Mark vcpu as online as soon as it is added to vsched not in vmigration callThis is naturall and helps solving the followig problem.
When init is forked no migration calls have happened yet
and thus default_vsched's cpu_online_map is empty. Thus
set_cpus_allowed() for init won't set CPU_MASK_ANY to
init and all tasks will be binded to CPU0.
http://bugzilla.openvz.org/show_bug.cgi?id=350
vcpu_put() locking was broken a bit while porting on 2.6.18It should be always called with interrupts disabled.
Reworked vcpu_put() logic into finish_vsched_switch(),
which was added to 2 variants of finish_lock_switch().
http://bugzilla.openvz.org/show_bug.cgi?id=346
Signed-Off-BY: Kirill Korotaev <dev@openvz.org>
Signed-Off-BY: Pavel Emelianov <xemul@openvz.org>
[PATCH] move_task_off_dead_cpu() should be called with disabled intsmove_task_off_dead_cpu() requires interrupts to be disabled,
while migrate_dead() calls it with enabled interrupts.
Added appropriate comments to functions and added
BUG_ON(!irqs_disabled()) into double_rq_lock() and
double_lock_balance() which are the real source of such bugs.
http://bugzilla.openvz.org/show_bug.cgi?id=346
Signed-Off-By: Kirill Korotaev <dev@openvz.org>
Signed-Off-By: Pavel ...
Fix misrouted interrupts deadlocksWhile testing kernel on machine with "irqpoll" option I've caught such a
lockup:
__do_IRQ()
spin_lock(&desc->lock);
desc->chip->ack(); /* IRQ is ACKed */
note_interrupt()
misrouted_irq()
handle_IRQ_event()
if (...)
local_irq_enable_in_hardirq();
/* interrupts are enabled from now */
...
__do_IRQ() /* same IRQ we've started from */
spin_lock(&desc->lo...
Check for NULL in ve->forward_table before freeing itSuch error path may occur:
do_env_create() /* error before initialising devinet sysctls */
real_put_ve()
real_do_env_free()
free_ve_sysctl
devinet_sysctl_free
free_sysctl_clone() /* is called with NULL argument, but sholdn't */
All the rest that is called in real_do_env_free() calls
kfree() which already has builtin check for NULL.
Make kernel compile w/o magic sysrqshow_top_slabs() introduced in 028test003 kernel was
placed under #ifdef CONFIG_MAGIC_SYSRQ and was used
in oom killer.
http://bugzilla.openvz.org/show_bug.cgi?id=340
Port of recent bind_conflict fix to ipv6reverted check as it is done for ipv4
http://git.openvz.org/?p=linux-2.6.18-openvz;a=commit;h=e3b9585d9d6789e71ff4ec74e3c584cca7c07178
In journal=ordered or journal=data mode retry in ext3_prepare_write() breaks the requirements of journaling of data with respect to metadata. Fix is to call commit_write to commit zero blocks before retry.Author: Andrey Savochkin <saw@sw.ru>
Signed-Off-By: Kirill Korotaev <dev@openvz.org>
Signed-Off-By: Pavel Emelianov <xemul@openvz.org>
[CPT] Fix restore of inode attributesDue to silly mistake wrong inode mode were set on restore
(cpt_mode were used instead of cpt_i_mode).
(#71135, #71161)
This patch implements safer printk from places like NMI or from under critical locks (runqueue lock and so on).Printk from NMI watchdog was left unmodified in this version, since
other patches are also fiddling with it. NMI watchdog requires great
care and consideration, and is left for future inspection.
Fix potential race in accessing dentry flagWhen mounting new tree DCACHE_VIRTUAL may be added to
mnt_root->d_flags and this is done w/o d_lock.
Since this is done after graft_tree() mnt_root may
be returned by link_path_walk()->follow_mount() pair.
Thus flag is to be set before dentry becomes visible.
Make flag inheritance before calling graft_tree.
KSYM's were rewritten to always access any KSYM'bol via pointer.This simplifies code and makes it work independant of
CONFIG_XXX=y or m settings.
http://bugzilla.openvz.org/show_bug.cgi?id=284
http://bugzilla.openvz.org/show_bug.cgi?id=283
[VETH] We need to be sure that pair device will not be unregistered while we are in veth_xmit().So we close VE0 device and clear pair pointers before
unregistering network devices to avoid this race. New
function veth_pair_del() is created to avoid code duplication.
dev_close() and unregister_netdevice() are protected with rtnl_lock().
(#70690)
Move declarations of virtalized filesystems to headers.This includes
* proc
* sysfs
* devpts
* tmpfs
Since they can not be made as module it's OK to
use them as is.
[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
Ported from 026 kernel.
[PATCH] usbfs: private mutex for open, release, and removeThe usbfs code doesn't provide sufficient mutual exclusion among open,
release, and remove. Release vs. remove is okay because they both
acquire the device lock, but open is not exclusive with either one. All
three routines modify the udev->filelist linked list, so they must not
run concurrently.
Apparently someone gave this a minimum amount of thought in the past by
explicitly acquiring the...
[PATCH] md: check bio address after mapping through partitions.Partitions are not limited to live within a device. So
we should range check after partition mapping.
Note that 'maxsector' was being used for two different things. I have
split off the second usage into 'old_sector' so that maxsector can be
still be used for it's primary usage later in the function.
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off...
[PATCH] IPV6: fix lockup via /proc/net/ip6_flowlabel [CVE-2006-5619]There's a bug in the seqfile handling for /proc/net/ip6_flowlabel, where,
after finding a flowlabel, the code will loop forever not finding any
further flowlabels, first traversing the rest of the hash bucket then just
looping.
This patch fixes the problem by breaking after the hash bucket has been
traversed.
Note that this bug can cause lockups and oopses, and is trivially invoked
by an unpr...
[PATCH] tcp: cubic scaling errorDoug Leith observed a discrepancy between the version of CUBIC described
in the papers and the version in 2.6.18. A math error related to scaling
causes Cubic to grow too slowly.
Patch is from "Sangtae Ha" <sha2@ncsu.edu>. I validated that
it does fix the problems.
See the following to show behavior over 500ms 100 Mbit link.
Sender (2.6.19-rc3) --- Bridge (2.6.18-rt7) ------- Receiver (2.6....
[PATCH] JMB 368 PATA detectionThe Jmicron JMB368 is PATA only so has the PATA on function zero. Don't
therefore skip function zero on this device when probing
Signed-off-by: Alan Cox <alan@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
OpenVZ team
Chris Wright
