Thomas Gleixner
f2530dc71c
kthread: Prevent unpark race which puts threads on the wrong cpu ...
The smpboot threads rely on the park/unpark mechanism which binds per
cpu threads on a particular core. Though the functionality is racy:
CPU0 CPU1 CPU2
unpark(T) wake_up_process(T)
clear(SHOULD_PARK) T runs
leave parkme() due to !SHOULD_PARK
bind_to(CPU2) BUG_ON(wrong CPU)
We cannot let the tasks move themself to the target CPU as one of
those tasks is actually the migration thread itself, which requires
that it starts running on the target cpu right away.
The solution to this problem is to prevent wakeups in park mode which
are not from unpark(). That way we can guarantee that the association
of the task to the target cpu is working correctly.
Add a new task state (TASK_PARKED) which prevents other wakeups and
use this state explicitly for the unpark wakeup.
Peter noticed: Also, since the task state is visible to userspace and
all the parked tasks are still in the PID space, its a good hint in ps
and friends that these tasks aren't really there for the moment.
The migration thread has another related issue.
CPU0 CPU1
Bring up CPU2
create_thread(T)
park(T)
wait_for_completion()
parkme()
complete()
sched_set_stop_task()
schedule(TASK_PARKED)
The sched_set_stop_task() call is issued while the task is on the
runqueue of CPU1 and that confuses the hell out of the stop_task class
on that cpu. So we need the same synchronizaion before
sched_set_stop_task().
Reported-by: Dave Jones <davej@redhat.com>
Reported-and-tested-by: Dave Hansen <dave@sr71.net>
Reported-and-tested-by: Borislav Petkov <bp@alien8.de>
Acked-by: Peter Ziljstra <peterz@infradead.org>
Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: dhillf@gmail.com
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1304091635430.21884@ionos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2013-04-12 14:18:43 +02:00
Merge branch 'next' of git://git.infradead.org/users/vkoul/slave-dma
2013-02-26 09:24:48 -08:00
Fairly unexciting MTD merge for 3.9:
2013-03-02 16:33:54 -08:00
can: rework skb reserved data handling
2013-01-28 18:17:25 -05:00
libceph: add support for HASHPSPOOL pool flag
2013-02-26 15:03:06 -08:00
arm-soc: soc-specific updates
2013-02-21 15:27:22 -08:00
Merge branch 'testing' of github.com:ceph/ceph-client into v3.8-rc5-testing
2013-01-30 07:54:34 -06:00
extcon: gpio: Rename filename of extcon-gpio.c according to kernel naming style
2013-02-13 08:35:43 -08:00
hsi: fix kernel-doc warnings
2013-03-01 13:39:00 -08:00
Input: atmel_mxt_ts - Support for touchpad variant
2013-03-08 16:03:29 -08:00
iio: Fix build error seen if IIO_TRIGGER is defined but IIO_BUFFER is not
2013-03-02 16:31:21 +00:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
2013-02-20 11:00:43 -08:00
ImgTec Meta architecture changes for v3.9-rc1
2013-03-03 12:06:09 -08:00
Merge branch 'for-3.9' of git://linux-nfs.org/~bfields/linux
2013-02-28 18:02:55 -08:00
char/misc driver fixes for 3.9-rc4
2013-03-28 15:51:33 -07:00
Main batch of InfiniBand/RDMA changes for 3.9:
2013-02-26 11:41:08 -08:00
mmc: sh_mobile_sdhi: remove unused .pdata field
2013-02-24 14:37:22 -05:00
mtd: nand: reintroduce NAND_NO_READRDY as NAND_NEED_READRDY
2013-03-14 12:48:54 +00:00
Merge branch 'master' of git://1984.lsi.us.es/nf
2013-02-26 17:24:26 -05:00
nfsd: Store ex_anon_uid and ex_anon_gid as kuids and kgids
2013-02-13 06:16:08 -08:00
pinctrl: generic: rename input schmitt disable
2013-02-15 09:52:29 +01:00
Fairly unexciting MTD merge for 3.9:
2013-03-02 16:33:54 -08:00
bq2415x_charger: Add support for offline and 100mA mode
2013-02-16 13:34:10 -08:00
regulator: core: update kernel documentation for regulator_desc
2013-03-01 06:07:22 +00:00
sched: move RR_TIMESLICE from sysctl.h to rt.h
2013-02-22 09:20:11 -08:00
Staging tree update for 3.9-rc1
2013-02-21 12:11:44 -08:00
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next into for-davem
2013-02-08 13:16:17 -05:00
NFS client bugfixes for Linux 3.9
2013-03-02 16:46:07 -08:00
usb: add find_raw_port_number callback to struct hc_driver()
2013-03-25 10:39:17 -07:00
x86, ACPI, mm: Revert movablemem_map support
2013-03-02 09:34:39 -08:00
gpiolib-acpi: Add ACPI5 event model support to gpio.
