gmane.linux.lfs.beyond.devel

BLFS Status

Bruce Dubbs — 2012-05-25T17:28:54

As of right now, we have

344 packages marked with lfs70_checked
274 packages marked with lfs71_checked
   0 packages marked with lfs6?_checked

There is one package not marked that I know of, icedtea, and that is 
being worked.

There are 15 open tickets.  Two are new apps, 11 are upgrades to 
existing packages (including icedtea).  Two tickets do not involve 
package upgrades.

   -- Bruce

ssd drives

Bruce Dubbs — 2012-05-23T23:00:14

I decided to explore using an ssd drive.  I purchased a 40G Intel SSD 
and so far it works well with SVN-20120514.

I created a GPT partition table with a 10G partition using parted and 
formatted as ext4.

The performance seems to be good.  htparm gives me about 230 GB/s which 
is more than twice as fast as my usual drives that give about 105 GB/s.

My question is how to best use the new drive in BLFS.  I thought of /opt 
and /usr.  I don't think /home would be very good and of course I could 
try to mount it as /mnt/lfs and use it as /.

What would you try first?

   -- Bruce

icedtea notes

Bruce Dubbs — 2012-05-23T04:26:52

I've been working on icedtea today.  The notes DJ used seem to build 
fine, but there are a lot of errors in the tests.  I'm going through the 
build a second time.  The first time I tested via ssh and a lot of the 
tests were just timing out so I aborted and started over.

The second time I was in an xfce session and set:

xset s off
xset dpms 0 0 0

to keep the screen blanker from kicking in.

What I've noticed is that the reason the tests take so long is that many 
time out.  If a test is going to pass, it seems to do it fairly quickly, 
but there are a lot of failures where the system is just sleeping.  The 
load factors run down to very small values.

I'm still going through the tests for a second time and these are the 
failures so far:

FAILED: com/sun/awt/SecurityWarning/GetSizeShouldNotReturnZero.java
FAILED: com/sun/awt/Translucency/WindowOpacity.java
FAILED: com/sun/jdi/connect/spi/DebugUsingCustomConnector.java
FAILED: com/sun/jdi/redefine/RedefineTest.java
FAILED: com/sun/jdi/redefineMethod/RedefineTest.java
FAILED: com/sun/jdi/sde/FilterMangleTest.java
FAILED: com/sun/jdi/sde/MangleStepTest.java
FAILED: com/sun/jdi/sde/MangleTest.java
FAILED: com/sun/jdi/sde/SourceDebugExtensionTest.java
FAILED: com/sun/jdi/sde/TemperatureTableTest.java
FAILED: com/sun/jdi/AccessSpecifierTest.java
FAILED: com/sun/jdi/AfterThreadDeathTest.java

Right now the load factor is down to 0.26.  A lot of the failures seem 
to have something to do with threads/events.

The second time through the build/test not going through ssh seems to 
have a lot fewer errors but I've probably still got a couple of hours to 
go.  I'll give a final result tomorrow.

------------

The failures do have a log,  For instance the FilterMangleTest.jtr is 
the 'Java Test Report'.  It tells me the following, but I don't know how 
to interpret it.

   -- Bruce

Execution failed: `main' threw exception: java.lang.NullPointerException
...
RROR: transport error 202: connect failed: Connection timed out
ERROR: JDWP Transport dt_socket failed to initialize, TRANSPORT_INIT(510)
JDWP exit error AGENT_ERROR_TRANSPORT_INIT(197): No transports 
initialized [../../../src/share/back/debugInit.c:741]
FATAL ERROR in native method: JDWP No transports initialized, 
jvmtiError=AGENT_ERROR_TRANSPORT_INIT(197)
...

  Target VM failed to initialize.
java.lang.NullPointerException
    at VMConnection.open(VMConnection.java:196)
    at TestScaffold.connect(TestScaffold.java:632)
    at TestScaffold.startUp(TestScaffold.java:363)
    at TestScaffold.startTo(TestScaffold.java:373)
    at TestScaffold.startToMain(TestScaffold.java:368)
    at FilterMangleTest.runTests(FilterMangleTest.java:160)
    at TestScaffold.startTests(TestScaffold.java:429)
    at FilterMangleTest.main(FilterMangleTest.java:96)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at 
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
    at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:601)
    at 
com.sun.javatest.regtest.MainWrapper$MainThread.run(MainWrapper.java:94)
    at java.lang.Thread.run(Thread.java:722)

nss

Bruce Dubbs — 2012-05-23T01:24:38

Looking at nss, the intro says:

* Estimated disk space required: 70 MB (more than double this to run the 
test suite)

* Estimated build time: 1.0 SBU (at least an additional 3.5 SBU to run 
the test suite)

But the instructions say:

   "This package does not come with a test suite."

