[OE-core] [PATCH 0/6] linux-yocto: consolidated pull request (5.0 kernel intro)

[Bruce Ashfield]

On Thu, Mar 7, 2019 at 8:04 AM Richard Purdie
<richard.purdie at linuxfoundation.org> wrote:
>
> On Wed, 2019-03-06 at 23:23 -0500, Bruce Ashfield wrote:
> > On Wed, Mar 6, 2019 at 11:21 PM Bruce Ashfield <
> > bruce.ashfield at gmail.com> wrote:
> > > On Wed, Mar 6, 2019 at 6:05 PM Richard Purdie
> > > <richard.purdie at linuxfoundation.org> wrote:
> > > > https://autobuilder.yoctoproject.org/typhoon/#/builders/60/builds/365/steps/7/logs/step1c
> > > >
> > > > https://autobuilder.yoctoproject.org/typhoon/#/builders/63/builds/365/steps/7/logs/step1c
> > > >    RESULTS - kernelmodule.KernelModuleTest.test_kernel_module -
> > > > Testcase 1541: FAILED
> > >
> > > Fixed, patches out shortly.
> > >
> > > > d) perf do_compile race seen on qemuarm64 (didn't reproduce
> > > > locally):
> > > >
> > > > https://autobuilder.yoctoproject.org/typhoon/#/builders/42/builds/366/steps/7/logs/step1b
> > >
> > > I can't reproduce this one either.
> >
> > I did notice that the AB log seems to show 4.19 being built for this,
> > is that just a red herring ?
> >
> > NOTE: recipe iptables-1.6.2-r0: task do_package_write_rpm: Succeeded
> > NOTE: recipe iptables-1.6.2-r0: task do_package_qa: Succeeded
> > NOTE: recipe ofono-1.25-r0: task do_package_write_rpm: Succeeded
> > NOTE: recipe linux-yocto-4.19.19+gitAUTOINC+41fe45f7c5_11e0e616ed-r0:
> > task do_package_write_deb: Succeeded
> > NOTE: recipe ofono-1.25-r0: task do_package_write_deb: Succeeded
> > NOTE: recipe linux-yocto-4.19.19+gitAUTOINC+41fe45f7c5_11e0e616ed-r0:
> > task do_package_write_rpm: Succeeded
>
> Yes, well spotted, perf is failing to build with the 4.19 kernel.
>
> Thanks for the other fixes, they improve things a lot. This brings us
> down to the above perf issue with 4.19 and:
>
> a) KMACHINE_qemuarm = "qemuarma15" not working?

I mentioned on IRC that the kernel parts are working for me, and
by that, the mapping to the qemuarma15 bsp definition and hence
configs is correct.

I seem to be missing some machine definition parts to get a valid
userspace to work against that new kernel .. is there a series of
patches that I'm missing ?

>
> b) qemuarm64 graphics not working with 5.0

I don't think I've ever had this working, but I'll have a look at it as well.
But more eyes on something like this, will definitely help.

Bruce

>
> Getting there! :)
>
> Cheers,
>
> Richard
>
>
>


-- 
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II