Re: install/57100: NetBSD sysint seems to be riddled with bugs and memory corruption

[Sad Clouds <cryintothebluesky%gmail.com@localhost>]

The following reply was made to PR install/57100; it has been noted by GNATS.

From: Sad Clouds <cryintothebluesky%gmail.com@localhost>
To: Martin Husemann <martin%duskware.de@localhost>
Cc: gnats-bugs%netbsd.org@localhost
Subject: Re: install/57100: NetBSD sysint seems to be riddled with bugs and
 memory corruption
Date: Sat, 10 Dec 2022 18:13:47 +0000

 On Sat, 10 Dec 2022 18:40:52 +0100
 Martin Husemann <martin%duskware.de@localhost> wrote:
 
 > Why would you care about start and end sectors, or file system types?
 > This is the "broad overview" and you only select how much of your disk
 > you offer to NetBSD - or something like that. On the second stage you
 > typically just accept whatever the first stage created ;-)
 
 Because it is a strange workflow. My understanding is that GPT is
 different from the old disklabel, where you had one NetBSD partition,
 which was subdivided into a, b, c, d, etc sub-partitions. So with GPT
 people can have different layouts.
 
 You need to know start/end sectors, so that you can check partition
 alignment. So with 4k drives you could have misaligned partitions and
 end up with poor performance.
 
 Essentially, it would be good to have "advanced partition" workflow,
 where you can manually create, delete, modify whichever partition
 layouts you want and not have sysint get in the way.