Comment 13 for bug 1822870

On Fri, 12 Apr 2019, Michael Ranweiler wrote:

> Hi Manoj - were you going to spin a new test kernel with the extra patches
> or did you want to stick where we are?  It's getting late, so I wanted to
> make sure that one of those versions could go in.  The extra 3 aren't, I
> think, critical, or could be added later, though it'd be good to have them.
> Also wanted to get them posted - should I go ahead and do that?

The latest test kernel that is available in the PPA
https://launchpad.net/~ubuntu-power-triage/+archive/ubuntu/lp1822870
includes all that patches in the tarball and the patch "51c3c62b58b3
powerpc: Avoid code patching freed init sections."' In comment #11
Ellerman@ibm verified that this kernel passes his tests in the simulator.

We would like to keep the patchset to the min number of required critical
patches. The patches to clean up code etc could trigger the kernel team to
reject the SRU because these are not critical fixes.

If you are satisfied with the set of patches that are currently
in the PPA kernel I can submit these for SRU review.

Here is the list of patches in the PPA kernel, please note the shaids wont
match with yours (upstream) because this is from the Ubuntu 4.15 kernel
source.

8db52d2ad2ce (HEAD -> spectre-1822870) powerpc: Avoid code patching freed
init sections
9358c01bc816 KVM: PPC: Book3S: Add count cache flush parameters to
kvmppc_get_cpu_char()
4c3a23aaf32f powerpc/security: Fix spectre_v2 reporting
9e74ec03431a powerpc/fsl: Add nospectre_v2 command line argument
60ec56d8b314 powerpc/fsl: Fix spectre_v2 mitigations reporting
5c2a9f5f9d9e powerpc/powernv: Query firmware for count cache flush
settings
ce74f68a8bfd powerpc/pseries: Query hypervisor for count cache flush
settings
e4ebd9989cdc powerpc/64s: Add support for software count cache flush
1d49623704bf powerpc/64s: Add new security feature flags for count cache
flush
7aa198fb1644 powerpc/asm: Add a patch_site macro & helpers for patching
instructions
4b97b64ac5cd powerpc/lib/feature-fixups: use raw_patch_instruction()
6508ea530cb7 powerpc/lib/code-patching: refactor patch_instruction()
f06866d8777e powerpc/64: Make meltdown reporting Book3S 64 specific
a40b1c85cc50 powerpc/64: Call setup_barrier_nospec() from setup_arch()
c9c86099ae24 powerpc/64: Add CONFIG_PPC_BARRIER_NOSPEC
82cf510fab3f powerpc/64: Make stf barrier PPC_BOOK3S_64 specific.
b4d6e3d336c3 powerpc/64: Disable the speculation barrier from the command
line
4b94b84138dd powerpc64s: Show ori31 availability in spectre_v1 sysfs file
not v2
6d63c784648f powerpc/64s: Enhance the information in cpu_show_spectre_v1()
e556e8721b0f powerpc/64: Use barrier_nospec in syscall entry
b09734c0aef5 powerpc: Use barrier_nospec in copy_from_user()
59abe6f37d98 powerpc/64s: Enable barrier_nospec based on firmware settings
f9c9cbba5139 powerpc/64s: Patch barrier_nospec in modules
b7860091936e powerpc/64s: Add support for ori barrier_nospec patching

>
> Thanks!
>
> Mike
>
> Mike Ranweiler
> <email address hidden>
>
>
> -----Michael Ranweiler/Rochester/IBM wrote: -----To: Manoj Iyer
> <email address hidden>
> From: Michael Ranweiler/Rochester/IBM
> Date: 04/11/2019 12:21AM
> Cc: <email address hidden>
> Subject: Re: Could you please test the PPA kernel?
>
> Manoj,
>
> I tested it, but we're missing a patch.  So I will attach that after I've
> had a chance to retest the series - sorry for this, we reviewed the list,
> but this one got dropped.
>
> Thanks,
>
> Mike
>
> Mike Ranweiler
> <email address hidden>
>
>
> -----Manoj Iyer <email address hidden> wrote: -----To: Michael Ranweiler
> <email address hidden>
> From: Manoj Iyer <email address hidden>
> Date: 04/10/2019 03:48PM
> Cc: <email address hidden>
> Subject: Could you please test the PPA kernel?
>
> Michael,
>
> I have test kernel with the patches cherry-picked/backported to 4.15
> bionic. We do not have a dd2.3 hw in-house, could you please validate that
> this kernel works for you and report back in the bug report?
>
> https://launchpad.net/~ubuntu-power-triage/+archive/ubuntu/lp1822870/
>
> Once I get the validation results I can send a pull request to the kernel
> team.
>
> Thanks
> --
> ============================
> Manoj Iyer
> Ubuntu/Canonical
> ============================
>
>
>
>

--
============================
Manoj Iyer
Ubuntu/Canonical
============================