linux-kernel - Re: [PATCH] modules: CONFIG_MODULE_COMPRESS: add hint that userspace support may easily be missing.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20150604200927.GB15890@rhlx01.hs-esslingen.de>
Date:	Thu, 4 Jun 2015 22:09:27 +0200
From:	Andreas Mohr <andi@...as.de>
To:	Stephen Rothwell <sfr@...b.auug.org.au>
Cc:	Rusty Russell <rusty@...tcorp.com.au>,
	Lucas De Marchi <lucas.de.marchi@...il.com>,
	Andreas Mohr <andi@...as.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Bertrand Jacquin <beber@...eeweb.net>,
	Marco d'Itri <md@...ux.it>,
	linux-modules <linux-modules@...r.kernel.org>,
	lkml <linux-kernel@...r.kernel.org>, Jon Masters <jcm@...hat.com>
Subject: Re: [PATCH] modules: CONFIG_MODULE_COMPRESS: add hint that userspace
 support may easily be missing.

Hi,

On Thu, Jun 04, 2015 at 01:19:46PM +1000, Stephen Rothwell wrote:
> Hi Rusty,
> > But disappointing that Debian doesn't configure with it, and there's no
> > easy way to check it.  Looks like Ubuntu vivid is the same.
> 
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=772628

Crap, that URL now is an implicit strong suggestion
that I didn't do my homework ;)
(it already described large parts
prior to me regurgitating this issue
due to being unaware of this history here).


Thanks to R.Russell for a very nice shortening
and extension of the help text!


Since the issue of .gz vs. .xz redundancy came up
(with people "threatening" to support only one alternative),
I want to mention that
when having to choose one
I'd tend to activating the .xz library dependency:
- while it has higher compression demands,
  hardware is getting beefier all the time,
  thus it should not matter
  (especially vs. the dominantly many decompression runs)
- it's simply the "more modern" and future-proof option,
  thus it should be favoured slightly
  since the system as a whole
  would want to make reasonably quick
  development/evolution "forward progress"
  rather than sticking to less favourable mechanisms
- .xz has been available for some time already
  i.e. the time window of "distro support maturity" is a given
[a counter-point might be that module-init-tools supports .gz only,
but then modern binary setups which chose .xz
would already have been shipped with kmod only]

OK, having said that,
I'm unsure what to think of the Debian package's decision
to not support compression so far
(and that even in times
where kmod does not provide a runtime option yet
to query the actual set of support flags).
>From a library dependency POV it may be attractive to skip compression,
and since Debian usually has a fixed setup
where non-compressed files are a given,
this seems like a valid choice.
It's just that for e.g. these situations:
- people with many custom kernels installed
- space-constrained systems
this is quite a nuisance that is a wee bit too unexpected (/show-stopper) -
took me roughly > 30 minutes to get it researched / resolved.

Thanks,

Andreas
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/