Quarter of the Hurd 2011-q4

[Arne Babenhauserheide]

Hi,

I finished the draft for the qoth-2011-q4. Please tell me what’s missing - and 
what needs fixing!

→ http://www.bddebian.com/~hurd-web/contributing/web_pages/news/moth_next/

And remember: It’s a wiki, so feel free to hack it to make it better!

------

A quarter of the Hurd: *Nix-based builds*, *slab allocator*, and *life 
filesystem debugging*.

This quarter Ludovic Courtès added a [Continuously-built Nix-based QEMU 
image](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00047.html), 
raising the count of Hurd distributions to 3: Debian, Arch and now Nix. His 
build is still pretty basic, but continuous integration is a great feature for 
testing images.

Also he added a [modern Autoconf 
initialization](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00068.html),
 [allowed /hurd/init to be a 
symlink](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00032.html) and 
made the Hurd [build with Savannah’s libc 
(2.14+)](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00025.html), 
getting the Nix distribution even more towards the cutting edge.

Samuel Thibault followed up with a [new debian disk 
set](http://lists.debian.org/debian-hurd/2011/12/msg00095.html) as christmas 
gift, turned /dev/urandom into a [native 
translator](http://lists.debian.org/debian-hurd/2011/11/msg00092.html) and 
[identified](http://lists.debian.org/debian-hurd/2011/11/msg00095.html) three 
easy porting cases with solutions: 

- undefined reference to dl_foo: add -ldl for building
- undefined reference to `main': missing gnu* case in the linking part of 
configure.ac or .in (pd-* packages are already being handled by their 
maintainer)
- undefined reference to clock_gettime or crypt: add -lrt or -lcrypt

These should help all those who want to help [[porting_packages|hurd/porting/]].

Maksym Planeta and Richard Braun committed [the last 
patches](http://lists.gnu.org/archive/html/bug-hurd/2011-12/msg00046.html) for 
integration of the slab allocator, making the memory allocator of GNU Mach SMP 
ready again. From IRC, freenode, #hurd, 2011-11-14:

    <braunr> there shouldn't be any noticeable difference with the
      master branch  
    <braunr> a bit less fragmentation  
    <braunr> more memory can be reclaimed by the VM system  
    <braunr> there are debugging features  
    <braunr> it's SMP ready  
    <braunr> and overall cleaner than the zone allocator  
    <braunr> although a bit slower on the free path (because of
      what's performed to reduce fragmentation)  
    <braunr> but even "slower" here is completely negligible

Also Richard Braun added Exodar (exodar.debian.net), a new porter box. It is 
faster and more reliable, which facilitates testing and bug hunting.

Additionally [[documentation_of_the_I/O_Path|hurd/io_path/]] was improved and 
Sergio Lopez documented his work on 
[[Better_Memory_Management_and_memfs|user/Sergio_Lopez/]] to make it easier for 
other hackers to work on the Hurd. 

Our hackers also used the quarter for porting a good number of packages. Sergio 
Lopez took care of 
[webkitgtk+](http://lists.debian.org/debian-hurd/2011/10/msg00025.html), while 
Svante Signell from [Arch Hurd](http://archhurd.org) went on a porting rampage 
with [pax](http://lists.debian.org/debian-hurd/2011/10/msg00105.html), 
[abiword](http://lists.debian.org/debian-hurd/2011/11/msg00035.html), 
[syslog-ng](http://lists.debian.org/debian-hurd/2011/11/msg00060.html), 
[ecl](http://lists.debian.org/debian-hurd/2011/11/msg00058.html), 
[fakeroot](http://lists.debian.org/debian-hurd/2011/12/msg00022.html), 
[daemon](http://lists.debian.org/debian-hurd/2011/12/msg00025.html) and 
[procps](http://lists.debian.org/debian-hurd/2011/12/msg00046.html) and Samuel 
Thibault from [Debian GNU/Hurd](http://www.debian.org/ports/hurd/) added 
[packagekit](http://lists.debian.org/debian-hurd/2011/10/msg00071.html), 
[evolution](http://lists.debian.org/debian-hurd/2011/10/msg00070.html), 
[emacs23 ](http://lists.debian.org/debian-hurd/2011/12/msg00018.html), 
[gcc-4.7](http://lists.debian.org/debian-hurd/2011/12/msg00065.html) and 
[iceweasel 
(firefox)](http://lists.debian.org/debian-hurd/2011/12/msg00080.html). Also 
Bouju Alain 
[submitted](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00079.html) 
[patches](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00084.html) to 
support cpuinfo in the /proc interface.

Additionally many bugs were found and squashed, by Pino Toscano ([recvfrom() 
with null](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00161.html)), 
Maksym Planeta 
([tmpfs](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00118.html)), 
Samuel Thibault 
([libtool](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00073.html), 
[mknod](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00070.html), [Fix 
POSIX 2008 
visibility](http://lists.gnu.org/archive/html/bug-hurd/2011-12/msg00004.html), 
[sudo setresuid](http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=645285)), Jim 
Meyering ([gnu tools with user id 
4294967295](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00094.html)), 
Paul Eggert ([Add error-checking on 
GNU](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00130.html)), Pino 
Toscano and Thomas Schwinge ([improved 
pthread](http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00013.html)) and 
Svante Signell ([e2fsprogs 
quota](http://lists.debian.org/debian-hurd/2011/10/msg00015.html)) to list only 
the ones who posted their fixes on the mailing list. 

Also Guillem Jover [fixed Mach's int vs long 
discrepancy](http://lists.debian.org/debian-hurd/2011/10/msg00053.html), which 
takes us the first step towards porting the Hurd to x64.

And there are now social network sites for GNU Hurd on 
[Google+](https://plus.google.com/114942488385711891227#114942488385711891227/posts)
 and the AGPL licensed [identi.ca](http://identi.ca/group/hurd), to make it 
easier to stay in touch.

As a final note, we want to share a story about real-life debugging with the 
Hurd:

    <youpi> yay GNU/Hurd  
    <youpi> I have added i_translator check in e2fsck, it was missing  
    <youpi> I had a volume that was keeping making ext2fs crash  
    <youpi> with a reproductible scenario  
    <youpi> could easily work out it was i_translator, then add a
      check to e2fsck, run it, which indeed fixed, them, and voilà,
      ext2fs was working again
        <youpi> all that on the same machine with *no* system reboot  
    <youpi> just ext2fs restart :)

So if you want to join us in our journey to realize more of the promises of the 
architecture of the Hurd, please [[get_in_contact|contact_us]] -- and maybe 
already grab the [[source_code|source_repositories]] and have fun hacking on 
free software!

------ 

Best wishes,
Arne