On 13 July 2017 at 08:27, Alistair Francis <alistai...@gmail.com> wrote: > On Tue, Jul 11, 2017 at 6:35 PM, Peter Maydell <peter.mayd...@linaro.org> > wrote: >> On 11 July 2017 at 15:33, Alistair Francis <alistai...@gmail.com> wrote: >>> On Tue, Jul 11, 2017 at 1:17 PM, Peter Maydell <peter.mayd...@linaro.org> >>> wrote: >>>> +#define TYPE_MPS2_MACHINE "mps2" >>> >>> This public name seems too common as well. I am just worried it'll >>> conflict with something one day. >> >> I can see the issue, but on the other hand the type name is not >> external ABI, so we can always rename it later if we have to. > > That's true, but it's still not ideal to be changing device names willy nilly. > >> The board names themselves (mps2-an385 and mps2-an511) on the >> other hand we have to get right from the start. I'm not sure >> tacking an arm- on the front of those helps; they're already >> pretty ungainly. > > I think the boards are already long enough to be unique and probably > don't need an 'arm-'. Although I think having the vendor name does > clarify what the board is. Imagine a user is querying QEMU for what > boards it supports and sees a board name they don't recognise. It will > be a lot easier to figure out what board this is is you search "ARM > MPS2" instead of just MPS2.
"ARM" is in the long description you get from -M help: mps2-an385 ARM MPS2 with AN385 FPGA image for Cortex-M3 mps2-an511 ARM MPS2 with AN511 DesignStart FPGA image for Cortex-M3 (compare existing realview-pb-a8 ARM RealView Platform Baseboard for Cortex-A8 etc) Having thought about it a bit I think there are two options: (1) stick with what we have here, just using "mps2" in type and filenames (2) use "arm-mps2" instead (in filenames, typenames and the board names) since I don't much like having the base machine type be 'arm-mps2' but the board names just 'mps2-an385' &c. But of the two I overall prefer (1) still. thanks -- PMM
