Message-ID: <8d6b1fcc-1a21-1707-dd8e-43529e1d644c@landley.net>
Date:   Thu, 1 Feb 2018 17:34:41 -0600
From:   Rob Landley <rob@...dley.net>
To:     Taras Kondratiuk <takondra@...co.com>,
        Arvind Sankar <nivedita@...m.mit.edu>,
        Mimi Zohar <zohar@...ux.vnet.ibm.com>
Cc:     initramfs <initramfs@...r.kernel.org>,
        Victor Kamensky <kamensky@...co.com>,
        linux-security-module <linux-security-module@...r.kernel.org>,
        Al Viro <viro@...IV.linux.org.uk>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH] rootfs: force mounting rootfs as tmpfs



On 02/01/2018 04:41 PM, Taras Kondratiuk wrote:
> Quoting Mimi Zohar (2018-02-01 13:51:52)
>> On Thu, 2018-02-01 at 11:09 -0600, Rob Landley wrote:
>>> On 02/01/2018 09:55 AM, Mimi Zohar wrote:
>>>> On Thu, 2018-02-01 at 09:20 -0600, Rob Landley wrote:
>>>>
>>>>>> With your patch and specifying "root=tmpfs", dracut is complaining:
>>>>>>
>>>>>> dracut: FATAL: Don't know how to handle 'root=tmpfs'
>>>>>> dracut: refusing to continue
>>>>>
>>>>> [googles]... I do not understand why this package exists.
>>>>>
>>>>> If you're switching to another root filesystem, using a tool that
>>>>> wikipedia[citation needed] says has no purpose but to switch to another
>>>>> root filesystem, (so let's reproduce the kernel infrastructure in
>>>>> userspace while leaving it the kernel too)... why do you need initramfs
>>>>> to be tmpfs? You're using it for half a second, then discarding it,
>>>>> what's the point of it being tmpfs?
>>>>
>>>> Unlike the kernel image which is signed by the distros, the initramfs
>>>> doesn't come signed, because it is built on the target system.  Even
>>>> if the initramfs did come signed, it is beneficial to measure and
>>>> appraise the individual files in the initramfs.
>>>
>>> You can still shoot yourself in the foot with tmpfs. People mount a /run
>>> and a /tmp and then as a normal user you can go
>>> https://twitter.com/landley/status/959103235305951233 and maybe the
>>> default should be a little more clever there...
>>>
>>> I'll throw it on the todo heap. :)
>>>
>>>>> Sigh. If people are ok with having rootfs just be tmpfs whenever tmpfs
>>>>> is configured in, even when you're then going to overmount it with
>>>>> something else like you're doing, let's just _remove_ the test. If it
>>>>> can be tmpfs, have it be tmpfs.
>>>>
>>>> Very much appreciated!
>>>
>>> Not yet tested, but something like the attached? (Sorry for the
>>> half-finished doc changes in there, I'm at work and have a 5 minute
>>> break. I can test properly this evening if you don't get to it...)
>>
>> Yes, rootfs is being mounted as tmpfs.
> 
> I don't think you can unconditionally replace ramfs with initramfs by
> default. Their behavior is different in some cases (e.g. pivot_root vs
> switch_root)

Both are switch_root, you can't pivot_root off of either one. (Yes, I
hit that bug and reported it, and they fixed it, back in the day...
http://lists.busybox.net/pipermail/busybox/2006-March/053529.html )

> and it can break many systems that expect ramfs by default.

The use case I told Mimi about off-list (since they stopped cc:ing the
list in one of their replies but the conversation continued) was the guy
who was extracting an initramfs bigger than 50% of system memory, which
worked with initramfs but failed with initmpfs. A quick google didn't
find the original message but it resulted in this blog entry from the
affected party:

http://www.lightofdawn.org/blog/?viewDetailed=00128

I.E. yeah, I know, I need to redo these patches tonight.

Rob

Powered by blists - more mailing lists