qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v3 1/8] target-ppc: Add helper for KVM_PPC_RTAS_


From: Laszlo Ersek
Subject: Re: [Qemu-devel] [PATCH v3 1/8] target-ppc: Add helper for KVM_PPC_RTAS_DEFINE_TOKEN
Date: Fri, 30 Aug 2013 22:05:38 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130806 Thunderbird/17.0.8

Hello Benjamin,

On 08/27/13 02:43, Benjamin Herrenschmidt wrote:

> Am I dreaming ? Those patches were written about a year ago and you guys
> are still nitpicking on names ? They should have been merged a LONG time
> ago ...
> 
> I'm seriously wondering how people get anything done with qemu KVM when
> it takes about a year of bike shed painting to get the most basic
> functionality merged.

(apologies for answering a rhetorical (non-)question)

"people" don't get anything done here. A handful of maintainers are
responsible for 80% of the changes (who can easily get in patches,
frequently without review, based on personal connections & trust between
each other), and the remaining 20% is done by hundreds of lowly droids,
including myself, who are put through the bikeshedding that you so aptly
describe (reposts up to v10 are commonplace).

http://thread.gmane.org/gmane.comp.emulators.qemu/228260
http://pastebin.com/Q85q2MUj

This two-caste process is celebrated as the pinnacle of quality assurance.

http://wiki.qemu.org/Contribute/SubmitAPatch

> Alex, you know, a maintainer's job is not to bounce a patch over and
> over again until it looks EXACTLY what you would have written in the
> first place. If that was the case you may as well write it yourself.

Now why are these thoughts so familiar... Oh wait:

http://thread.gmane.org/gmane.comp.emulators.qemu/230604/focus=230654

Many powerful people on the list are firmly and unwaveringly enamored of
their own voices.

> In this specific case, it's even more annoying because that comment
> could have been done ages ago, and because the name you propose is even
> more horrible than what was there... If you are really keen about that
> pick up at least something that doesn't suck such as
> 
>       kvmppc_define_rtas_kernel_token()
> 
> What you propose is even more confusing as to what the function is for.
> 
> At the end of the day, I despair though. Really.

qemu-devel is possibly more in-bred and more elitist than lkml.

If you don't have the networking capital, $DEITY help you with your
patches. (Well illustrated by several recent series of a submitter with
spotless track record, but maybe not with the necessary networking
capital, being ignored for months.)

Note that with this email I'm not trying to elevate myself to your level
-- I'm just a nameless droid as I said, but I can't help notice the
overlap between your points and mine.

qemu maintainers: when a Linux arch maintainer calls your process
"systematic bikeshed painting", maybe you should try to get over
yourselves once in a while.

Laszlo, butthurt




reply via email to

[Prev in Thread] Current Thread [Next in Thread]