Planning/2.5

From QEMU
Revision as of 15:31, 11 December 2015 by Eblake (talk | contribs) (→‎Known issues: [x-]blockdev-insert-medium)

Release Schedule

2015-08-11 Beginning of development phase
2015-10-15 Soft feature freeze. All features

should have patches on the list by this date; major features should have initial code committed.

2015-11-12 2015-11-13 Hard feature freeze. Tag -rc0,

only bug fixes committed after this point

2015-11-19 2015-11-20 Tag v2.5.0-rc1
2015-11-26 Tag v2.5.0-rc2
2015-12-03 2015-12-07 Tag v2.5.0-rc3
2015-12-11 Tag v2.5.0-rc4
2015-12-10 2015-12-16 Tag v2.5.0

Known issues

Please list all patch series or bugs which need to be fixed for the release here. That way I can be sure I don't miss any when I'm rolling release candidates.

Still unfixed

  • qemu-iotests 028 sometimes fails; looks like a race causing one less '(qemu) ' prompt than expected in the .out file (probably not show-stopper)
  • mark blockdev-{insert,remove}-medium as experimental

Fixed in rc4

  • crash on migration if virtio-9p device was hot-unplugged http://patchwork.ozlabs.org/patch/553966/
  • MAINTAINERS update for virtio-9p
  • Update translation files with `make -C po update` and commit the result.
  • xbzrle migration crash/corruption (see Fix xbzrle vs last_sent_block update)

Fixed in rc3

Fixed in rc0

  • netmap.c compile failures
  • Mac OS 9 compatibility improvements (upstream rework)
  • fix Highbank model by adding board setup blob
  • migration: fix analyze-migration.py script
  • revert of VHOST_RESET_DEVICE change
  • bugs in ARM gdbstub singlestep and breakpoint handling
  • configure barfs if there's no c++ compiler (fortify-source test is assuming $cxx is defined)
  • Zynq ADC controller
  • migration doesn't work on OSX (migrate_init zeroes out the mutex)
  • migration/ram.c 32-bit compile fixes

Also see Planning/2.5/Testing.

Targeted Features

See the ChangeLog/2.5 for full details.