen/... VMs in certain cases.

Thanks,
Cole

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 0/3] fix reuseaddr regression josef@toxicpanda.com - 2017-09-18 18:30 +0200
  [PATCH 1/3] net: set tb->fast_sk_family josef@toxicpanda.com - 2017-09-18 18:40 +0200
  Re: [PATCH 0/3] fix reuseaddr regression Cole Robinson <crobinso@redhat.com> - 2017-09-18 19:50 +0200
  Re: [PATCH 0/3] fix reuseaddr regression David Miller <davem@davemloft.net> - 2017-09-19 23:00 +0200
    Re: [PATCH 0/3] fix reuseaddr regression Josef Bacik <josef@toxicpanda.com> - 2017-09-23 02:30 +0200

csiph-web