Re: [PATCH 1/5] kernel-configs: switch to using defconfigs

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Fri, May 19, 2017 at 10:39 AM, Eric Biggers <ebiggers3@xxxxxxxxx> wrote:
> Hi Amir,
>
> On Fri, May 19, 2017 at 10:05:35AM +0300, Amir Goldstein wrote:
>> On Thu, May 18, 2017 at 9:44 PM, Theodore Ts'o <tytso@xxxxxxx> wrote:
>> > On Wed, May 17, 2017 at 04:32:29PM -0700, Eric Biggers wrote:
>> >> From: Eric Biggers <ebiggers@xxxxxxxxxx>
>> >>
>> >> Currently, the kernel configs for kvm-xfstests and gce-xfstests are
>> >> difficult to maintain because they are so long.  Fix this by changing
>> >> them to 'defconfig' format, so they only list the non-default settings.
>> >>
>> >> As documented, there is one additional step introduced when building a
>> >> kernel for the first time: you now must run 'make olddefconfig' after
>> >> copying a defconfig file to .config.  However I think this is an
>> >> acceptable tradeoff for having much more maintainable config files.
>> >> Running 'make olddefconfig' also means that people won't have to answer
>> >> random config questions if they're testing the latest kernel.
>> >>
>> >> This change was scripted by running 'make savedefconfig' for each config
>> >> within the appropriate source tree (stable/linux-X.X-y).
>> >>
>> >> Signed-off-by: Eric Biggers <ebiggers@xxxxxxxxxx>
>> >
>> > Thanks, applied.  (This patch was 1.4MB, so it would have been
>> > rejected by vger.kernel.org.)
>> >
>>
>> Ted,
>>
>> Applied where? github is 2 weeks behind. kernel.org is 2 weeks behind.
>>
>> BTW, I am curious to know why the kernel-configs in this repository have
>> no obvious debug config options.
>>
>> Aren't you guys testing regularly with lockdep etc? does everyone carry
>> their own home brewed debug configs like I do [1]?
>>
>> What do you say about including -debug variants to the maintained
>> kernel-configs in this repo? or turn on the obvious debugging features
>> by default?
>>
>
> There are already a number of debugging options turned on in the 4.9 x86_64
> config, including lockdep, various other CONFIG_DEBUG_* options, and the crypto
> self-tests.  It definitely should still be improved, though.  I've been planning
> to go through it again and see if there are any other debug options that should
> be turned on, as well as unused features which should be disabled.  Also the
> older configs tend to be outdated and should be synced up with the latest

Right, I actually diffed my debug config with latest and missed the
fact that lockdep
was turned on in upstream sometime after I forked my config.

Interesting configs I found in my diff you may want to consider:
-CONFIG_MESSAGE_LOGLEVEL_DEFAULT=7
-CONFIG_DYNAMIC_DEBUG=y
-CONFIG_DETECT_HUNG_TASK=y
-CONFIG_DEBUG_INFO_REDUCED=y
-CONFIG_HARDENED_USERCOPY=y
-CONFIG_DEBUG_ATOMIC_SLEEP=y

Cheers,
Amir.
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux