UML not maintained anymore?

UML not maintained anymore?

[Natale Patriciello]

It seems there is no interest in fixing bugs (such as [1]). Moreover,
same guest filesystem (same host os, distribution, etc.) on two
different machines (i7-2630 the first, i7-7700HQ the second) yield
different results, with crashes and corruption of filesystem in the
modern computer.  So, can I consider UML as a legacy thing in the Linux
kernel? With what I can replace it (I'm doing TCP research, and I focus
on the networking stack)?

Thank you

[1] https://sourceforge.net/p/user-mode-linux/mailman/message/35663374/ 

Re: UML not maintained anymore?

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 852 bytes --]

On Tue 2017-02-28 16:28:11, Natale Patriciello wrote:
> It seems there is no interest in fixing bugs (such as [1]). Moreover,
> same guest filesystem (same host os, distribution, etc.) on two
> different machines (i7-2630 the first, i7-7700HQ the second) yield
> different results, with crashes and corruption of filesystem in the
> modern computer.  So, can I consider UML as a legacy thing in the Linux
> kernel? With what I can replace it (I'm doing TCP research, and I focus
> on the networking stack)?

Well.. if it worked before and does not work now, that's a regression
and will be fixed. Bisect would be useful.

If it never worked on new CPUs, that's different situation...
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: UML not maintained anymore?

[Richard Weinberger]

On Fri, Mar 17, 2017 at 4:46 PM, Pavel Machek <pavel@ucw.cz> wrote:
> On Tue 2017-02-28 16:28:11, Natale Patriciello wrote:
>> It seems there is no interest in fixing bugs (such as [1]). Moreover,
>> same guest filesystem (same host os, distribution, etc.) on two
>> different machines (i7-2630 the first, i7-7700HQ the second) yield
>> different results, with crashes and corruption of filesystem in the
>> modern computer.  So, can I consider UML as a legacy thing in the Linux
>> kernel? With what I can replace it (I'm doing TCP research, and I focus
>> on the networking stack)?
>
> Well.. if it worked before and does not work now, that's a regression
> and will be fixed. Bisect would be useful.
>
> If it never worked on new CPUs, that's different situation...

Yep, a bisect would be good.
Regressions are rather easy to fix.

UML is more or less in legacy mode an I work on it purely in my very limited
spare time. That's why I don't have time to track down and solve every
single bug.

-- 
Thanks,
//richard