[Debian] Re: OpenVZ 028test018 patch for Debian

Kir Kolyshkin kir at openvz.org
Fri Mar 9 10:54:10 EST 2007


Ola Lundqvist wrote:
> Privet Vasily
>
> On Fri, Mar 09, 2007 at 06:22:28PM +0300, Vasily Tarasov wrote:
>   
>> Hello, Ola,
>>
>> The most recent released OpenVZ patch against Debian kernel is ready and
>> can be downloaded from
>> http://download.openvz.org/~vass/diff-ovz-028.18-deb.gz
>>     
>
> Thanks a lot! Do you have a list of changes that was made in this
> version?
>   
015 to 018?

Short info is here: 
http://openvz.org/news/updates/devel-kernel-2.6.18-028test018.1
Let me quote: "Merged mainstream stable updates up to 2.6.18.8. Critical 
fix in task/UBC handling. Fixes in ext3, NFS for VE, simfs. 
Checkpointing fixes and improvements".

Log can be taken from:
  http://git.openvz.org/?p=linux-2.6.18-openvz;a=log;h=028test018
or directly from git, if you have openvz git tree cloned locally, for 
example
   git log 028test015..028test018
See git-log(3) for details.

Here's a list of patches:

linux-2.6.18-028test018 released
Missed ve context switch in NFS RPC code.
ext3 error behavior was broken in linux kernels since 2.5.x versions
task puts UBC before the task becomes invisible for all (e.g. /proc),
EXT3_ERRORS_CONTINUE should be taken from the superblock as default 
value for
EXT2_ERRORS_CONTINUE should be read from the sb as default error behaviour.
Revert diff-ms-ext3-retries-20061109 until all the issues are resolved.
Patch from mainstream:
Same story as with p4-clockmod. Driver does set_cpus_allowed(cpu), then
Merge git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-2.6.18.y
Linux 2.6.18.8
fix umask when noACL kernel meets extN tuned for ACLs
Fix for shmem_truncate_range() BUG_ON()
make ppc64 current preempt-safe
fix msync error on unmapped area
read_zero_pagealigned() locking fix
Fix incorrect user space access locking in mincore() (CVE-2006-4814)
x86_64: fix 2.6.18 regression - PTRACE_OLDSETOPTIONS should be accepted
V4L: buf_qbuf: fix videobuf_queue->stream corruption and lockup
V4L: tveeprom: autodetect LG TAPC G701D as tuner type 37
V4L: fix ks0127 status flags
V4L: Fix quickcam communicator driver for big endian architectures
v4l: cx88: Fix leadtek_eeprom tagging
v4l: cx2341x audio_properties is an u16, not u8
dvb-core: fix bug in CRC-32 checking on 64-bit systems
IB/mad: Fix race between cancel and receive completion
hfs_fill_super returns success even if no root inode (CVE-2006-6056)
grow_buffers() infinite loop fix (CVE-2006-5757, CVE-2006-6060)
i2c: fix broken ds1337 initialization
IB/srp: Fix FMR mapping for 32-bit kernels and addresses above 4G
SCSI: add missing cdb clearing in scsi_execute()
Don't leak NT bit into next task
bcm43xx: Fix for oops on ampdu status
bcm43xx: Fix for oops on resume
Linux 2.6.18.7
[PATCH] Fix a free-wrong-pointer bug in nfs/acl server (CVE-2007-0772)
Over uncharged privvmpages in 32bit VE on x86_64
Replace wrong rcu_lock by rcu_unlock
linux-2.6.18-028test017 released
Change default for per/UB TW buckets limitations
[BC] Don't uncharge resources in RCU callbacks
[BC] Fix uncharging of privvmpages on error path
[BC] Fix compilation warning in ub_proc.c
Bad inode -EIO screwup. CVE-2006-5753 ported from mainstream.
[CPT] use BUILD_BUG_ON() where appropriate
[BC] Return ENOMEM if pty charge fails
[SIMFS] get lower vfsmount on simfs mount
Introduce and use for_each_ve()
Unresolved symbols should abort build.
IPC walking symbols are VE-related, not CPT.
linux-2.6.18-028test016 released
[CPT] Convert some printk-s to eprintks
[CPT] Unmap ptep in page_get_desc early
Make it possible to change page state info dumped from /proc/meminfo
[IA64] do not reset loadrs after clone
[CPT] Change ubc image format to remove magic numbers like 6 and 12
[CPT] adjust ubc limits on restore
[CPT] fix ubc save/restore procedure and change order of ubc params in image
[CPT] Add check for image version
[BC] Revert commits that add per-vma RSS accounting
Allow process access /proc/$PID/fd after setuid()
[CPT] do not forget to release mm semaphore in error path


More information about the Debian mailing list