On Wed, 06 Sep 2017 15:53:14 +0200, Richard Brown wrote:
On 6 September 2017 at 14:55, Jean Delvare <firstname.lastname@example.org> wrote:
Takashi, my own recommendation: if nobody complained about what you were doing so far, then it must be right, stick to it, unless YOU think there's a problem with it.
Takashi, my own recommendation is to favour reducing your work and the variance between the Leap Kernel and the SLE Kernel as much as possible. But I, like the rest of the community, obviously trust you if you think the additional work is justified.
Thanks both of you, I decided to keep the current scheme, as there is no obvious problem with it, so far. That is, Leap gets slightly more frequent updates than SLE, but the release is aligned. It's good for both sides (users on Leap can get fixes more quickly, and SLE can get more test coverage).
My only additional stipulation as Chairman would be to add that the communities expectation is that the Leap kernel is as reliable as, and is 'broadly compatible' with, the SLE kernel. Hence my original reply to this thread, which suggests mimicking whatever is being done in SLE.
That's always OK with Leap kernel, as we share the very same code as SLE kernel. Generally speaking, they are "fully compatible" in the source code level. The major difference is the kernel config, e.g. Leap supports more drivers than SLE. Takashi -- To unsubscribe, e-mail: email@example.com To contact the owner, e-mail: firstname.lastname@example.org