I looked at the Makefile and could not any test or check.  Is the intro 
data about a test suite just an oversight from prior versions?

The size and md5sum were correct and the build size and time were pretty 
close to what I had.

   -- Bruce

colord check

Bruce Dubbs — 2012-05-21T21:13:34

I'm getting a failure in 'make check' in colord:

libcolord:ERROR:cd-self-test.c:971:colord_client_func: assertion failed 
(error == NULL): Could not connect: No such file or directory 
(g-io-error-quark, 1)
/bin/sh: line 5: 18614 Aborted                 ${dir}$tst
FAIL: cd-self-test

Has anyone else seen this?  The only dependency I haven't built is sane.

                     colord 0.1.19
                   ===================

         prefix:                    /usr
         datadir:                   ${datarootdir}
         compiler:                  gcc
         cflags:                    -g -O2
         cppflags:
         gobject-introspection:     yes
         PolicyKit support:         yes
         Reverse engineering tools: no
         File descriptor fallback:  yes
         External volume searching: yes
         SANE support:              no
         GUDEV support:             yes
         LCMS2 DICT support:
         Vala API generator:        no
         Daemon user:               root


   -- Bruce

New Trac Tickets

Armin K. — 2012-05-21T19:21:16

I was in process of fetching source packages for my new LFS build 
(Current LFS SVN + new packages reported in LFS Trac) and I stumbled 
upon those. I'll try to do most of them myself, but help is welcome if 
anyone wants to help. If you want to help, just assign it to yourself. 
I'll mark rest when I get to those. Currently running check on full gcc 
suite which will take a long time I think.

Holiday

Andrew Benton — 2012-05-20T00:55:23

Hello All,
I'll be away on holiday for a week so I won't be able to do any work
for a bit.

Andy

pan?

Lars Bamberger — 2012-05-19T10:44:34

Looking for pan ...

I found a commit dated 11 May 2012 from Bruce in which pan was removed
from the book. Pan-0.137 was released just 1 May 2012 and compiles and
runs fine (CMMI with an optional switch to use gtk+3)

Was there any particular reason why this package was deleted?

Lars
(please excuse if I'm not up to date on all the changes in the book, I
have been offline for the last couple of months.)

guile-2.0.5

Lars Bamberger — 2012-05-19T08:23:16

Hello,

trying to rebuild my productive system (gnucash ... ARGH!) .. it's a
long way.

Anyway: Current guile-2.0.5 is broken. I get this when I try to start guile:


This is a known error and fixed in guile commit
5de0053178b4acc793ae62838175e5f3ab56c603 and will be in guile-2.0.6.
(http://debbugs.gnu.org/cgi/bugreport.cgi?bug=11083) However,
guile-2.0.6 is not released yet. I've installed guile-2.0.5.155-20e2d (a
snapshot from guile's homepage:
http://hydra.nixos.org/job/gnu/guile-2-0/tarball/latest)

This version installs OK, make check runs without any failures (there
was one with guile-2.0.5).

slib-3b3 now installs OK as per blfs instructions. (One more small step
towards gnucash.)


The book should be updated to guile-2.0.6 as soon it's out. Meanwhile,
I'd put a note on the wiki, but I don't have write permissions.

Cheers,
Lars

Observations about firefox and xulrunner

Bruce Dubbs — 2012-05-18T23:41:27

I can build both Firefox and Xulrunner on my new SVN-20120514 system but 
had some issues that may or may not be significant.

I could not seem to get FF to build with the xulrunner linkage, although 
each built OK.

For FF, the command

tar -xvf firefox-build-dir/dist/firefox-12.0.en-US.linux-$(uname 
-m).tar.bz2 -C /usr/lib/firefox-12.0 --strip-components=1

hung on me.  I already had a /usr/lib/firefox-12.0 directory with items 
installed from previous builds, but deleting all the files there allowed 
the instruction ot complete.  We may want to do something like:

rm -rf /usr/lib/firefox-12.0 &&
mkdir /usr/lib/firefox-12.0 &&
tar -xvf \
    firefox-build-dir/dist/firefox-12.0.en-US.linux-$(uname -m).tar.bz2 \
   -C /usr/lib/firefox-12.0 --strip-components=1

-----------

I tried to build xulrunner and use it for firefox, but the instructions 
in the book are a bit confusing.