2013-01-29 11:47:06 +01:00
libata: Use integer return value for atapi_command_packet_set
2013-04-03 19:33:40 -04:00
bdi: allow block devices to say that they require stable page writes
2013-02-21 17:22:19 -08:00
watchdog: bcm47xx_wdt.c: use platform device
2013-03-01 12:47:16 +01:00
fs/exec.c: make bprm_mm_init() static
2013-02-26 02:46:13 -05:00
Input: bma150 - make some defines public and fix some comments
2013-02-15 18:52:45 -08:00
memory-hotplug: common APIs to support page tables hot-remove
2013-02-23 17:50:12 -08:00
Btrfs: move fs/btrfs/ioctl.h to include/uapi/linux/btrfs.h
2013-02-20 09:37:28 -05:00
bug.h, compiler.h: introduce compiletime_assert & BUILD_BUG_ON_MSG
2013-02-21 17:22:16 -08:00
clockevents: Add generic timer broadcast function
2013-01-31 22:15:36 +01:00
clocksource: always define CLOCKSOURCE_OF_DECLARE
2013-01-31 19:57:30 -08:00
coda: Restrict coda messages to the initial user namespace
2013-02-13 06:00:53 -08:00
mm: compaction: make __compact_pgdat() and compact_pgdat() return void
2013-02-23 17:50:10 -08:00
Fix breakage in MIPS siginfo handling
2013-03-19 19:15:52 +01:00
compiler-gcc.h: Add gcc-recommended GCC_VERSION macro
2013-02-21 17:22:15 -08:00
compiler-gcc{3,4}.h: Use GCC_VERSION macro
2013-02-21 17:22:15 -08:00
compiler{,-gcc4}.h, bug.h: Remove duplicate macros
2013-02-21 17:22:15 -08:00
bug.h, compiler.h: introduce compiletime_assert & BUILD_BUG_ON_MSG
2013-02-21 17:22:16 -08:00
sched: add wait_for_completion_io[_timeout]
2013-02-15 16:45:06 +01:00
Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux
2013-02-25 16:46:44 -08:00
context_tracking: Export context state for generic vtime
2013-01-26 17:37:17 +01:00
cpufreq: Remove unnecessary use of policy->shared_type
2013-02-02 01:29:32 +01:00
cpuidle: remove vestage definition of cpuidle_state_usage.driver_data
2013-02-11 18:49:51 -05:00
vfs: kill FS_REVAL_DOT by adding a d_weak_revalidate dentry op
2013-02-26 02:46:09 -05:00
Revert "lockdep: check that no locks held at freeze time"
2013-03-31 11:38:33 -07:00
PM / devfreq: Fix compiler warnings for CONFIG_PM_DEVFREQ unset
2013-04-02 01:28:41 +02:00
dm: add target num_write_bios fn
2013-03-01 22:45:49 +00:00
Driver core patches for 3.9-rc1
2013-02-21 12:05:51 -08:00
dm kcopyd: introduce configurable throttling
2013-03-01 22:45:49 +00:00
dma-buf: implement vmap refcounting in the interface logic
2013-02-27 15:13:36 +05:30
dmaengine: add dma_request_slave_channel_compat()
2013-02-27 22:37:34 +05:30
dmaengine: dw_dmac: move to generic DMA binding
2013-02-28 09:53:21 +05:30
eCryptfs: allow userspace messaging to be disabled
2013-03-03 23:59:59 -08:00
EDAC: Merge mci.mem_is_per_rank with mci.csbased
2013-03-16 06:32:30 +01:00
A few fixes to reduce places where pstore might hang
2013-02-21 09:38:18 -08:00
Merge branch 'for-3.9/core' of git://git.kernel.dk/linux-block
2013-02-28 12:52:24 -08:00
default SET_PERSONALITY() in linux/elf.h
2013-02-26 02:46:08 -05:00
include/linux/eventfd.h: fix incorrect filename is a comment
2013-02-27 19:10:23 -08:00
memory-hotplug: remove /sys/firmware/memmap/X sysfs
2013-02-23 17:50:12 -08:00
Revert "lockdep: check that no locks held at freeze time"
2013-03-31 11:38:33 -07:00
fs: Limit sys_mount to only request filesystem modules.
