Re: [PATCH 4.4 000/124] 4.4.146-stable review

[Greg Kroah-Hartman]

On Sun, Aug 05, 2018 at 03:19:33PM +0000, Dan Rue wrote:
> On Sun, Aug 05, 2018 at 03:49:52PM +0200, Greg Kroah-Hartman wrote:
> > On Sun, Aug 05, 2018 at 05:24:21PM +0530, Naresh Kamboju wrote:
> > > On 4 August 2018 at 14:29, Greg Kroah-Hartman
> > > <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > > > This is the start of the stable review cycle for the 4.4.146 release.
> > > > There are 124 patches in this series, all will be posted as a response
> > > > to this one.  If anyone has any issues with these being applied, please
> > > > let me know.
> > > >
> > > > Responses should be made by Mon Aug  6 08:26:39 UTC 2018.
> > > > Anything received after that time might be too late.
> > > >
> > > > The whole patch series can be found in one patch at:
> > > >         https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.4.146-rc1.gz
> > > > or in the git tree and branch at:
> > > >         git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.4.y
> > > > and the diffstat can be found below.
> > > >
> > > > thanks,
> > > >
> > > > greg k-h
> > > 
> > > Results from Linaroâs test farm.
> > > No regressions on arm64, arm and x86_64.
> > 
> > Really?  How did you all miss the netfilter bug that was introduced by
> > this series?
> > 
> > Anyway, thanks for testing, but I'm worried...
> 
> I took a look through our test suites to see where we may have coverage for
> this. There is a netns series of tests in LTP [1], but we've had trouble
> getting them running successfully in the past and so they are being skipped. I
> don't know if it would have caught this particular issue, but I'll find out,
> and also see if we can get them working and running again.
> 
> We also run the net/ tests in kselftest, but there aren't that many, and of
> course running kselftest the way we do is not a well supported usecase (running
> all of 4.17 kselftest against 4.4 kernels, skipping tests that are broken).
> 
> So I would ask, how would net/ expect such things to be tested, and what can we
> do to help?

I don't know.  Somehow this was caught when running Android's userspace,
which seems to be a great stress test for networking.  What about
running the Android networking tests on kernels?  Those don't need to be
run on an android system to work properly last I checked.

thanks,

greg k-h