OpenVZ-legacy
  1. OpenVZ-legacy

linux-2.6.16-openvz

Public
No branch, tag or commit 'refs/tags/026test012.1' exists in repository 'linux-2.6.16-openvz'
AuthorCommitMessageCommit dateIssues
Pavel EmelianovPavel Emelianov
46d1e25bce7[UBC] Set correct precharge values for init_task.Otherwise file freeing will happen in "swapper" context and will spoil all statistics due to "negative" unsigned long value. http://bugzilla.openvz.org/show_bug.cgi?id=322
OpenVZ teamPavelOpenVZ team
d55df66c6e0linux-2.6.16-026test020 released
Dmitry MishinPavel EmelianovDmitry Mishin
d59101ac42f[CPT] Fix self-deadlock in CPT codeRemoved excessive ip_conntrack lock in restoring.
Denis V. LunevPavel EmelianovDenis V. Lunev
5e970db13f6Fix venet and veth statistics broken by NETIF_F_LLTX addition.This is a fix for previous patch.
Andrey MirkinPavel EmelianovAndrey Mirkin
7b4df337954[CPT] The second part of previous commit.Size of ip_conntrack_help union differs on i386 and x86_64 because of pointers used in ip_ct_pptp_master structure:
Andrey MirkinPavel EmelianovAndrey Mirkin
0d0674785b1Size of ip_conntrack_help union differs on i386 and x86_64 because of pointers used in ip_ct_pptp_master structure:union ip_conntrack_help { struct ip_ct_pptp_master ct_pptp_info; struct ip_ct_ftp_master ct_ftp_info; struct ip_ct_irc_master ct_irc_info; }; struct ip_ct_pptp_master { enum pptp_ctrlsess_state sstate; enum pptp_ctrlcall_state cstate; u_int16_t pac_call_id; u_int16_t pns_call_id; struct ip_ct_gre_keymap *keymap_orig, *keymap_repl...
Pavel EmelianovPavel Emelianov
f7defed6e41[VZDQ] Exported needed user_get_super() symbol.
Pavel EmelianovPavel Emelianov
6ec8dd465a8Fixed cpt/Makefile for x86_64...
Dmitry MishinPavel EmelianovDmitry Mishin
02b2da38165Virtualized ipt_owner module. Compatibility issue: ipt_mac bit mask is changed, all users of this module should recompile it's vzctl with new headers.Made so for compatibility with 2.6.9 branch.
Dmitry MishinPavel EmelianovDmitry Mishin
d7936989124Ported missed compat-related patches from 2.6.9 branch.
Denis V. LunevPavel EmelianovDenis V. Lunev
92e3fba0507This patch fixed venet and veth statistics broken by NETIF_F_LLTX addition.
Andrey MirkinPavel EmelianovAndrey Mirkin
27ef07e0fa6Enables creation of veth pair in VE0. - ve0.is_running is initialized to 1 - ve0.op_sem is initialized in vecalls_init()
Alexey KuznetsovPavel EmelianovAlexey Kuznetsov
fbec7e02e5c[CPT] mlockall() prevents restoreIf a program in does mlockall(), one can make configurations with unreadable mlocked VMAs, which are not really in core. This is sort of a linux feature. The reason is that mlock*() set VM_LOCKED even if they cannot bring in pages. The fix is to ignore -EFAULT, returned by mlock().
Alexey KuznetsovPavel EmelianovAlexey Kuznetsov
f89badcef13[CPT] [PATCH 2/2] massive module refcnt leakage while restore(module's part for previous patch)
Alexey KuznetsovPavel EmelianovAlexey Kuznetsov
113404ea1cf[CPT] [PATCH 1/2] massive module refcnt leakage while restoreBug: detach of passed FDs is made directly in af_unix.c, bypassing skb destructor sometimes, so we leak module refcnt grabbed, when we attached our private destructor.
Alexey KuznetsovPavel EmelianovAlexey Kuznetsov
9382f03f8a6[CPT] crash in cpt_resume().Actually, it is known bug, which has been fixed in hurry and the fix did not cover all the places. Task can have tsk->sighand==NULL, if it is already released.