2013-03-03 19:36:31 -08:00
userns: Don't allow creation if the user is chrooted
2013-03-27 07:49:29 -07:00
new helper: file_inode(file)
2013-02-22 23:31:31 -05:00
ftrace: Do not call stub functions in control loop
2013-04-08 12:24:23 -04:00
tracing: Make a snapshot feature available from userspace
2013-01-30 11:02:06 -05:00
block: fix part_pack_uuid() build error
2013-02-26 03:10:52 +11:00
irq: Remove IRQ_EXIT_OFFSET workaround
2013-02-22 00:05:07 +01:00
kprobes: Make hash_64() as always inlined
2013-03-18 10:21:23 +01:00
hlist: drop the node parameter from iterators
2013-02-27 19:10:24 -08:00
video: Add generic HDMI infoframe helpers
2013-02-22 08:20:05 +01:00
Merge branches 'for-3.9/logitech', 'for-3.9/multitouch', 'for-3.9/ntrig', 'for-3.9/thingm' and 'for-3.9/upstream' into for-linus
2013-02-21 10:45:01 +01:00
mm: remove unused memclear_highpage_flush()
2013-02-23 17:50:16 -08:00
mm/rmap: rename anon_vma_unlock() => anon_vma_unlock_write()
2013-02-23 17:50:17 -08:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
2013-02-26 20:16:07 -08:00
hwmon: Retire SENSORS_LIMIT
2013-01-25 21:03:54 -08:00
Drivers: hv: vmbus: Consolidate all offer GUID definitions in hyperv.h
2013-01-25 11:17:31 -08:00
idr: deprecate idr_pre_get() and idr_get_new[_above]()
2013-03-13 15:21:47 -07:00
mac80211: advertise operating mode notification capability
2013-02-15 09:41:42 +01:00
macvlan: add multicast filter
2013-02-06 15:59:47 -05:00
hlist: drop the node parameter from iterators
2013-02-27 19:10:24 -08:00
vlan: adjust vlan_set_encap_proto() for its callers
2013-02-23 21:00:06 -05:00
ipv6: by default join ff01::1 and in case of forwarding ff01::2 and ff05:2
2013-02-11 14:16:26 -05:00
ipv4: introduce address lifetime
2013-01-29 13:59:57 -05:00
sched/rt: Add <linux/sched/rt.h> header to <linux/init_task.h>
2013-02-19 08:06:01 +01:00
iommu: Add DOMAIN_ATTR_WINDOWS domain attribute
2013-02-06 10:47:28 +01:00
ipmi: remove superfluous kernel/userspace explanation
2013-02-27 19:10:21 -08:00
ipv6 flowlabel: Convert np->ipv6_fl_list to RCU.
2013-01-30 22:41:13 -05:00
x86, msi: Use IRQ remapping specific setup_msi_irqs routine
2013-01-28 12:17:25 +01:00
irq_work.h: fix warning when CONFIG_IRQ_WORK=n
2013-03-22 16:41:19 -07:00
jbd2: use module parameters instead of debugfs for jbd_debug
2013-02-09 16:29:20 -05:00
printk: Provide a wake_up_klogd() off-case
2013-03-22 16:41:20 -07:00
cputime: Generic on-demand virtual cputime accounting
2013-01-27 19:23:27 +01:00
x86: Add Crash kernel low reservation
2013-01-29 19:32:58 -08:00
kgdb: remove #include <linux/serial_8250.h> from kgdb.h
2013-02-04 15:35:26 -08:00
ksm: remove old stable nodes more thoroughly
2013-02-23 17:50:19 -08:00
KVM: Allow cross page reads and writes from cached translations.
2013-04-07 13:05:35 +03:00
KVM: Allow cross page reads and writes from cached translations.
