|
From: | Anthony Liguori |
Subject: | Re: [Qemu-devel] [PATCH 2/5] boards: rename machine type functions |
Date: | Fri, 24 Feb 2012 10:57:40 -0600 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.23) Gecko/20110922 Lightning/1.0b2 Thunderbird/3.1.15 |
On 02/24/2012 10:51 AM, Luiz Capitulino wrote:
On Fri, 24 Feb 2012 10:15:52 -0600 Anthony Liguori<address@hidden> wrote:On 02/24/2012 10:12 AM, Peter Maydell wrote:On 24 February 2012 14:13, Luiz Capitulino<address@hidden> wrote:Perform the following renames: o qemu_register_machine() -> machine_register() o find_machine() -> machine_find() o find_default_machine() -> machine_find_default() Signed-off-by: Luiz Capitulino<address@hidden> 52 files changed, 96 insertions(+), 96 deletions(-)I said I was going to nack the next change-whole-tree change, so here it is: nack. At some point we're going to have to actually start converting boards to be QOM objects themselves, but (a) I hope we can make that incremental so we can do things in batches the way we did with MemoryRegion conversions and (b) until then the qemu_register_machine rename just looks like unnecessary churn to me.I think I agree with you here. I don't see the value compared to the churn here given that we're going to have to touch all of this again anyway soon.Do you guys see value in patch 5/5, which moves the machine function to the boards file?
Yes. And it's not that the other practicals aren't useful, it's just that I agree with Peter's valid point that the churn outweighs the benefit right now.
Regards, Anthony Liguori
If you don't I'll keep only patch 1/5 (which can be submitted to qemu-trivial).
[Prev in Thread] | Current Thread | [Next in Thread] |