[Mageia-dev] How to Test the new uClibc?

[Colin Guthrie]

'Twas brillig, and Thierry Vignaud at 09/08/12 08:10 did gyre and gimble:
> On 9 August 2012 08:31, Shlomi Fish <shlomif at iglu.org.il> wrote:
>>> wrote:
>>>>> since the new busybox (which I maintain in Mageia) package did not
>>>>> build with the uClibc version that was present in Cauldron, I grab
>>>>> maintainership for uClibc too, and commited a new uClibc package
>>>>> to the svn (based a lot on the Mandriva Cooker package).
>>>>>
>>>>> However, now I'm wondering how to best test it to make sure
>>>>> nothing got broken. Which programs are using it on a Mageia system
>>>>> that I can use to test?
>>>>
>>>> Not that much.
>>>> If busybox tools still work, just go on...
>>>
>>> Ping! It's time to make cauldron hot!
>>
>> I don't know if you attended that IRC conversation, so let me quote it (I am rindolf):
> 
> I didn't.
> 
> (...)
> 
>> So I'm back at square 1, and don't know how to further proceed. If someone more capable than me can take
>> it upon themselves, then it would be great.
> 
> I think's Neoclust worries for nothing.
> 
> There not that much packages using uclibc:
> 
> busybox-1.20.2-1.mga3.src
> CCfits-2.4-1.mga3.src
> liblzo-2.06-1.mga2.src
> libx86-1.1-11.mga2.src
> lvm2-2.02.95-5.mga3.src
> newt-0.52.14-2.mga3.src
> pciutils-3.1.10-1.mga3.src
> slang-2.2.4-4.mga3.src
> sysfsutils-2.1.0-14.1.mga3.src
> util-linux-2.21.2-3.mga3.src
> wmbattery-2.40-1.mga2.src
> wmfire-1.2.4-1.mga2.src
> zlib-1.2.7-4.mga3.src
> 
> Note that of those are unused
> (busybox/pciutils/newt/slang/sysfsutils/zlib) unlike
> mdv who choose to bloat installer by compiling with uclibc instead of dietlibc
> 
> What's more, the remaining ones (util-linux, lvm2 (maybe busybox?)) were
> used with old mkinitrd but I think dracut now use glibc linked ones.
> Colin, can you confirm?

Dracut just uses what is on the system, so yeah it'll use glibc unless
some module pulls it in. There is a busybox module in dracut to allow it
to be installed but it's not enabled by default AFAICT.

> So I don't think it's that much critical.
> At worst, we can check that regenerating initrd on a LVM machine still
> boots.

My machine is an LVM one, so I can likely test this, but as my current
initrd does not have ulibc in it, I doubt very much this test will show
anything other than success :)

Cheers

Col



-- 

Colin Guthrie
colin(at)mageia.org
http://colin.guthr.ie/

Day Job:
  Tribalogic Limited http://www.tribalogic.net/
Open Source:
  Mageia Contributor http://www.mageia.org/
  PulseAudio Hacker http://www.pulseaudio.org/
  Trac Hacker http://trac.edgewall.org/