Alexey KuznetsovPavel EmelianovAlexey Kuznetsov
e0ee32c5938[CPT] another bug in restoring deleted af_unix socketsOne case was missed. We assumed that if path_lookup() fails with -ENOENT, it means that we can bind to this name. But directory can be deleted! So, instead, switch to attempt to bind() to name. And if it fails, bind() to temporary name instead.
Denis V. LunevPavel EmelianovDenis V. Lunev
0cade7cc35f[VZDQ] Always fail dentry revalidate check for /proc quotafile
Vasily TarasovPavel EmelianovVasily Tarasov
5879b3b33f5[VZDQ] This patch sets correct size on /proc/vz/aquota/*/aquota.* (#59920)Signed-off-by: Vasily Tarasov <vtaras@openvz.org> Signed-off-by: Denis Lunev <den@openvz.org>
Denis V. LunevPavel EmelianovDenis V. Lunev
50389c2d083[VZDQ] Restore quota symlink to /proc on vz_quota_on corrupted by quotacheck(#66949)
Vasily AverinPavel EmelianovVasily Averin
730a319cd71Remove VE init pid + 1024 (virtual init pid).(#68754).
Denis V. LunevPavel EmelianovDenis V. Lunev
0becb52714bFix of errorneous 'nighbour table overflow' message.The problem: - VE should have exactly one ARP entry - it is allocated from UBC slab and the allocation is failed due to UBC - EBUFS is returned to the caller The cure: - ENOMEM in such a case (#65836)
Denis V. LunevPavel EmelianovDenis V. Lunev
a111609e911[VENET] TX for venet device does not require device queue locking
Andrey MirkinPavel EmelianovAndrey Mirkin
8abf2dc2f6bFix local port range patch compilation on ia64
Denis V. LunevPavel EmelianovDenis V. Lunev
1ed3e55da64[VETH] TX for veth device do not require device queue locking
Dmitry MishinPavel EmelianovDmitry Mishin
dbdf80eb550Fixed printed debug info in case of "busy inodes"(Port of 2.6.9 patch, #68575)
OpenVZ teamPavelOpenVZ team
9fc9c9920dblinux-2.6.16-026test019 released
OpenVZ teamPavel EmelianovOpenVZ team
aeae84ef040MMerged 2.6.16.30-pre1 from /linux/kernel/git/stable/linux-2.6.16.y
Alexandr AndreevPavel EmelianovAlexandr Andreev
00296bea5caUnstatic iptables init/fini functions to work when built-in.http://bugzilla.openvz.org/show_bug.cgi?id=283
Alexandr AndreevPavel EmelianovAlexandr Andreev
5dcb5cbd694[VETH] Unstatic veth_open() to work when built-in.
Alexandr AndreevPavel EmelianovAlexandr Andreev
dbf73d94393[CPT] Use KSYM_EQUAL_TO() for veth_open checks.
Alexandr AndreevPavel EmelianovAlexandr Andreev
7c858f54102Add KSYM_EQUAL_TO() macro for CONFIG_MODULES=n case work
Andrey MirkinPavel EmelianovAndrey Mirkin
c04fd6bb990We should check return code of execve("iptables") to be -ENOENT, not ENOENT.
Vasily TarasovPavel EmelianovVasily Tarasov
1f57d9a7d07cap_netlink_recv() should check for CAP_NET_ADMIN also.Now zebra should work. http://forum.openvz.org/index.php?t=tree&goto=6283&#msg_6283
Dmitry MishinPavel EmelianovDmitry Mishin
5bb070631aaFixed iptables modules refcount on VE stop.In 2.6.16 ipt_flush functions are excessive, as cleanup is performed by ipt_unregister_table() funcs.
Andrey MirkinPavel EmelianovAndrey Mirkin
7e5ea9c1072Remove /proc/vz/veth entry from VPS.It is a bug that this proc entry exists in both VE0 and VPS. This patch fixes this. http://bugzilla.openvz.org/show_bug.cgi?id=271
Dmitry MishinPavel EmelianovDmitry Mishin
8aeb60b8da9Additional iptables security checks.This patch fixes possible oopses or memory disclosure on new iptable entries addition. VE owner can break the node. (#65826)
Pavel EmelianovPavel Emelianov
63491c5c23f[UBC] Fix of kmemsize optimisation accounting for NUMA caseskmem_cache_alloc_node() is used in NUMA only and calls ub_slab_charge() with enabled IRQs which may spoil kmem prechrged value for current task. Potential fix for http://bugzilla.openvz.org/show_bug.cgi?id=288
Pavel EmelianovPavel Emelianov
9eda540ced8Fix dcache accounting optimisation for NUMA=y case.When walking nodes need to walk each online node.
