Sounds like a pretty cool software. Will take a look later. On Sep 22, 2015 16:42, "christopher@ira-kunststoffe.de" < christopher@ira-kunststoffe.de> wrote:

Hello,

Im the developer of rmlint (http://rmlint.rtfd.org) and mantain also the rmlint-git AUR package (https://aur4.archlinux.org/packages/rmlint-git/). I'd love to see it also in the official community repository, especially since some users don't like to install the build dependencies (scons, sphinx). For that, of course, a TU would be needed and I figured the best place to find one would be here. I think the number of votes should be enough.

If changes in the PKGBUILD are required, I will fix that beforehand of course. There is also another package by someone else that is not build from git, but from release tarballs: https://aur.archlinux.org/packages/rmlint/

Thanks, Chris

P.S: Sorry if that got posted twice.

Am Di, 22. Sep, 2015 um 5:29 schrieb Eli Schwartz :

On 09/22/2015 10:42 AM, christopher@ira-kunststoffe.de wrote:

...

Hello,

Im the developer of rmlint (http://rmlint.rtfd.org) and mantain also the rmlint-git AUR package (https://aur4.archlinux.org/packages/rmlint-git/).

Slightly offtopic, but have you considered cleaning up the pkgver() to use the release tag, as recommended here: https://wiki.archlinux.org/index.php/VCS_package_guidelines#Git

Thanks for the hint. Changed it to produce pkgvers like `2.2.0.r509.gc0b8e23`.

Also, you are missing the conflicts field (the non-git version has the conflict, instead, but I believe the general usage is to specify that in the development version).

Can't hurt it to have it in the -git version too. I've updated it.

No, sorry, I'm quite busy. I will try to get to this tomorrow. On Sat, Sep 26, 2015 at 4:46 PM, christopher@ira-kunststoffe.de < christopher@ira-kunststoffe.de> wrote:

Hi,

Sorry if this sounds impatient: Are there any news on this?

Am Di, 22. Sep, 2015 um 5:16 schrieb Sven-Hendrik Haase

...

:

...

Sounds like a pretty cool software. Will take a look later. On Sep 22, 2015 16:42, "christopher@ira-kunststoffe.de" < christopher@ira-kunststoffe.de> wrote:

Hello,

...

Im the developer of rmlint (http://rmlint.rtfd.org) and mantain also the rmlint-git AUR package (https://aur4.archlinux.org/packages/rmlint-git/ ). I'd love to see it also in the official community repository, especially since some users don't like to install the build dependencies (scons, sphinx). For that, of course, a TU would be needed and I figured the best place to find one would be here. I think the number of votes should be enough.

If changes in the PKGBUILD are required, I will fix that beforehand of course. There is also another package by someone else that is not build from git, but from release tarballs: https://aur.archlinux.org/packages/rmlint/

Thanks, Chris

P.S: Sorry if that got posted twice.

Hello Massimiliano, To 1) Indeed only `libelf` might be needed for detecting non-stripped binaries. It is an optional dependency and if it's not installed, it won't be build with it. I was not aware that libelf is packaged as separate package. Here is the relevant code btw: https://github.com/sahib/rmlint/blob/master/lib/utilities.c#L182 To 2) Indeed, gettext can be likely moved to build deps then. To 3) The gui can be found in the develop branch, it's not yet in master: https://github.com/sahib/rmlint/tree/develop/gui It will be included in the next release which will be probably somewhen this month. More information here: http://rmlint.readthedocs.org/en/latest/gui.html Regards, Christopher Am Di, 13. Okt, 2015 um 10:43 schrieb Massimiliano Torromeo :

Hi, I'm looking at rmlint right now and I have cleaned-up the dependencies a bit. Since you are the developer, it would be nice to have your confirmation on a couple of things:

1. I cound't find any reason why rmlint should depend on elfutils. My guess is that you need the "strip" command but that is provided by binutils, while elfutils has "eu-strip". 2. It seems to me that gettext is a build time dependency. The binary is not linked to it and I don't see calls to its tools outside the build scripts. 3. What are the optdepends for? I don't see any GUI in the source tree.

Yay, thanks a lot Massimiliano! I updated my PKGBUILD in the AUR too with the points you mentioned. If I release a new version, would it be a good idea to post a short notice here? Alternatively, an interested maintainer could register this feed to get notifed: https://github.com/sahib/rmlint/releases.atom Am Di, 13. Okt, 2015 um 6:04 schrieb Massimiliano Torromeo :

Congratulations, I just moved rmlint to [community] [1] :)

[1] https://www.archlinux.org/packages/community/x86_64/rmlint/

I just noticed that rmlint crashes on my devices with an "Illegal hardware instruction". This is most likely due to the fact that our build system tries to be clever about enabling sse4.2 support for CityHash. If it detects that your machine works with SSE4.2 (Intel CPU I guess?) it will be enabled, which causes the above error on CPUs that do not support it. I've disabled this cleverness and added `--with-sse` for people that built rmlint themselves: https://github.com/sahib/rmlint/commit/d2b44bee176966f93fc48465afcf942a5ee9e... This commit is included in the 2.2.1 release which I just pushed. Am Di, 13. Okt, 2015 um 11:51 schrieb Massimiliano Torromeo :

Il giorno mar 13 ott 2015 alle ore 19:18 christopher@ira-kunststoffe.de < christopher@ira-kunststoffe.de> ha scritto:

...

Yay, thanks a lot Massimiliano!

No problem!

...

If I release a new version, would it be a good idea to post a short notice here? Alternatively, an interested maintainer could register this feed to get notifed:

https://github.com/sahib/rmlint/releases.atom

No need to notify the ML. In fact I did already subscribe to the atom feed. https://www.archlinux.org/packages/community/x86_64/rmlint/

Il giorno mer 14 ott 2015 alle ore 01:08 christopher@ira-kunststoffe.de < christopher@ira-kunststoffe.de> ha scritto:

I've disabled this cleverness and added `--with-sse` for people that built rmlint themselves:

That's better. I am updating it now. Il giorno mer 14 ott 2015 alle ore 02:02 Levente Polyak < anthraxx@archlinux.org> ha scritto:

Looks orphan to me, any chance you forgot to adopt it after the upload? :P

Yep, it happens all the times :P

Hi Christopher, On 14/10/15, christopher@ira-kunststoffe.de wrote:

Sorry for the inconvinience, it still crashes with "illegal instruction". This is most likely due to the fact that I didn't actually make --without-sse the default. Should be (hopefully!) corrected in v2.2.2. I shouldn't push changes late night. :)

Just tried it on AMD FX 8210 from the package and had the same Illegal hardware and code dump. I dowload the PKGBUILD of the package and bumped the version to 2.2.2. Compiled and worked. Can't you make the code use the hardware SSE 4.2 CRC32C if the hardware is there or use a software implementation of it if not? What's the difference in performance with and without SSE4.2 enabled - can we have some metrics? -- Leonidas Spyropoulos Compiled in Vim, sent through mutt.