[PULL,Zesty] arm64 fixes for zesty

Submitted by dann frazier on March 8, 2017, 7:59 p.m.

Details

Message ID CALdTtnueGPXkR3KNoTGHugCD4A0=COv43Pv_hxzPfUU6PAiPfg@mail.gmail.com
State New
Headers show

Pull-request

git://git.launchpad.net/~dannf/ubuntu/+source/linux/+git/linux for-zesty

Commit Message

dann frazier March 8, 2017, 7:59 p.m.
Here's a set of miscellaneous arm64 changes for zesty. It includes
fixes (erratum/quirking) for the QDF2400 platform, a follow-up fix for
the previously merged VFIO backport, and a GICv3 ITS performance fix.

All are cherry picks from upstream, with minor backporting required
for the pl011 workaround.

The following changes since commit ba46da7c1cc57d83f6af66bfe8f10516151c7923:

  powerpc: Update to new option-vector-5 format for CAS (2017-03-08
10:29:45 -0700)

are available in the git repository at:

  git://git.launchpad.net/~dannf/ubuntu/+source/linux/+git/linux for-zesty

for you to fetch changes up to a1d65588ad163760adbc548ea208f8b2bc6cb06e:

  irqchip/gic-v3-its: Enable cacheable attribute Read-allocate hints
(2017-03-08 12:19:29 -0700)

----------------------------------------------------------------
Eric Auger (1):
      iommu: Fix static checker warning in iommu_insert_device_resv_regions

Shanker Donthineni (2):
      arm64: Avoid clobbering mm in erratum workaround on QDF2400
      irqchip/gic-v3-its: Enable cacheable attribute Read-allocate hints

Sinan Kaya (1):
      PCI: Add ACS quirk for Qualcomm QDF2400 and QDF2432

Sudeep Holla (1):
      tty: pl011: Work around QDF2400 E44 stuck BUSY bit

 arch/arm64/mm/proc.S             |  2 +-
 drivers/acpi/spcr.c              | 23 +++++++++++++++++++++++
 drivers/iommu/iommu.c            |  2 +-
 drivers/irqchip/irq-gic-v3-its.c | 10 +++++-----
 drivers/pci/quirks.c             | 23 +++++++++++++++++++++++
 drivers/tty/serial/amba-pl011.c  | 66
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-------
 6 files changed, 112 insertions(+), 14 deletions(-)

Comments

Seth Forshee March 8, 2017, 9:37 p.m.
On Wed, Mar 08, 2017 at 12:59:59PM -0700, dann frazier wrote:
> Here's a set of miscellaneous arm64 changes for zesty. It includes
> fixes (erratum/quirking) for the QDF2400 platform, a follow-up fix for
> the previously merged VFIO backport, and a GICv3 ITS performance fix.

These patches all need buglinks, otherwise they look fine.
dann frazier March 8, 2017, 10:04 p.m.
On Wed, Mar 8, 2017 at 2:37 PM, Seth Forshee <seth.forshee@canonical.com> wrote:
> On Wed, Mar 08, 2017 at 12:59:59PM -0700, dann frazier wrote:
>> Here's a set of miscellaneous arm64 changes for zesty. It includes
>> fixes (erratum/quirking) for the QDF2400 platform, a follow-up fix for
>> the previously merged VFIO backport, and a GICv3 ITS performance fix.
>
> These patches all need buglinks, otherwise they look fine.

Ah, OK. I'll file bugs and resubmit. Also, I should probably know this
by now, but at what stage in the release does the buglink requirement
set in? I think historically I've used lazy discovery by submitted
cherry-picks w/o buglinks until I get an e-mail like this one, then
transition to buglink mode for the rest of the cycle :)
Seth Forshee March 9, 2017, 1:11 p.m.
On Wed, Mar 08, 2017 at 03:04:06PM -0700, dann frazier wrote:
> On Wed, Mar 8, 2017 at 2:37 PM, Seth Forshee <seth.forshee@canonical.com> wrote:
> > On Wed, Mar 08, 2017 at 12:59:59PM -0700, dann frazier wrote:
> >> Here's a set of miscellaneous arm64 changes for zesty. It includes
> >> fixes (erratum/quirking) for the QDF2400 platform, a follow-up fix for
> >> the previously merged VFIO backport, and a GICv3 ITS performance fix.
> >
> > These patches all need buglinks, otherwise they look fine.
> 
> Ah, OK. I'll file bugs and resubmit. Also, I should probably know this
> by now, but at what stage in the release does the buglink requirement
> set in? I think historically I've used lazy discovery by submitted
> cherry-picks w/o buglinks until I get an e-mail like this one, then
> transition to buglink mode for the rest of the cycle :)

You should have buglinks for everything after feature freeze. There's no
reason you can't include them before then though ;-)