I built xulrunner by the current instructions (-j1):

SBU=25.628
77188 /usr/src/firefox/firefox-12.0.source.tar.bz2 size (75.378 MB)
985768 kilobytes build size (962.664 MB)
md5sum : 80c3e5927274de7f181fb5f931ac5fd4

But the files are in:

/usr/lib/xulrunner-12.0
/usr/lib/xulrunner-devel-12.0

The instructions on the firefox page says:

  If you have linked against an already installed Xulrunner, as the root 
user:

make -C firefox-build-dir install &&
...

My problem was that I didn't have a firefox-build-dir directory or a 
/usr/lib/firefox-12.0/application.ini.

I think we may need to clarify that make -f client.mk is still required 
but the options

# ac_add_options --with-system-libxul
# ac_add_options --with-libxul-sdk=/usr/lib/xulrunner-devel-12.0

will bypass most of the long build.

----------

I tried to run ff over ssh, but it died.  That is probably because my X 
server is fairly old -- well really old (6.8.2 from 2005).  :)

On the new system, FF started and ran fine.

---------

When I started FF, I looked at Help->About.  Since I was using twm, I 
couldn't close that window!  There is no close button for 'About' and 
twm doesn't have a window close function that I could find.  This isn't 
a BLFS issue, but I thought it was worthy of comment.

   -- Bruce

pulseaudio

Bruce Dubbs — 2012-05-18T18:47:35

When running make check for pulse audio, I get a hand.  It says 
'sleeping (5s)' but it never completes.  Has anyone else seen this.

   -- Bruce

libvorbis

Bruce Dubbs — 2012-05-18T03:24:38

For some reason, I have a problem with the libvorbis tests. 'make check' is not 
finding libm.so.

I fixed it with:

LDFLAGS=-lm ./configure --prefix=/usr  &&
make                                   &&
make check

Can anyone confirm this?

   -- Bruce

IcedTea-Web-1.2 with OpenJDK-1.7.0.3

Fernando de Oliveira — 2012-05-17T16:18:13

Just finished build and install.

New changed location:

ln -s ${JAVA_HOME}/jre/lib/IcedTeaPlugin.so /usr/lib/mozilla/plugins/

So, when upgrading the book to OpenJDK-1.7.0.3, it is necessary:

sed -i -e '#/<arch>##'
       -e '/, where <arch> is either i386 or amd64//'

Works well with FF and SM, but not Midori.

FF:

<quote>
Verifying Java Version
A newer version of Java is available

Please click the download button to get the recommended Java for your computer.
Your Java version: 1.7.0_03-icedtea

If you recently completed your Java software installation, you may need to restart your browser (close all browser windows and re-open) before verifying your installation.
Download Free Java Software
Version 7 Update 4
Download Java Now 
</quote>

SM: similar.

Midori: crash

I will try to rebuild both, M and WebKit, if nothing works, will ask in support.


[]s,
Fernando

freeglut

Bruce Dubbs — 2012-05-16T20:01:06

I'm having a problem with freeglut and xorg being in /opt/xorg.

I get:

/bin/sh ../../../libtool --tag=CC   --mode=link gcc -I../../../include
 -g -O2 -Wall -pedantic -export-dynamic  -o CallbackMaker
CallbackMaker-CallbackMaker.o ../../../src/libglut.la -lGL -lXext
-lX11

/usr/bin/ld: cannot find -lGL

Yet I have:

$ cat /etc/ld.so.conf
# Begin /etc/ld.so.conf
/usr/local/lib
/opt/lib
/opt/xorg/lib

and

$ ls /opt/xorg/lib/libGL*
/opt/xorg/lib/libGL.so    /opt/xorg/lib/libGL.so.1.2  /opt/xorg/lib/libGLU.so.1
/opt/xorg/lib/libGL.so.1  /opt/xorg/lib/libGLU.so
/opt/xorg/lib/libGLU.so.1.3.08000

I can work around this, but why do I need to?  Any ideas?

  -- Bruce

Recent build notes

DJ Lucas — 2012-05-15T22:44:14

Couple of things I noticed on my latest build.

The DRM backend is broken in cairo with Xorg-7.7, no idea if this is the 
result of the export symbols patch or latest xorg. Anybody built it? 
(probably unused)

Ghostscript lists lcms as a dependency, it should be lcms2.

Udev 182 didn't rebuild with pci-utils without a decompressed pci.ids file.

