[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#683674: Changelogs for intel-microcode/1.20120606.3



Changelogs since the version in Wheezy:

intel-microcode (1.20120606.3) unstable; urgency=low

   * initramfs: while creating the initramfs, if we need to iucode_tool
     --scan-system, attempt to modprobe cpuid if cpu/cpuid device is missing,
     and report an error if it doesn't work.  Thanks to Sebastian Andrzej
     Siewior for a good suggestion on how to fix it (closes: #683161)
   * README.Debian: add "modprobe cpuid" to example
   * debian/control: use better Vcs-browser URI that is properly
     handled by the current alioth redirector.

 -- Henrique de Moraes Holschuh <hmh@debian.org>  Sun, 29 Jul 2012 11:09:44 -0300

intel-microcode (1.20120606.2) unstable; urgency=low

   * Fix README.source to reflect that cpu-signatures.txt processing
     was moved to the toplevel Makefile
   * Update diff-latest-pack.sh to really find iucode_tool

 -- Henrique de Moraes Holschuh <hmh@debian.org>  Sat, 21 Jul 2012 18:10:47 -0300

intel-microcode (1.20120606.1) unstable; urgency=low

   * Change to ABI 1:
     + Ship binary microcode in /lib/firmware
     + Add initramfs helpers to install and load binary microcode on boot
     + Call update-initramfs on package upgrades and removals
     + Use non-deprecated kernel interface to interact with kernel
       (sysfs+fw loader)
   * Include microcode for older processors.  This should help some
     older boxes for which microcode was not being shipped by Intel
     anymore but which still have users, with the trade-off that we
     will also ship some useless and mostly useless microcode
   * Do not ship i686-only microcode in the amd64 binary package.  We
     still ship all microcode in the i386 binary package, to support
     64bit processors running i?86 userspace transparently
   * Switch myself to maintainer, and Giacomo to uploader to better
     reflect who is responsible for any bugs this could cause...
   * Switch to the 3.0 (native) package format as it doesn't make practical
     sense to base the source package on the Intel upstream tarball anymore
     because we use all past Intel microcode releases as source
     + Use xz to compress the tarball, it does a _much_ better job than
       bzip2 and gzip for this package
     + Override lintian warning about switch to native packaging, as it was
       done on purpose.  It can be removed in the future
   * Drop CDBS, switch to classic (less obfuscated/much better documented)
     debhelper build
   * Switch to debhelper v7, which is good enough for Debian Lenny and later
     don't use a newer mode for now, to facilitate backporting
   * Document in README.source:
     + this package must be trivial to backport to oldstable and stable
       (i.e. Debian Lenny and Debian Squeeze ATM)
     + how to add new upstream microcode packs and microcode overrides
     + other relevant details related to the lack of Intel changelogs
   * Build-Depend on iucode-tool to handle binary microcode, merge
     microcode packs and overrides, and split into firmware files
   * Drop support for microcode.ctl, as it cannot handle binary
     microcode or the non-deprecated kernel interface
   * Update package short and long descriptions
   * Add a NEWS file to explain all the behaviour changes
   * Recommend iucode-tool to support optional selective microcode
     selection for the initramfs (reduces microcode size greatly)
   * Change to priority: standard.  This package should be installed in
     every Intel-based Debian system, which is unfortunately impossible
     since it is non-free, but at least mark it as such
   * add debian/diff-latest-pack.sh utility (not shipped in the binary
     package) to help produce the "upstream changelogs"
   * debian/control: add Vcs-* fields

 -- Henrique de Moraes Holschuh <hmh@debian.org>  Fri, 13 Jul 2012 15:23:23 -0300 

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh


Reply to: