For discussion of Alpine Linux development and developer support

3 2

[alpine-devel] Re: [alpine-aports] [PATCH] main/mkinitfs: backport cryptdiscards patch

Natanael Copa
Details
Message ID
<20170616181142.28b8a187@ncopa-desktop.copa.dup.pw>
Sender timestamp
1497629502
DKIM signature
missing
Download raw message
On Wed, 14 Jun 2017 23:44:34 +0200
Sören Tempel <soeren@soeren-tempel.net> wrote:

> On 14.06.17, Natanael Copa wrote:
> > What do you think? Should we go for:
> > 
> >   cryptroot=$device cryptdm=$dmname cyptdiscards=yes cryptheader=$header cryptoffset=$offset
> > 
> > or should we do:
> > 
> >   cryptdevice=$device:$dmname:allow-discards cryptkey=$header:$offset  
> 
> Honestly I don't have a strong opinion on this. I believe the latter is
> harder to read and besides we would break backwards compatibility if we
> would switch to this scheme.
> 
> Do you have an opinion on this? Would you like to switch to the latter?

Does anybody on alpine-devel have opinion on what the boot args should be for LUKS devices?

-nc



---
Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
Help:         alpine-devel+help@lists.alpinelinux.org
---
Sören Tempel
Details
Message ID
<20170622154836.GA3263@francium>
In-Reply-To
<20170616181142.28b8a187@ncopa-desktop.copa.dup.pw> (view parent)
Sender timestamp
1498146516
DKIM signature
missing
Download raw message
On 16.06.17, Natanael Copa wrote:
> Does anybody on alpine-devel have opinion on what the boot args should
> be for LUKS devices?

Doesn't seem to be the case. How about merging this into the mkinitfs
repository without backporting it for now? We can still change the name
of the kernel parameters later on.

Cheers,
Sören


---
Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
Help:         alpine-devel+help@lists.alpinelinux.org
---
Sören Tempel
Details
Message ID
<20170714145955.GA17438@francium>
In-Reply-To
<20170622154836.GA3263@francium> (view parent)
Sender timestamp
1500044395
DKIM signature
missing
Download raw message
Ping?

On 22.06.17, Sören Tempel wrote:
> On 16.06.17, Natanael Copa wrote:
> > Does anybody on alpine-devel have opinion on what the boot args should
> > be for LUKS devices?
> 
> Doesn't seem to be the case. How about merging this into the mkinitfs
> repository without backporting it for now? We can still change the name
> of the kernel parameters later on.
> 
> Cheers,
> Sören
> 
> 
> ---
> Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
> Help:         alpine-devel+help@lists.alpinelinux.org
> ---
> 


---
Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
Help:         alpine-devel+help@lists.alpinelinux.org
---
7heo
Details
Message ID
<1865c521-d1f1-3e7e-eb3f-e32898a3e7c2@mail.com>
In-Reply-To
<20170714145955.GA17438@francium> (view parent)
Sender timestamp
1500281200
DKIM signature
missing
Download raw message
On 7/14/2017 4:59 PM, Sören Tempel wrote:
> Ping?
>
> On 22.06.17, Sören Tempel wrote:
>> On 16.06.17, Natanael Copa wrote:
>>> Does anybody on alpine-devel have opinion on what the boot args should
>>> be for LUKS devices?
>>
>> Doesn't seem to be the case. How about merging this into the mkinitfs
>> repository without backporting it for now? We can still change the name
>> of the kernel parameters later on.
>>
>> Cheers,
>> Sören

Hey there,

Sorry for the lag, I couldn't answer in Friday and then the week end got 
in the way.

However, I prefer the separated-variables way, and by far. Here is the 
reasoning:

- The parsing would happen in shell (as opposed as in the initfs binary) 
and I don't think doing unnecessary (for appearance only) text parsing 
in shell is the sane way to go.

- Using ':' (or anything else really) as separator is assuming that this 
character isn't used in any identifier in the list. In the past, present 
and future. For any of those identifiers, ever. And really, we cannot 
assume that.

- Separating variables with their name is way more explicit and way less 
magic than establishing an arbitrary order and relying on the users to 
read the documentation on what a variable does, and in what order.

- Having multiple variables grouped where it is technically incorrect 
will be misleading and confusing, teaching newcomers the wrong lessons.

* * *

Long story short, I do not think that the cost (purely visual) of having 
the variables separated in the configuration file is worth the technical 
effort and technical risk of concatenating them. KISS.

HTH
7heo


---
Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
Help:         alpine-devel+help@lists.alpinelinux.org
---