Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 19 Nov 2015 10:08:57 -0500
From:      Pedro Giffuni <pfg@freebsd.org>
To:        Dan Partelly <dan_partelly@rdsor.ro>
Cc:        Adrian Chadd <adrian.chadd@gmail.com>, freebsd-current <freebsd-current@freebsd.org>
Subject:   Re: DDB patches 
Message-ID:  <FE623ABB-F5D7-43AB-A7AA-8D3E5C09CE98@freebsd.org>
In-Reply-To: <22918FB9-4DC2-438D-B9F0-C295DD273B50@rdsor.ro>
References:  <B6FDC307-13EB-48AC-8130-C597AB8C06F4@rdsor.ro> <B6BF97C8-9FE3-4ADE-A047-33AF0B879781@freebsd.org> <22918FB9-4DC2-438D-B9F0-C295DD273B50@rdsor.ro>

next in thread | previous in thread | raw e-mail | index | archive | help

> Il giorno 19/nov/2015, alle ore 04:57, Dan Partelly =
<dan_partelly@rdsor.ro> ha scritto:
>=20
> Hey Pedro,
>=20
> Thanks a lot , mate.=20
>=20
> I=E2=80=99m reluctant to put it up as a PR, since some PR are =
outstanding for years. =20
>=20

Well, that=E2=80=99s the way the project works: you cannot really depend =
on me, or
anyone else keeping old patches around. If you want a record of your
submission bugzilla is the place to keep it. And of course there is no =
guarantee
anyone will look at it but your chances are much better in bugzilla than
in a mailinglist.



> Adrian,
>=20
> since Pedro has issue with hardware, could you try the patch and give =
a resolution on it ? I reviewed it mentally (no FreeBSD atm machine on =
which I could actually  patch the kernel)  and apart style changes it =
looks OK . Physically i can test it again fro a couple of days.

Mental reviews don=E2=80=99t count much: if you are not running FreeBSD =
and standing
behind your patch the chances of being taking seriously are slim.


>  Getting this reviewed & tested / committed or rejected would give me =
an idea on how things actually work around here. This is actual code =
which you can commit or reject not commentaries only like in the thread =
regarding the binary code reuse. =20
>=20
>=20

I recall you stated the patch was =E2=80=9Cnot ready=E2=80=9D when you =
posted it. I haven=E2=80=99t really
done anything to say it is ready. Unless someone else finds time to do =
real
testing it won=E2=80=99t happen.

Adrian tends to do some particularly valuable contributions to the =
project. I
would prefer if he spends his time on more important tasks.

> [qute from libxo thread ]
>>> It's all fine and good making technical decisions based on drawings =
and handwaving and philosophizing, but at some point someone has to do
>>> the code.
>>> The reason is simple - someone offered to do the work and push it =
through. This isn't a commercial thing where we get to make project =
>>decisions and allocate resources - the juniper folk came up with a =
solution that
>=20
> Once I see how things work around here once someone wrote  the code,  =
and get this done one way or another , we could proceed to the =
libification of ifconfig, should you so desire, and you believe we can =
all benefit from it.=20
>=20

Wrong approach. You can=E2=80=99t really blackmail someone into taking =
your changes.

Things work like this:

- You discuss your idea and try to get some consensus in the =
lists/IRC/conferences.
- You *write* a specific proof of concept and get it discussed.
- You finish your prototype.
- Your work gets rejected until you get something some committer is =
willing to support.
- When there are no objections and a committer feels like it, your work =
gets committed,
 which doesn=E2=80=99t necessarily mean it will stay.
- You are expected to maintain it.

Libxo already went through this process.

We are particularly NOT interested in code where the original =
contributor will walk
away as soon as he/she receives criticism or has plans that do not match =
ours.
If this is not your ideal workflow =E2=80=A6 fork your own BSD, a lot of =
intelligent
people do just that.

Pedro.

>=20
> Dan
>=20
>=20
>=20
>> On 19 Nov 2015, at 11:17, Pedro Giffuni <pfg@freebsd.org> wrote:
>=20
>>=20
>> Hello;
>>=20
>>> Il giorno 19/nov/2015, alle ore 02:34, Dan Partelly =
<dan_partelly@rdsor.ro> ha scritto:
>>>=20
>>> Hey Pedro,
>>>=20
>>> some times ago you got some DDB patches from me in which I added =
relational ops support from it. The patch was a bit clobbered,=20
>>> but last I know you cleaned it up and put it somewhere on =
freebsd.org (prolly your page) up for review.=20
>>>=20
>>=20
>> It=E2=80=99s here:
>> https://people.freebsd.org/~pfg/patches/ddb.patch
>>=20
>> I haven=E2=80=99t tested it though.
>>=20
>>> Could you or Adrian review the patch set , and if it is OK =
potentially proceed with a commit ? Or if it is not ok for a commit , =
please advice on a follow up.=20
>>>=20
>>=20
>> I am having hardware issues so I won=E2=80=99t be able to do much in =
a while.
>> Perhaps you should review it and submit it as a PR.
>>=20
>> Pedro.
>>=20
>=20




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?FE623ABB-F5D7-43AB-A7AA-8D3E5C09CE98>