ACK/cmnt: [PATCH v2 0/3][SRU Bionic] Decode ARM CPER records in kernel

Stefan Bader stefan.bader at canonical.com
Tue Jun 12 09:13:26 UTC 2018


On 11.06.2018 19:00, dann frazier wrote:
> On Mon, Jun 4, 2018 at 3:59 PM, Stefan Bader <stefan.bader at canonical.com> wrote:
>> On 23.05.2018 17:05, dann frazier wrote:
>>> Clean cherry-picks from upstream. Tested on a QDF2400 (the only ARM
>>> system I know of that supports EINJ, needed for simulating errors).
>>>
>>> v2: Explicitly enable CONFIG_UEFI_CPER_ARM
>>>
>>> Tyler Baicar (2):
>>>   efi: Move ARM CPER code to new file
>>>   efi: Parse ARM error information value
>>>
>>> dann frazier (1):
>>>   UBUNTU: [Config] CONFIG_UEFI_CPER_ARM=y
>>>
>>>  debian.master/config/config.common.ubuntu |   1 +
>>>  drivers/firmware/efi/Kconfig              |   5 +
>>>  drivers/firmware/efi/Makefile             |   1 +
>>>  drivers/firmware/efi/cper-arm.c           | 356 ++++++++++++++++++++++
>>>  drivers/firmware/efi/cper.c               | 122 +-------
>>>  include/linux/cper.h                      |  48 +++
>>>  6 files changed, 414 insertions(+), 119 deletions(-)
>>>  create mode 100644 drivers/firmware/efi/cper-arm.c
>>>
>>
>> Somehow more nice to have than required but ok, limited to specific arch and not
>> too complex either. So only thing I probably would change is to have the patch
>> changing the ubuntu config last.
> 
> Sorry, just noticed your feedback here. I actually used to put the
> [Config] changes last, but noticed they were getting applied earlier.
> Turns out there was a reason - bisection. IIRC, the idea is to avoid
> having a commit where the Kconfig setting exists, but we haven't yet
> defined it in the Ubuntu config.

You are right and I am not sure what exactly I was thinking when I wrote that
comment. I guess it was to have everything in place and then activate it. Just
that this breaks normal build as soon as this requires to make some decision
(turn on/off/build as module). Bisection itself is more independent as one needs
to twiddle config options often between steps.
So maybe (as of todays thinking) the safest option is to put the ubuntu config
change either as the first change (as long as the option(s) do not exist in
Kconfig they would just be ignored) or directly before the commit that makes the
Kconfig change.

-Stefan
> 
>   -dann
> 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <https://lists.ubuntu.com/archives/kernel-team/attachments/20180612/c7e95e38/attachment.sig>


More information about the kernel-team mailing list