udev alpha testing

[Harald Hoyer]

I would be glad, if the really brave could compile, install and test:
ftp://people.redhat.com/harald/udev/076-1/

Beware this could end in a non-bootable system.

This version removes the hardware initializing/module loading phase from rc.sysinit. Udev
should do most, if 
not all job. I am interested in any missing modules, that were loaded before.

Thx for testing.

--
Harald

[Bart Vanbrabant]

Harald Hoyer wrote:
>  I would be glad, if the really brave could compile, install and
> test:
>  ftp://people.redhat.com/harald/udev/076-1/
>
>  Beware this could end in a non-bootable system.
>
>  This version removes the hardware initializing/module loading phase
>  from rc.sysinit. Udev should do most, if not all job. I am interested
>  in any missing modules, that were loaded before.
>
>  Thx for testing.
>
>  -- 
>  Harald
>  
The udev srpm doesn't seem to compile on the lastest rawhide. I get this
output when I do $ rpmbuild --rebuild udev-076-1.src.rpm

[root@laptop-bart tmp]# rpmbuild --rebuild udev-076-1.src.rpm
Installing udev-076-1.src.rpm
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
warning: user harald does not exist - using root
warning: group harald does not exist - using root
Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.57665
+ umask 022
+ cd /usr/src/redhat/BUILD
+ LANG=C
+ export LANG
+ unset DISPLAY
+ cd /usr/src/redhat/BUILD
+ rm -rf udev-076
+ /usr/bin/bzip2 -dc /usr/src/redhat/SOURCES/udev-076.tar.bz2
+ tar -xf -
+ STATUS=0
+ '[' 0 -ne 0 ']'
+ cd udev-076
++ /usr/bin/id -u
+ '[' 0 = 0 ']'
+ /bin/chown -Rhf root .
++ /usr/bin/id -u
+ '[' 0 = 0 ']'
+ /bin/chgrp -Rhf root .
+ /bin/chmod -Rf a+rX,u+w,g-w,o-w .
+ echo 'Patch #1 (udev-075-daemon.patch):'
Patch #1 (udev-075-daemon.patch):
+ patch -p1 -b --suffix .udevdaemon -s
+ cp /usr/src/redhat/SOURCES/udev.html .
+ exit 0
Executing(%build): /bin/sh -e /var/tmp/rpm-tmp.3874
+ umask 022
+ cd /usr/src/redhat/BUILD
+ cd udev-076
+ LANG=C
+ export LANG
+ unset DISPLAY
+ make USE_KLIBC=false USE_SELINUX=true USE_STATIC=true STRIP=/bin/true
udevdir=/dev USE_LOG=false DEBUG=false 'EXTRAS=        
extras/scsi_id          extras/ata_id         extras/usb_id        
extras/edd_id         extras/volume_id    ' all
Creating udev_version.h
Compiling sysfs_class.c:                                              [OK]
Compiling sysfs_device.c:                                             [OK]
Compiling sysfs_dir.c:                                                [OK]
Compiling sysfs_driver.c:                                             [OK]
Compiling sysfs_utils.c:                                              [OK]
Compiling dlist.c:                                                    [OK]
Creating library libsysfs.a:                                          [OK]
Running ranlib:                                                       [OK]
Compiling udev_event.c:                                               [OK]
Compiling udev_device.c:                                              [OK]
Compiling udev_config.c:                                              [OK]
Compiling udev_add.c:                                                 [OK]
Compiling udev_remove.c:                                              [OK]
Compiling udev_db.c:                                                  [OK]
Compiling udev_rules.c:                                               [OK]
Compiling udev_rules_parse.c:                                         [OK]
Compiling udev_utils.c:                                               [OK]
Compiling udev_utils_string.c:                                        [OK]
Compiling udev_utils_file.c:                                          [OK]
Compiling udev_utils_run.c:                                           [OK]
Compiling udev_libc_wrapper.c:                                        [OK]
Compiling udev_selinux.c:                                             [OK]
Creating library libudev.a:                                           [OK]
Running ranlib:                                                       [OK]
Compiling udev.c:                                                     [OK]
Linking udev:                                                        
[ERROR]
  gcc -Wl,-warn-common -static udev.o -o udev libudev.a
libsysfs/libsysfs.a -ls
  elinux -lsepol
  libudev.a(udev_device.o): In function `udev_init_device':
  /usr/src/redhat/BUILD/udev-076/udev_device.c:87: undefined reference
to `__ct
  ype_b(a)GLIBC_2.0'
  libudev.a(udev_config.o): In function `parse_config_file':
  /usr/src/redhat/BUILD/udev-076/udev_config.c:128: undefined reference
to `__c
  type_b(a)GLIBC_2.0'
  libudev.a(udev_config.o): In function `get_key':
  /usr/src/redhat/BUILD/udev-076/udev_config.c:56: undefined reference
to `__ct
  ype_b(a)GLIBC_2.0'
  libudev.a(udev_rules.o): In function `import_keys_into_env':
  /usr/src/redhat/BUILD/udev-076/udev_rules.c:179: undefined reference
to `__ct
  ype_b(a)GLIBC_2.0'
  libudev.a(udev_rules.o): In function `get_key':
  /usr/src/redhat/BUILD/udev-076/udev_rules.c:98: undefined reference to
`__cty
  pe_b(a)GLIBC_2.0'
 