2013-04-07 13:05:35 +03:00
leds: leds-pwm: Convert to use devm_get_pwm
2013-02-01 17:47:04 -08:00
libata: Set max sector to 65535 for Slimtype DVD A DS8A8SH drive
2013-04-03 19:33:43 -04:00
list: Fix double fetch of pointer in hlist_entry_safe()
2013-03-14 13:18:30 -07:00
xen-blkfront: drop the use of llist_for_each_entry_safe
2013-02-19 15:17:08 -05:00
Merge branch 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
2013-02-22 19:25:09 -08:00
lib/lzo: Update LZO compression to current upstream version
2013-02-20 19:36:01 +01:00
ARM / highbank: add support for pl320 IPC
2013-02-02 00:01:15 +01:00
x86, ACPI, mm: Revert movablemem_map support
2013-03-02 09:34:39 -08:00
mm: refactor inactive_file_is_low() to use get_lru_size()
2013-02-23 17:50:20 -08:00
memory-hotplug: export the function try_offline_node()
2013-02-23 17:50:13 -08:00
drivers/net/phy/micrel_phy: Add support for new PHYs
2013-01-28 18:42:10 -05:00
mm: remove offlining arg to migrate_pages
2013-02-23 17:50:19 -08:00
Revert "mm: introduce VM_POPULATE flag to better deal with racy userspace programs"
2013-03-28 17:45:51 -07:00
mm: fold page->_last_nid into page->flags where possible
2013-02-23 17:50:17 -08:00
Revert "mm: introduce VM_POPULATE flag to better deal with racy userspace programs"
2013-03-28 17:45:51 -07:00
mm: fix wrong comments about anon_vma lock
2013-02-05 20:38:48 +11:00
mm: zone_end_pfn is too small
2013-03-22 16:41:20 -07:00
vfs: Add a mount flag to lock read only bind mounts
2013-03-27 07:50:04 -07:00
ARM: video: mxs: Fix mxsfb misconfiguring VDCTRL0
2013-03-19 11:07:58 +08:00
v4 GRE: Add TCP segmentation offload for GRE
2013-02-15 15:17:11 -05:00
net: count hw_addr syncs so that unsync works properly.
2013-04-05 00:18:46 -04:00
netpoll: Fix __netpoll_rcu_free so that it can hold the rtnl lock
2013-02-11 19:19:33 -05:00
nfsd: Handle kuids and kgids in the nfs4acl to posix_acl conversion
2013-02-13 06:16:06 -08:00
nfs: Convert idmap to use kuids and kgids
2013-02-13 06:15:29 -08:00
NFS client bugfixes for Linux 3.9
2013-03-02 16:46:07 -08:00
of: Create function for counting number of phandles in a property
2013-02-13 10:09:31 +00:00
gpio: Make of_count_named_gpios() use new of_count_phandle_with_args()
2013-02-13 10:11:53 +00:00
PM / OPP: fix condition for empty of_init_opp_table()
2013-02-23 00:30:08 +01:00
mm: fold page->_last_nid into page->flags where possible
2013-02-23 17:50:17 -08:00
s390/mm: implement software dirty bits
2013-02-14 15:55:23 +01:00
mm: remove MIGRATE_ISOLATE check in hotpath
2013-02-23 17:50:15 -08:00
mm: only enforce stable page writes if the backing device requires it
2013-02-21 17:22:19 -08:00
constify path_get/path_put and fs_struct.c stuff
2013-03-01 23:51:07 -05:00
PCI: Add PCI ROM helper for platform-provided ROM images
2013-03-26 17:19:41 -06:00
Merge branch 'linux_next' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-edac
2013-02-28 20:42:33 -08:00
percpu_counter.h: comment code for better readability
2013-02-05 11:54:06 +01:00
perf,x86: fix link failure for non-Intel configs
2013-03-17 15:59:15 -07:00
hlist: drop the node parameter from iterators
2013-02-27 19:10:24 -08:00
pm / runtime: introduce pm_runtime_set_memalloc_noio()
2013-02-23 17:50:16 -08:00
pm / runtime: introduce pm_runtime_set_memalloc_noio()
2013-02-23 17:50:16 -08:00
Driver core patches for 3.9-rc1
2013-02-21 12:05:51 -08:00
pps: Add pps_lookup_dev() function
2013-02-13 09:40:35 -08:00
spinlocks and preemption points need to be at least compiler barriers
2013-04-09 10:48:33 -07:00
printk: Provide a wake_up_klogd() off-case
2013-03-22 16:41:20 -07:00
procfs: add proc_remove_subtree()
2013-04-09 14:09:17 -04:00
pwm: Changes for v3.9-rc1
2013-02-26 09:34:29 -08:00
spi/pxa2xx: add support for Intel Low Power Subsystem SPI
2013-02-08 13:14:40 +00:00
