On Thu, Jul 19, 2018 at 3:00 PM, Michael Orlitzky wrote:
> On 07/19/2018 05:51 PM, Ben Kohler wrote:
>> Hello,
>>
>> I'd like to propose adding USE=udev to our linux profiles (in
>> profiles/default/linux/make.defaults probably). This flag is already
>> enabled on desktop profiles but it also aff
Hi Mart,
Mart Raudsepp writes:
> That said, I would question such a choice. Does it technically not
> work or what's the problem with it?
It works partially. Most of the time they does not bulid.
The host OS handles /dev for Gentoo Prefix, be it mdev or udev.
> But it's up the prefix projec
Ühel kenal päeval, R, 20.07.2018 kell 17:01, kirjutas Benda Xu:
> Hi Mart,
>
> Mart Raudsepp writes:
>
> > That said, I would question such a choice. Does it technically not
> > work or what's the problem with it?
>
> It works partially. Most of the time they does not bulid.
That sounds like
On Thu, 19 Jul 2018 23:51:05 -0400 Aaron Bauman wrote:
> You are the minimalist... Not the rest. Provide a reasonable scenario please.
Such setup is quite simple: secure server or container usually for
a single task with minimal setup of packages and USE flags to
reduce attack surface.
Best rega
On Fri, Jul 20, 2018 at 1:58 AM Michael Orlitzky wrote:
>
> On 07/20/2018 01:06 AM, Mart Raudsepp wrote:
> >>
> >> * They can't be undone. It's next to impossible for me to undo
> >> USE=udev when set in a profile that is inherited by all others.
> >
> > You set USE=-udev in your make.conf.
On 07/19/18 20:54, Mikle Kolyada wrote:
> +1. widely used profiles should have as least flags enabled by default
> as possible, I would not be happy with +udev on my servers.
>
I disagree with this premise. The default and most widely used profiles
should fit the most common use cases.
I'd be
On 07/19/18 22:40, Benda Xu wrote:
>
> To represent the Gentoo Prefix users, we would like to have USE=udev
> turned off or even hard masked on linux-prefix profiles.
>
> Yours,
> Benda
>
I believe this is an argument in favor of moving the default to profiles
then, out of IUSE defaults, right?
On 07/19/18 23:04, Michael Orlitzky wrote:
>
> No I'm not. I'm saying add them per-package, because it's a better
> design. We have package.use in profiles now, not just IUSE defaults.
>
> Global defaults have problems:
>
> * They can't be undone. It's next to impossible for me to undo
>
Hi,
July 20, 2018 2:26 PM, "Ben Kohler" wrote:
> On 07/19/18 23:04, Michael Orlitzky wrote:
>
>>
>>
>> If you really want to enable it globally after being told that it's bad
>> engineering and downright annoying, go do it in a profile that I can
>> avoid and not "linux".
>
> I believe you'r
On Fri, Jul 20, 2018 at 8:39 AM wrote:
>
>
> Why not introducing a new level in the hierarchy ? Something like "common"
> could be fit.
>
> default/linux/amd64/13.0
> default/linux/amd64/13.0/common
> default/linux/amd64/13.0/common/desktop
> default/linux/amd64/13.0/common/developer
> ...
>
> By
July 20, 2018 2:55 PM, "Rich Freeman" wrote:
> On Fri, Jul 20, 2018 at 8:39 AM wrote:
>
>> Why not introducing a new level in the hierarchy ? Something like "common"
>> could be fit.
>>
>> default/linux/amd64/13.0
>> default/linux/amd64/13.0/common
>> default/linux/amd64/13.0/common/desktop
>
On 20/07/18 13:20, Ben Kohler wrote:
> On 07/19/18 20:54, Mikle Kolyada wrote:
>
>> +1. widely used profiles should have as least flags enabled by default
>> as possible, I would not be happy with +udev on my servers.
>>
> I disagree with this premise. The default and most widely used profiles
> s
On Fri, Jul 20, 2018 at 9:05 AM wrote:
>
> I’m not sure I was clear enough in what 13.0 would mean : basically, its
> current content would be
> delegated to common, and 13.0 would keep only things needed to have minimal
> breakages/conflicts.
> And we would keep the current directory-like inher
On 20/07/18 13:39, n...@troglodyte.be wrote:
> Hi,
>
> July 20, 2018 2:26 PM, "Ben Kohler" wrote:
>
>> On 07/19/18 23:04, Michael Orlitzky wrote:
>>
>>>
>>>
>>> If you really want to enable it globally after being told that it's bad
>>> engineering and downright annoying, go do it in a profile th
On Fri, Jul 20, 2018 at 9:17 AM M. J. Everitt wrote:
>
> The hierarchy method is indeed flawed, it would be better to have
> something akin to USE flags for profiles (PROFLAGS?) .. so that you
> could mingle different aspects without replicating sections of the
> 'tree' to get the common configura
On 07/20/2018 07:55 AM, Rich Freeman wrote:
>
> While I agree that setting USE=-udev isn't the same as leaving it to
> package defaults, you further claim that setting this globally causes
> severe breakage in some cases. Can you provide an example of this?
>
https://bugs.gentoo.org/640226
Or
On Fri, Jul 20, 2018 at 9:47 AM Michael Orlitzky wrote:
>
> On 07/20/2018 07:55 AM, Rich Freeman wrote:
> >
> > While I agree that setting USE=-udev isn't the same as leaving it to
> > package defaults, you further claim that setting this globally causes
> > severe breakage in some cases. Can you
Mart Raudsepp wrote:
> > * USE=udev means different things for different packages. You think
> > it "makes udev work" or whatever, but nobody has any idea what it
> > does for half of the packages that use it. The meaning is package-
> > specific, so the default should be package-specific.
>
On 07/20/2018 03:37 AM, Matt Turner wrote:
>>
>> If I want to undo your new flag, I have to set USE="-udev" globally, and
>> that clobbers any important per-package defaults that maintainers have set.
>
> I understand the concern at least in theory. But can you please give
> me a concrete example
On 2018.07.20 14:14, Rich Freeman wrote:
> While you can get Gentoo
> running with busybox and such and I completely support having profiles
> to enable this, I'm not sure this is the sort of thing that we want to
> point new users towards as a starting point.
>
> --
> Rich
>
New to Linux user
On Thu, Jul 19, 2018 at 11:01 AM, Michał Górny wrote:
> Python 3.5+ introduces parallel build support in distutils. Take
> advantage of that by passing appropriate -j option. Since distutils
> does not support an equivalent of --load-average, default to the number
> of CPUs+1 when unspecified.
>
Hi, everyone.
I'm working on improving OpenPGP commit verification on Infra end.
The changes so far shouldn't visibly affect developers whose accounts
are configured correctly. However, if you have trouble pushing, please
contact me (or other Infra members) immediately to investigate.
The change
22 matches
Mail list logo
