On 11/05/12 06:57, Somebody in the thread at some point said:
On Fri, 2012-05-11 at 00:46 +0200, Alexander Sack wrote:
On Fri, May 11, 2012 at 12:09 AM, Jon Medhurst (Tixy)tixy@linaro.org wrote:
On Thu, 2012-05-10 at 23:34 +0400, Andrey Konovalov wrote:
Now it is time to move the focus to the linux-linaro tree. For one week it will use the mainline tip as the base. Then, on next Thursday the most recent -rc will be selected as the base, and won't be changed until 12.05 is released. Most probably it will be v3.4-rc7.
I may have misunderstood but....
Doesn't this mean on next Wednesday you be tracking Wednesdays tip, then on Thursdays move back in time to this Sundays rc7 release?
Yeah, I wondered about the same. In general I am very suspicious if we say we would have to go back and feel we might duplicate work in a direction where we shouldn't invest...
How bad is the tip revision you aim for Andrey? Maybe we can check how well that works and if there are problems collaboratively try to fix that with the goal to release from tip?
Should we not release based on a specific Linux rc or final release rather than some random intermediate commit. It seems a lot neater and easier to communicate.
We don't have to take this 'tracking' thing too far :-)
Tracking is tracking, it's best done at whatever current basis head is.
But what is a 'release' here? For normal mortals it's when Linus puts out a tag without the -rc bit, due to Linus' judgement that the breakages are largely ironed out.
When we talk about applying the Linaro Release-industrial complex with its mailing lists and rules and dedicated staff to a 'release' of tracking content, especially on this febrile unified tree, I think there's a bit of an impedence mismatch coming.
Being on some -rc or some other intermediate HEAD isn't going to make much odds, either that kernel performs overall better than a recent one at another HEAD or it doesn't, that's the only "release criteria". If the current one performs best and is on a random HEAD commit, we certainly shouldn't wind it backwards to last -rc that performs worse just because that's "easier to communicate".
Likewise in unified case, there might not be much choice about which recent kernels had most LTs participating with workable content, if that's on an intermediate HEAD we can't be sniffy.
-Andy