hlist: drop the node parameter from iterators
2013-02-27 19:10:24 -08:00
Merge branches 'doctorture.2013.01.29a', 'fixes.2013.01.26a', 'tagcb.2013.01.24a' and 'tiny.2013.01.29b' into HEAD
2013-01-28 22:25:21 -08:00
Merge remote-tracking branch 'regmap/topic/no-bus' into regmap-next
2013-02-14 17:11:09 +00:00
include/linux/res_counter.h needs errno.h
2013-03-13 15:21:44 -07:00
ring-buffer: Add stats field for amount read from trace ring buffer
2013-01-30 11:01:53 -05:00
mm/rmap: rename anon_vma_unlock() => anon_vma_unlock_write()
2013-02-23 17:50:17 -08:00
Driver core patches for 3.9-rc1
2013-02-21 12:05:51 -08:00
lib/scatterlist: use page iterator in the mapping iterator
2013-02-27 19:10:10 -08:00
kthread: Prevent unpark race which puts threads on the wrong cpu
2013-04-12 14:18:43 +02:00
lsm: add the missing documentation for the security_skb_owned_by() hook
2013-04-10 15:40:39 -04:00
seqlock: Use seqcount infrastructure
2013-02-19 08:43:34 +01:00
pps: Move timestamp read into PPS code proper
2013-02-13 10:13:58 -08:00
Fix breakage in MIPS siginfo handling
2013-03-19 19:15:52 +01:00
netfilter: don't reset nf_trace in nf_reset()
2013-04-05 15:38:10 -04:00
smp: make smp_call_function_many() use logic similar to smp_call_function_single()
2013-02-21 17:22:20 -08:00
stop_machine: Mark per cpu stopper enabled early
2013-02-26 22:25:17 +01:00
VSOCK: Introduce VM Sockets
2013-02-10 19:41:08 -05:00
spinlocks and preemption points need to be at least compiler barriers
2013-04-09 10:48:33 -07:00
srcu: Remove checks preventing idle CPUs from calling srcu_read_lock()
2013-02-07 15:15:00 -08:00
rtc: stmp3xxx: add wdt-accessor function
2013-03-01 12:40:36 +01:00
pps: Fix build breakage from decoupling pps from tty
2013-02-13 12:15:50 -08:00
PM: Introduce suspend state PM_SUSPEND_FREEZE
2013-02-09 22:30:44 +01:00
vmscan: change type of vm_total_pages to unsigned long
2013-02-23 17:50:22 -08:00
x86: Don't panic if can not alloc buffer for swiotlb
2013-01-29 19:36:53 -08:00
burying unused conditionals
2013-02-14 09:21:15 -05:00
sysfs: Functions for adding/removing symlinks to/from attribute groups
2013-01-25 21:51:13 +01:00
tcp: adding a per-socket timestamp offset
2013-02-13 13:22:15 -05:00
ARM: tegra: add function to read chipid
2013-01-28 11:19:06 -07:00
thermal: shorten too long mcast group name
2013-03-20 17:56:58 -04:00
Merge branch 'nohz/printk-v8' into irq/core
2013-02-05 00:48:46 +01:00
arm-soc: cleanups
2013-02-21 14:58:40 -08:00
cputime: Use accessors to read task cputime stats
2013-01-27 19:23:31 +01:00
tty: Document required behavior of tty driver close()
2013-02-04 15:40:28 -08:00
pps: Move timestamp read into PPS code proper
2013-02-13 10:13:58 -08:00
udp: add encap_destroy callback
2013-03-20 12:10:38 -04:00
uprobes: Introduce uprobe_apply()
2013-02-08 18:28:04 +01:00
USB: add usb_hcd_{start,end}_port_resume
2013-01-25 15:06:01 -08:00
userns: Restrict when proc and sysfs can be mounted
2013-03-27 07:50:08 -07:00
mfd: vexpress: Add pseudo-GPIO based LEDs
2013-02-14 00:22:58 +01:00
virtio: Add module driver macro for virtio drivers.
2013-02-13 17:00:32 +10:30
mm: don't wait on congested zones in balance_pgdat()
2013-02-23 17:50:15 -08:00
mm: numa: handle side-effects in count_vm_numa_events() for !CONFIG_NUMA_BALANCING
2013-02-23 17:50:16 -08:00
Merge branch 'console-fixes' into drm-next
2013-02-08 12:13:43 +10:00
cputime: Safely read cputime of full dynticks CPUs
2013-01-27 20:35:47 +01:00
wanrouter: delete now orphaned header content, files/drivers
2013-01-31 19:56:35 -05:00
watchdog: core: dt: add support for the timeout-sec dt property
2013-03-01 12:48:36 +01:00
wlcore: move wl12xx_platform_data up and make it truly optional
2013-02-08 10:05:02 +02:00
workqueue: rename cpu_workqueue to pool_workqueue
2013-02-13 19:29:12 -08:00
