On Thu, Aug 02, 2012 at 08:45:18AM +0100, Lee Jones wrote:
Over time, the requests for Maintainers have Snowballed (pun intended). My task now seems to be enabling drivers for Device Tree _and_ fix all associated driver bugs, including any requested restructuring and API
One thing to bear in mind with device tree is that it's all about defining an ABI - it's supposed to be stable and OS agnostic so it puts a lot of pressure on to really address quality problems that become visible in the bindings. This stuff is much less critical with platform data, it's relatively easy to change.
adoption. What you fail to notice is that I am only one person, and hopping all over the code-base trying to do everyone's bidding is no mean feat. In
It's fairly obvious that it's only you, or at least only you posting stuff (I know sometimes there are bigger teams behind people) - the pressure you're under to get something in is very clear. A big part of what I'm saying here is that it would be really helpful if could you slow down a bit, discuss problems more and avoid cutting corners so much. This is likely to save you time overall, you'll have a much higher success rate and you'll get much better feedback if it's clear what's going on and that there's an interest in understanding the issues.
If you need to focus on device tree enablement and there are underlying problems in the subsystems you're looking at perhaps you need to push back on whoever's asking you to do the work and say you need the domain experts to pitch in and help you out.
reality I am no more obliged to fix driver bugs than you are. In fact as the Maintainer of some of these subsystems, perhaps you could even help out a bit?
You're not telling us about the problems you see so it's very difficult for anyone to help you.
For example with this patch the only information you've sent is the patch and the fact that you're seeing the error you're ignoring going off on the system you're working with (which I had to ask to find out about...). You've not shown the error message or provided any other hint which would help anyone understand why the error might be occuring and what a good solution to the problem is. Ola's guess seems very likely but nobody's got enough information to confirm that unless there's been some off-list ST/Linaro communication.
Obviously with the stuff that's device specific you also have to contend with the fact that you're working on hardware which just isn't all that widely available and quite possibly has closed datasheets.
We are all trying to do good things here. Please try not to be too over- critical. We all know Mainlining is a good thing. Perhaps less people would be so frightened of it, thus more people would do it if the reviews weren't such a ball-ache ( for want of a better expression :) ).
This is a two way thing - one of the tools that maintainers have to try to drive up the quality of the code is to push back on poor quality submissions and devote less bandwith to sources of those submissions.
It is true that there's a bunch of people who seem to view review as being just an annoying obstacle to be dealt with with the minimum possible effort but in practice all that does is make things more painful for everyone, they do tend to be more noticable because "applied, thanks" doesn't make for a big thread.