On 11/30/2016 10:47 AM, Thomas Huth wrote: > We've currently got 18 architectures in QEMU, and thus 18 target-xxx > folders in the root folder of the QEMU source tree. More architectures > (e.g. RISC-V, AVR) are likely to be included soon, too, so the main > folder of the QEMU sources slowly gets quite overcrowded with the > target-xxx folders. > Thus I'd like to suggest that we move the target-xxx folders into > a dedicated target/ folder, so that target-xxx/ simply becomes > target/xxx/ instead. > > The first patch is the most important one here, it prepares the > build system to be able to deal with both, target-xxx and target/xxx > folders. Once that has been applied, each target architecture > could be moved to the target/ folder when it is convenient for the > subsytem maintainer to do the move (i.e. we do not have to move all > files immediately in one go). So if we agree to do this change and > the first patch has been committed, I could rather send the following > patches step by step to the maintainers so they can queue the patch > for their subsystem for a good point in time. And new architectures > like RISC-V and AVR could start in the target/ folder immediately > instead. > > So please provide some feedback: Is this a good idea? Or do you > consider this rather just as unnecessary code churn instead?
Given the low amount of manpower for the stable work, I think this will cause more harm than it helps. So my gut feeling is to keep it as is. Christian
