linux-kernel - Re: [PATCH 1/3 V2] time: move timeconst.h into include/generated

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]

Date:	Sun, 12 Apr 2015 16:35:39 -0500
From:	Rob Landley <rob@...dley.net>
To:	Nicholas Mc Guire <hofrat@...dl.org>
Cc:	Michal Marek <mmarek@...e.cz>,
	Masahiro Yamada <yamada.m@...panasonic.com>,
	Sam Ravnborg <sam@...nborg.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	"H. Peter Alvin" <hpa@...or.com>, Joe Perches <joe@...ches.com>,
	John Stultz <john.stultz@...aro.org>,
	Andrew Hunter <ahh@...gle.com>, Paul Turner <pjt@...gle.com>,
	Aaron Sierra <asierra@...-inc.com>,
	Brian Norris <computersforpeace@...il.com>,
	Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/3 V2] time: move timeconst.h into include/generated

On Sun, Apr 12, 2015 at 7:13 AM, Nicholas Mc Guire <hofrat@...dl.org> wrote:
> kernel/time/timeconst.h is moved to include/generated/ and generated in
> an early build stage by top level Kbuild. This allows using timeconst.h
> in an earlier stage of the build.
>
> Signed-off-by: Nicholas Mc Guire <hofrat@...dl.org>

Sigh, I'll adjust my patch:

http://landley.net/hg/aboriginal/file/1698/sources/patches/linux-noperl-timeconst.patch

Backstory: Peter Anvin added perl to the kernel build in 2.6.25 and
something like the 9th time I submitted patches to remove it, several
years later when it looked like they'd finally go in, he submitted a
competing patch to one of my "just do this in C and shell" patch
series to instead add a dependency on the 'bc" tool, which is not in
busybox, wasn't in the linux from scratch or buildroot or openembedded
builds (everybody had to add it after his patch went in to keep
building the kernel), and which is actually hard to implement in a
posix compliant way for an embedded environment because it's defined
as requiring arbitrary precision math (all timeconst calculation needs
is 64 bit math, I.E. long long) and it's defined as being capable of
doing things like fractional exponentiation at arbitrary precision
(fixed point, I think).

I don't know why Peter is on a crusade to stamp out simple build
environments. He added perl as a build dependency to every tool he
maintained at the same time (not just the kernel but also klibc and
his bootloader), and then when he couldn't defend perl he added a new
dependency to break existing simple/audited build environments.

When I objected to him about the perl he said I was engaged in an
uninteresting "academic" exercise.
(https://lkml.org/lkml/2008/2/15/548) Oh yes, a simple auditable build
environment totally has no real world consequences:

http://www.kith.org/journals/jed/2015/03/15/15043.html

Yup, none at all. Snowden proved that, clearly. (And he was totally a
one-off, it's not like there was Manning or Daniel Ellsberg or Mark
Felt before him...)

Sigh. Don't mind me, I'll update my local patch, once again.

Rob
--
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/