Tejun Heo
2a81490811
writeback: implement foreign cgroup inode detection ...
As concurrent write sharing of an inode is expected to be very rare
and memcg only tracks page ownership on first-use basis severely
confining the usefulness of such sharing, cgroup writeback tracks
ownership per-inode. While the support for concurrent write sharing
of an inode is deemed unnecessary, an inode being written to by
different cgroups at different points in time is a lot more common,
and, more importantly, charging only by first-use can too readily lead
to grossly incorrect behaviors (single foreign page can lead to
gigabytes of writeback to be incorrectly attributed).
To resolve this issue, cgroup writeback detects the majority dirtier
of an inode and will transfer the ownership to it. To avoid
unnnecessary oscillation, the detection mechanism keeps track of
history and gives out the switch verdict only if the foreign usage
pattern is stable over a certain amount of time and/or writeback
attempts.
The detection mechanism has fairly low space and computation overhead.
It adds 8 bytes to struct inode (one int and two u16's) and minimal
amount of calculation per IO. The detection mechanism converges to
the correct answer usually in several seconds of IO time when there's
a clear majority dirtier. Even when there isn't, it can reach an
acceptable answer fairly quickly under most circumstances.
Please see wb_detach_inode() for more details.
This patch only implements detection. Following patches will
implement actual switching.
v2: wbc_account_io() now checks whether the wbc is associated with a
wb before dereferencing it. This can happen when pageout() is
writing pages directly without going through the usual writeback
path. As pageout() path is single-threaded, we don't want it to
be blocked behind a slow cgroup and ultimately want it to delegate
actual writing to the usual writeback path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Jan Kara <jack@suse.cz>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Greg Thelen <gthelen@google.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
2015-06-02 08:40:20 -06:00
libceph: announce support for straw2 buckets
2015-04-22 18:33:48 +03:00
ARM: SoC late changes for v4.1
2015-04-22 09:24:55 -07:00
crush: straw2 bucket type with an efficient 64-bit crush_ln()
2015-04-22 18:33:43 +03:00
Merge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma
2015-04-24 09:49:37 -07:00
fsl /bestcomm
iio: add support for hardware fifo
2015-03-29 16:17:10 +01:00
Initial ACPI support for arm64:
2015-04-24 08:23:45 -07:00
platform/chrome: Updates for v4.1
2015-04-26 13:36:02 -07:00
net/mlx4_core: Set initial admin GUIDs for VFs
2015-04-15 15:51:50 -04:00
net/mlx5_core: Extend struct mlx5_interface to support multiple protocols
2015-04-02 16:33:43 -04:00
Merge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma
2015-04-24 09:49:37 -07:00
mtd: spi-nor: factor out replace-able flash_{lock,unlock}
2015-03-27 10:37:38 -07:00
netlink: implement nla_put_in_addr and nla_put_in6_addr
2015-03-31 13:58:35 -04:00
netfilter: Pass nf_hook_state through arpt_do_table().
2015-04-04 13:26:52 -04:00
uapi: ebtables: don't include linux/if.h
2015-04-13 20:08:38 +02:00
netfilter: Pass nf_hook_state through ipt_do_table().
2015-04-04 12:47:04 -04:00
netfilter: Pass nf_hook_state through ip6t_do_table().
2015-04-04 12:52:06 -04:00
Merge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma
2015-04-24 09:49:37 -07:00
power: max17042_battery: Use reg type instead of chip type
2015-04-06 17:46:27 +02:00
md/raid6 algorithms: delta syndrome functions
2015-04-22 08:00:41 +10:00
Merge remote-tracking branches 'regulator/topic/mode', 'regulator/topic/notifier', 'regulator/topic/palmas', 'regulator/topic/qcom' and 'regulator/topic/stw481x' into regulator-next
2015-04-10 19:16:03 +01:00
soc /ti
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
2015-04-15 09:00:47 -07:00
NFS: NFSoRDMA Client Changes
2015-04-23 15:16:37 -04:00
Merge 4.0-rc7 into usb-next
2015-04-07 11:17:34 +02:00
include/linux: remove empty conditionals
2015-04-15 16:35:22 -07:00
Initial ACPI support for arm64:
2015-04-24 08:23:45 -07:00
irqchip: Add GICv2 specific ACPI boot support
2015-03-26 15:13:07 +00:00
fs: move struct kiocb to fs.h
2015-03-25 20:28:11 -04:00
arm-cci: Split the code for PMU vs driver support
2015-03-27 13:44:43 +00:00
md/raid5: activate raid6 rmw feature
2015-04-22 08:00:42 +10:00
libata: Implement support for sense data reporting
2015-03-27 11:59:22 -04:00
writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb()
2015-06-02 08:39:08 -06:00
writeback: make writeback_control track the inode being written back
2015-06-02 08:39:48 -06:00
MIPS: BCM47xx: Move NVRAM header to the include/linux/.
