[Skiboot] [PATCH v2 0/3] Support PCI Virtual Device

[Alistair Popple]

On Wed, 7 Sep 2016 06:13:39 PM Stewart Smith wrote:
> Alistair Popple <alistair at popple.id.au> writes:
> > On Thu, 11 Aug 2016 12:12:35 PM Gavin Shan wrote:
> >> The patches support PCI virtual device. The NVLinks are presented to 
Linux
> >> kernel via PCI virtual (emulated) device. As we're going to support 
NVLink2
> >> pretty soon, we need support PCI virtual device in common layer so that 
it
> >> can be shared by NVLINK1/2.
> >> 
> >> Changelog
> >> =========
> >> v1 -> v2:
> >>    * Rebased to master branch per requested by Stewart. No code changes
> >>      since v1. The last commit is f102c9d19078 ("Merge skiboot-5.3.1
> >>      release notes").
> >>    * Had quick test on Garrison1 and no obvious issues found.
> >
> > We should run actual GPU workloads to test any NPU changes (eg. a CUDA 
> > bandwidth test - let me know if you're not sure how). A simple boot test 
is 
> > not enough as it won't exercise many of the code paths that have been 
changed 
> > here. It seems patch 3 breaks something as the nVidia driver no longer 
works 
> > with nvlink on master - could you please take a look at this? Thanks!
> 
> I'm thinking we need something in op-test-framework that runs things
> like a CUDA workload (if possible), FWTS and a smoke test with HTX.

I believe at least one version of HTX includes some CUDA tests.

> However... that does run into the whole problem of having an OS to run
> all of these packages on, which creates a bit more work to make it
> happen.

Which for nVidia is even more fun as we need to have the out-of-tree driver 
complied against the kernel. Although buildroot does have a package for 
that...

> Do we have a victim^Wvolunteer to maybe work on that? :)

... which leads me to dob Sam B in :)

I know at one point he was working on a buildroot based system for testing 
qemu/kvm. Not sure how far he got but it might be a useful base as having 
automated testing for this stuff would be wonderful.

Regards,

Alistair

> 
> > Stewart - we may need to revert "[PATCH v2 3/3] hw/npu: Use PCI virtual 
> > device" if we can't get this fixed soon - will let you know.
> 
> ack.
> 
>