libudev.a(udev_rules.o):/usr/src/redhat/BUILD/udev-076/udev_rules.c:280:
more
   undefined references to `__ctype_b(a)GLIBC_2.0' follow
  collect2: ld returned 1 exit status
make: *** [udev] Error 1
error: Bad exit status from /var/tmp/rpm-tmp.3874 (%build)


Bart

-- 
Bart Vanbrabant <bart.vanbrabant(a)zoeloelip.be&gt;
PGP fingerprint: 093C BB84 17F6 3AA6 6D5E  FC4F 84E1 FED1 E426 64D1

[Felipe Alfaro Solana]

>  I would be glad, if the really brave could compile, install and
> test:
>  ftp://people.redhat.com/harald/udev/076-1/
>
>  Beware this could end in a non-bootable system.
>
>  This version removes the hardware initializing/module loading phase from rc.sysinit. Udev
> should do most, if
>  not all job. I am interested in any missing modules, that were loaded before. 
Impressive. Not only does it boot faster (with selinux=0), but I can't
see any real difference between 076-1 and 075-4: exactly the same
modules get loaded.

[dragoran]

Felipe Alfaro Solana wrote:

> >I would be glad, if the really brave could compile, install and
> test:
> >ftp://people.redhat.com/harald/udev/076-1/
> >
> >Beware this could end in a non-bootable system.
> >
> >This version removes the hardware initializing/module loading phase from rc.sysinit.
> Udev should do most, if
> >not all job. I am interested in any missing modules, that were loaded before.
> >    
> >
>
> Impressive. Not only does it boot faster (with selinux=0), but I can't
> see any real difference between 076-1 and 075-4: exactly the same
> modules get loaded.
>
>   
>  
why with selinux=0 ? is it slower with selinux enabled?

[Felipe Alfaro Solana]

>  why with selinux=0 ? is it slower with selinux enabled? 
It seems policy 2.X is having some trouble with udev.

[Harald Hoyer]

Felipe Alfaro Solana wrote:
> >I would be glad, if the really brave could compile, install and
> test:
> >ftp://people.redhat.com/harald/udev/076-1/
> >
> >Beware this could end in a non-bootable system.
> >
> >This version removes the hardware initializing/module loading phase from rc.sysinit.
> Udev should do most, if
> >not all job. I am interested in any missing modules, that were loaded before.
>  
>  
>  Impressive. Not only does it boot faster (with selinux=0), but I can't
>  see any real difference between 076-1 and 075-4: exactly the same
>  modules get loaded.
>   
well, did you update initscripts also?

[Felipe Alfaro Solana]

>  well, did you update initscripts also? 
Of course :-)