[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH for-2.9 v2] Move target-* CPU file into a target
From: |
Eduardo Habkost |
Subject: |
Re: [Qemu-devel] [PATCH for-2.9 v2] Move target-* CPU file into a target/ folder |
Date: |
Wed, 14 Dec 2016 15:55:46 -0200 |
User-agent: |
Mutt/1.7.1 (2016-10-04) |
On Wed, Dec 14, 2016 at 06:42:49PM +0100, Paolo Bonzini wrote:
>
>
> On 14/12/2016 17:11, Markus Armbruster wrote:
> > Thomas Huth <address@hidden> writes:
> >
> >> On 14.12.2016 12:33, Peter Maydell wrote:
> >>> On 14 December 2016 at 11:21, Markus Armbruster <address@hidden> wrote:
> >>>> Thomas Huth <address@hidden> writes:
> >>>>
> >>>>> 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.
> >>>>> To disburden the main folder a little bit, let's move the target-xxx
> >>>>> folders into a dedicated target/ folder, so that target-xxx/ simply
> >>>>> becomes target/xxx/ instead.
> >>>>
> >>>> The root directory is crowded, but hardly because of its 18 target-*
> >>>> entries:
> >>>>
> >>>> $ git-ls-tree --name-only v2.8.0-rc3 | wc -l
> >>>> 167
> >>>>
> >>>> Getting the 18 target-* out of the way would help a little, but I hate
> >>>> making the directory tree deeper.
> >>
> >> We already got folders in folders in a lot of places, so this is not
> >> increasing the maximum depths ;-)
> >
> > Let me rephrase: I hate making directory tree *branches* deeper.
> >
> >>>> We have 110 files in the root directory. Moving some of them to
> >>>> subdirectories looks more promising to me.
> >>>
> >>> True, but Thomas has done the work here already, so I don't think
> >>> we lose anything by merging it rather than throwing it away.
> >>
> >> Yes, and please consider that there are more targets coming soon ...
> >> RISC-V, AVR, ... and I am pretty sure that we won't stop there.
> >
> > Even doubling the number of target-* won't make our root directory
> > appreciatively messier.
>
> Cutting down the number of toplevel directories by 1/3rd makes some
> sense though.
>
> >> With regards to the other *.c files in the main directory, I agree with
> >> Markus, we should move some of them to suitable subdirectories, too. I
> >> could have a look at that when I've got some spare minutes...
> >
> > Thanks in advance!
>
> Yeah, there are some easy moves to migration/ and block/. (One old idea
> was also to move block layer drivers to block/{formats,protocols}/ to
> make room for the core block layer in block/). Also perhaps new
> directories monitor/, accel/ and tools/.
I was considering moving xen-* and kvm-* to xen/ and kvm/
directories, but an accel/ directory would be even better. I plan
to send a patch to do this:
tcg/ => accel/tcg/
kvm-* => accel/kvm/* (except for stubs)
xen-* => accel/xen/* (except for stubs)
qtest.c => accel/qtest/qtest.c
--
Eduardo