[vivid,SRU] Add X-Gene EDAC Driver

This patch set adds and enables the X-Gene EDAC driver and makes
additional
generic EDAC build system code cleanup.

BugLink: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1494357

The following changes since commit
c550dccb0f3996f87aae5f3352d3d287f1289483:

  UBUNTU: Ubuntu-3.19.0-29.31 (2015-09-09 16:15:35 +0100)

are available in the git repository at:

  git://kernel.ubuntu.com/cmagina/vivid-xgene.git 1494357

for you to fetch changes up to
ec70356f92beebf7b07930d3b003343f44aa6f57:

  UBUNTU: [Config] Add XGENE_EDAC, EDAC_SUPPORT and EDAC_ATOMIC_SCRUB
(2015-09-10 11:43:05 -0400)

----------------------------------------------------------------
Borislav Petkov (2):
      EDAC: Cleanup atomic_scrub mess
      EDAC, edac_stub: Drop arch-specific include

Craig Magina (1):
      UBUNTU: [Config] Add XGENE_EDAC, EDAC_SUPPORT and
EDAC_ATOMIC_SCRUB

Loc Ho (5):
      arm64: Enable EDAC on ARM64
      MAINTAINERS: Add entry for APM X-Gene SoC EDAC driver
      Documentation: Add documentation for the APM X-Gene SoC EDAC DTS
binding
      EDAC: Add APM X-Gene SoC EDAC driver
      arm64: Add APM X-Gene SoC EDAC DTS entries

 .../devicetree/bindings/edac/apm-xgene-edac.txt    |   78 ++
 MAINTAINERS                                        |    7 +
 arch/arm/Kconfig                                   |    2 +
 arch/arm/include/asm/edac.h                        |    5 +-
 arch/arm64/Kconfig                                 |    1 +
 arch/arm64/boot/dts/apm/apm-storm.dtsi             |   83 ++
 arch/mips/Kconfig                                  |    1 +
 arch/mips/include/asm/edac.h                       |    4 +-
 arch/powerpc/Kconfig                               |    2 +
 arch/powerpc/include/asm/edac.h                    |    4 +-
 arch/tile/Kconfig                                  |    1 +
 arch/tile/include/asm/edac.h                       |   29 -
 arch/x86/Kconfig                                   |    2 +
 arch/x86/include/asm/edac.h                        |    2 +-
 debian.master/config/config.common.ubuntu          |    3 +
 drivers/edac/Kconfig                               |   14 +-
 drivers/edac/Makefile                              |    1 +
 drivers/edac/edac_mc.c                             |    9 +-
 drivers/edac/edac_stub.c                           |    1 -
 drivers/edac/xgene_edac.c                          | 1254
++++++++++++++++++++
 20 files changed, 1461 insertions(+), 42 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/edac/apm-xgene
-edac.txt
 delete mode 100644 arch/tile/include/asm/edac.h
 create mode 100644 drivers/edac/xgene_edac.c

Thanks,

Looks good. There appears to be little chance of regression for 
non-arm64 platforms.

On 21.09.2015 15:53, Craig Magina wrote:
> git://kernel.ubuntu.com/cmagina/vivid-xgene.git 1494357

0001-EDAC-Cleanup-atomic_scrub-mess.patch
 This renames an inline function. For the in-kernel code not that much
 of a deal but if there are any external drivers anywhere which rely on
 asm/edac.h this API change may cause trouble.
0002-arm64-Enable-EDAC-on-ARM64.patch
 Looks ok.
0003-MAINTAINERS-Add-entry-for-APM-X-Gene-SoC-EDAC-driver.patch
 Looks safe.
0004-Documentation-Add-documentation-for-the-APM-X-Gene-S.patch
 Looks safe.
0005-EDAC-Add-APM-X-Gene-SoC-EDAC-driver.patch
 Not sure what exactly "dropped suggested merge changes" means but otherwise
 only adding a new driver. Should not cause regressions.
0006-arm64-Add-APM-X-Gene-SoC-EDAC-DTS-entries.patch
 Only device tree change, so at least no effect on other arches.
0007-EDAC-edac_stub-Drop-arch-specific-include.patch
 Dropping an include file. Should be harmless but is it needed?
0008-UBUNTU-Config-Add-XGENE_EDAC-EDAC_SUPPORT-and-EDAC_A.patch
 Looks ok.

So 0001 may pose a risk for breaking API. 0003 and 0004 might not strictly be
required but good to include for completeness and pose no risk. 0007 does not
look strictly required at first glance but maybe there is more to it than is
obvious and it should be safe as long as compile was tested on all supported
arches. The rest looks ok for the purpose of getting a new driver enabled.

-Stefan

Applied to Vivid master-next branch.

Cheers,
--
Luís