Because of build order (a very minimal target), the firefox.desktop file 
got installed as /usr/share/applications (need to check first, and 
create the directory if necessary).

Mesa

Bruce Dubbs — 2012-05-15T22:25:06

I've been going through a new install today.  I've run into a proglem with Mesa:

nouveau_vieux_dri.so.tmp: undefined reference to `nouveau_pushbuf_flush'

and about 15 similar references.  Where are these defined?  Either I
missed something or we will need to add how to get these referenced
defined.

  -- Bruce

Report on building GNOME

Pierre Labastie — 2012-05-15T21:35:49

Hi,

I have built GNOME according to the book instruction of May 13. It went 
rather smoothly (but lenghty). Here is my report.
Many of the observations are only typos, and often, they are about a non 
GNOME package.

Typos:
- libsoup page : in command explanations, first line : "specific 
depepdencies" instead of "dependencies"
- NSS page : in recommended dependencies : "internal sqlite is 
incompatable" instead of "incompatible" (well maybe incompatable is 
English and I do no know it)
- Evince: "command explanations" refers to $GNOME_PREFIX (that sentence 
should be suppressed actually)
- Mobile-broadband-provider-info: still has $GNOME_PREFIX instead of 
/usr in configure instructions.

Missing bits:
- libnl is lacking "short descriptions"
- SpiderMonkey too
- dhcpcd: Missing role="root" for "make install" and the following "sed..."

Defects:
- NetworkManager: I have not been able to build it without pppd, unless 
I passed --disable-ppp. So either the switch should be in the 
instructions, or pppd should be a recommended dep.
- Metacity: the run-time dependency notification-daemon has 
role="optional", while it is required at run-time (without it, GNOME 
does not start!)
- Telepathy-farstream requires gst-python, and I have not found a switch 
to prevent configure from testing it and bombing out if not present.
- Avahi does not build avahi-daemon nor avahi-dnsconfd (and maybe 
others) if --disable-daemon is passed to configure. Since those 
executables are in the short descriptions, and libdaemon is only 
optional, maybe something like 'only built if libdaemon is present and 
--disable daemon is removed from the instructions above' should be 
added. Furthermore, without those executables, the bootscript throws errors.

Race conditions. Those are not stated in the book, but could be a 
usefull information:
NSS and shared-mime-info do not build with make -jx (x>1)

There were also a few tickets I have filed and which have been already 
corrected.

GNOME starts and seems to behave OK.

Regards
Pierre
PS : built ~220 packages from "raw" lfs to GNOME (required and 
recommended deps).

OpenJDK downloads

DJ Lucas — 2012-05-14T05:16:33

I'm getting prepared to commit OpenJDK (IcedTea-2.1) to the book, 
however, I wanted to ask for opinions on how to do the downloads. The 
IcedTea tarball only contains the build environment. The makefile will 
download the needed files as part of the build process (alternately you 
can use the "download" make target). Unfortunately, the files are named 
with git sums and renamed to something useful in the source tree. In all 
other BLFS instructions, packages can be downloaded prior to building. 
So options are, download 7 files with names such as 7ceda3124828.tar.gz, 
host on anduin with the files already renamed, or just let the makefile 
handle it with a note saying that if you want to build offline, you'll 
need to download these files before hand?

xulrunner

Bruce Dubbs — 2012-05-13T03:51:33

Andy,

   I'm trying to build xulrunner and ran into the system-cairo build problem.  I 
saw your posts at https://bugzilla.mozilla.org/show_bug.cgi?id=722975

Can you tell me the latest status from your point of view.

   -- Bruce

libdrm dependencies

Ragnar Thomsen — 2012-05-12T19:39:34

I just updated libdrm, but I noticed libpthread-stubs and libatomic_ops are 
listed as required deps. I have neither of these installed, but libdrm still 
detects them on my system and builds without problems. Can they be removed as 
deps?

Configure output:
checking for PTHREADSTUBS... yes
......
checking for CAIRO... yes
checking for LIBUDEV... yes
checking for native atomic primitives... Intel
checking for PCIACCESS... yes
checking for VALGRIND... no
......

Cairo and valgrind seem to be dependencies too.

-Ragnar-

xml maintenance

Bruce Dubbs — 2012-05-12T19:39:33

I am going to make several large commits later today to clean up the book's xml. 
  For the most part I will be moving packages in the xml to an archive directory 
so that searches can exclude non-current files.

I don't anticipate any commits that will affect instructions.

   -- Bruce