Dmitry MishinPavel EmelianovDmitry Mishin
0086a8a9790gcc-4.1.1 compilation fix for cycles_t_to_jiffies()
Pavel EmelianovPavel Emelianov
4165f1fa371[UBC] Don't allow precharged files exhaust kmemsize When file is put it may be added to precharged value to some task thus holding UB_NUMFULE and UB_KMEMSIZE resources.The problem is that files do not start uncharging till ub_barrier_farnr() is hit for UB_NUMFILE. For ub0 ub_barrier_farnr() can happen only after hitting kmemsize barrier. Thus kmemsize reurce gets completely exhausted. On 2.6.9 this problem is not easyli reproducible as files are put in the context of closing task usually. On 2.6.16 files are put via RCU and thus - in other task's context. h...
Pavel EmelianovPavel Emelianov
43f07723152[UBC] Changes in oom killer This patch adjust the logic of oom killer to take into account UBC statistics: 1. ub with maxmum overdraft of resources is selected 2. task within this ub is selected and killedGroup with memory consumption below oomguarpages.barrier is not oom-killed. If several oom killers happen at once only one of them succeeds to prevent massive kills.
Adrian BunkAdrian Bunk
196be852be1Linux 2.6.16.30-pre1
Henk VergonetAdrian BunkHenk Vergonet
ce551cbbec5USB: add YEALINK phones to the HID_QUIRK_IGNORE blacklistKeys on Yealink based phones will not function properly when using the generic HID driver. This patch prevents the generic HID code from grabbing the device before the regular yealink driver can get a grip on it. Signed-off-by: Henk Vergonet <Henk.Vergonet@gmail.com> Signed-off-by: Vojtech Pavlik <vojtech@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Adrian Bunk <b...
Henk VergonetAdrian BunkHenk Vergonet
d93beca96a8USB: Fix unload oops and memory leak in yealink driverThis patch fixes a memory leak and a kernel oops when trying to unload the driver, due to an unbalanced cleanup. Thanks Ivar Jensen for spotting my mistake. Signed-off-by: Henk Vergonet <henk.vergonet@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Adrian Bunk <bunk@stusta.de>
Martin SchwidefskyAdrian BunkMartin Schwidefsky
579adde438bkernel/kmod.c: fix a race condition in usermodehelper.There is a race between call_usermodehelper_keys, __call_usermodehelper and wait_for_helper. It should only happen if preemption is enabled or on a virtualized system. If the cpu is preempted or put to sleep by the hypervisor in __call_usermodehelper between the creation of the wait_for_helper thread and the second check on sub_info->wait, the whole execution of wait_for_helper including the c...
Mark M. HoffmanAdrian BunkMark M. Hoffman
ab469df3639I2C: fix 'ignore' module parameter handlingThis patch fixes a bug in the handling of 'ignore' module parameters of I2C client drivers. Signed-off-by: Mark M. Hoffman <mhoffman@lightlink.com> Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Adrian Bunk <bunk@stusta.de>
Pavel EmelianovPavel Emelianov
f11ee2353f1[UBC] Move vma rss accounting under separate config optionEach vm_area_struct has it's own rss counter which was supposed to help debugging of unused_privvmpages and do not bring any new functionality to kernel. So make it possible to turn it off. New config option is CONFIG_UBC_DEBUG_VMRSS.
OpenVZ teamPavelOpenVZ team
cb8603399b0linux-2.6.16-026test018 released
Hartmut HackmannAdrian BunkHartmut Hackmann
cec8f39e59cV4L/DVB: TDA8290 updateThis patch - works around a bug in the I2C bridge that makes the initialization of the TDA10046 fail on recent LifeView cards - puts the AGC output to tristate in sleep mode. This is necessary for recent hybrid cards that switch the AGC via tristateing. Signed-off-by: Hartmut Hackmann <hartmut.hackmann@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org> Acked-by: Mich...