lp:ubuntu/lucid/sl-modem

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

17. By أحمد المحمودي (Ahmed El-Mahmoudy) on 2010-03-16

* debian/control:
  + Really remove quilt from sl-modem-source's Depends
  + Add execstack to Build-Deps
* debian/rules: Mark slmodemd as not requiring executable stack
  (LP: #537650)

16. By Mathias Gug on 2010-01-28

Drop kernel-package recommends to a suggests so that kernel-package can be
dropped to universe.

15. By أحمد المحمودي (Ahmed El-Mahmoudy) on 2009-09-13

Remove bashism from DKMS script (LP: #429071)

14. By Scott James Remnant (Canonical) on 2009-03-05

* debian/patches/autoload.diff: Patch drivers to export the right aliases
  for auto-load on open.
* debian/sl-modem-daemon.modutils: so we don't need this file anymore, the
  rest is done by udev anyway
* debian/rules: so we don't need to install the modprobe.d file
* debian/sl-modem-daemon.dirs: so we don't need the directory
* debian/sl-modem-daemon.preinst: so remove the file on upgrade

13. By Martin Pitt on 2009-02-25

debian/sl-modem-daemon.init: Also check for modems which are
subdevices, and thus only expose themselves in "aplay -l", not in
/proc/asound/cards. Thanks to Peteris Krisjanis for the patch!
(LP: #298424)

12. By أحمد المحمودي (Ahmed El-Mahmoudy) on 2009-02-08

debian/sl-modem-daemon.postinst: moved line to add Slmodemd user to be
before lines that install directories for that user. (Closes: #514699)

11. By أحمد المحمودي (Ahmed El-Mahmoudy) on 2008-10-21

* Added support for DKMS:
  + debian/control: added dkms to sl-modem-source's Recommends field.
  + Added debian/dkms.conf.in.
  + debian/rules: copy dkms.conf.in to
    debian/sl-modem-source/usr/src/modules/sl-modem/dkms.conf
  + Added sl-modem-source.prerm and sl-modem-source.postinst

10. By أحمد المحمودي (Ahmed El-Mahmoudy) on 2008-10-10

* New upstream release. (Closes: #354908, #489289, #457637, LP: #144090, #269855)
* Adopting sl-modem. (Closes: #471888)
* Removed the following patches since they are applied upstream:
  + patch to define URB_ASYNC_UNLINK if not set
  + patch that adds alternative MODULE_PARM function (module_param)
  + patch removing the inclusion of config.h
  + patch making a biarch build with lib32 package versions
  + patches fixing buils with kernel versions 2.6.16, 2.6.18 and 2.6.26
* Remove cmd.patch and pcireg.patch as they are applied upstream.
* Used quilt patch system.
  + Moved slmodem-class-simple-to-class-meins.diff to debian/patches/
  + 01_Makefile.diff: Makefile cleanup to not loose the user-specified
    kernel version string.
  + modem_group.diff: set the default device group to "dialout".
  + 02_dial_string_size.diff: enlarge dial string buffer to 128 bytes
  + 03_Makefile_2.6.12.4.diff: workaround in drivers/Makefile for a kernel
    2.4 check accidentialy matching 2.6.12.4.
  + 04_sregs_init.diff
  + 05_st7554_license.diff: changed the MODULE_LICENSE string to
    "Dual BSD/GPL" since the license is actually a BSD license clone.
  + 06_slmodem-class-simple-to-class-meins.diff: patch for
    class/simple_class transition.
  + 08_ungrab_winmodem.diff: helper module to tell the kernel to keep away
    from unsupported PCI id numbers.
  + 10_drop_privileges.diff: slmodemd privilege dropping patch.
  + 11_logfile.diff: place debug log in /var/log/slmodem.log.
  + kernel-ver.diff: workaround for pre-2.6.18 kernels.
  + README.diff: Dropped the TROUBLESHOOTING section from README and adapted
    the feedback section, mentioned the transfer of the modem business to
    Conexant.
  + debug_level.diff: debug log file is not for end users. Users shall use
    -l <number>.
* Added fortify_source.diff patch to fix FTBFS with -D_FORTIFY_SOURCE=2
  compile flag.
* Synch'ed ungrab-winmodem.c from
  http://linmodems.technion.ac.il/packages/smartlink/ungrab-winmodem-20080126.tar.gz.
* Remove debian/control.backup* as I don't see any use for them.
* Update Standards-Version to 3.8.0.
* Added watch file.
* debian/rules: slusb support ends at 2.6.24, hence add a check for kernel
  sublevel, to prevent the attempt to install non-existing slusb.ko. Thanks
  to Maurizio Avogadro for the suggestion.
* Translation updates: Swedish, German, Basque, French, Italian, Czech,
  Portuguese, Finnish, Japanese and Russian (Closes: #501038, #501043,
  #501065, #501102, #501108, #501213, #501225, #501701, #501632, #501593).
* New translation: Arabic (Closes: #501196).
* debian/control: Added Maurizio Avogadro to Uploaders field.
* debian/sl-modem-daemon.init: changes from Ubuntu
  + sl-modem-daemon initscript waits for aplay (from alsa-utils) to report
    the card, as it appears in /proc/asound/cards before it is actually
    available to slmodemd. This extra check defeats this race.
  + Remove hard and soft ulimits for locked-in-core memory. This daemon
    unfortunately needs quite a bit of memory and it needs to be in core
    because it's doing the real-time modem DSP. This deals with LP: #144468
    and probably other bugs too.

[Maurizio Avogadro]
* debian/rules: removed code to append extra lines to modprobe.d file, since
  they are statically appended in debian/sl-modem-daemon.modutils.
* The init script can now modprobe even if the slamr module has been
  blacklisted.
* Modified 11_logfile.diff: Fixed the log path as the slmodemd daemon runs
  as unprivileged user and didn't have write permissions on /var/log.
* Added class_device.diff patch: since 2.6.25 class_device_add() and
  class_device_destroy() has been renamed, modified st7554.c accordingly.
* Removed obsolete /etc/modutils file.
* Small enhancements to the modprobe.d file:
  + verify device permissions on module install
  + modprobe cleans /dev on module remove
* Added some troubleshooting info to README.Debian.

9. By Ian Jackson on 2007-09-26

Turn ulimit -l[SH] into ulimit -[HS]l so as to avoid upsetting bash.
(Wow, dashism in an init script.)

8. By Ian Jackson on 2007-09-25

Remove hard and soft ulimits for locked-in-core memory.
This daemon unfortunately needs quite a bit of memory and it
needs to be in core because it's doing the real-time modem DSP.
This deals with LP: #144468 and probably other bugs too.