2015-04-01 17:22:00 +02:00
blkcg: implement bio_associate_blkcg()
2015-06-02 08:33:34 -06:00
lib/bitmap.c: bitmap_[empty,full]: remove code duplication
2015-04-17 09:03:56 -04:00
lib: find_*_bit reimplementation
2015-04-17 09:03:53 -04:00
writeback, blkcg: associate each blkcg_gq with the corresponding bdi_writeback_congested
2015-06-02 08:33:35 -06:00
blk-mq: Shared tag enhancements
2015-06-01 14:35:56 -06:00
block, dm: don't copy bios for request clones
2015-05-22 08:58:57 -06:00
writeback, blkcg: restructure blk_{set|clear}_queue_congested()
2015-06-02 08:33:35 -06:00
bpf: Make internal bpf API independent of CONFIG_BPF_SYSCALL #ifdefs
2015-04-02 13:25:49 +02:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
2015-04-14 15:44:14 -04:00
kernel: conditionally support non-root users, groups and capabilities
2015-04-15 16:35:22 -07:00
cgroup, block: implement task_get_css() and use it in bio_associate_current()
2015-06-02 08:33:34 -06:00
cleancache: remove limit on the number of cleancache enabled filesystems
2015-04-14 16:49:03 -07:00
clk: clk-gpio-gate: Fix active low
2015-04-10 17:45:30 -07:00
clockevents: Cleanup dead cpu explicitely
2015-04-03 08:44:37 +02:00
Initial ACPI support for arm64:
2015-04-24 08:23:45 -07:00
mm: cma: constify and use correct signness in mm/cma.c
2015-04-14 16:49:04 -07:00
mm: allow compaction of unevictable pages
2015-04-15 16:35:17 -07:00
lib: make memzero_explicit more robust against dead store elimination
2015-05-04 17:49:51 +08:00
lib: make memzero_explicit more robust against dead store elimination
2015-05-04 17:49:51 +08:00
lib: make memzero_explicit more robust against dead store elimination
2015-05-04 17:49:51 +08:00
console: Add extensible console matching
2015-03-26 16:16:20 +01:00
cpu: Provide smpboot_thread_init() on !CONFIG_SMP kernels as well
2015-04-13 10:19:30 +02:00
cpuidle: remove state_count field from struct cpuidle_device
2015-04-03 13:15:50 +02:00
cpumask: resurrect CPU_MASK_CPU0
2015-04-16 12:33:51 +09:30
kernel: conditionally support non-root users, groups and capabilities
2015-04-15 16:35:22 -07:00
crypto: api - prevent helper ciphers from being used
2015-03-31 21:21:03 +08:00
VFS: Impose ordering on accesses of d_inode and d_flags
2015-04-15 15:05:28 -04:00
ipv4: dccp: handle ICMP messages on DCCP_NEW_SYN_RECV request sockets
2015-03-23 16:52:26 -04:00
PM / devfreq: event: Add const keyword for devfreq_event_ops structure
2015-03-30 17:09:41 +09:00
dm: remove unnecessary wrapper around blk_lld_busy
2015-03-31 12:03:49 -04:00
Power management and ACPI updates for v4.1-rc1
2015-04-14 20:21:54 -07:00
dma-buf: cleanup dma_buf_export() to make it easily extensible
2015-04-21 14:47:16 +05:30
dmaengine: vdma: Fix compilation warnings
2015-04-17 23:18:53 +05:30
x86/efi: Add a "debug" option to the efi= cmdline
2015-04-01 12:46:22 +01:00
mm: fold arch_randomize_brk into ARCH_HAS_ELF_RANDOMIZE
2015-04-14 16:49:05 -07:00
f2fs: add F2FS_INLINE_DOTS to recover missing dot dentries
2015-04-10 15:08:57 -07:00
fs: Add support FALLOC_FL_INSERT_RANGE for fallocate
2015-03-25 15:07:05 +11:00
filter: introduce SKF_AD_VLAN_TPID BPF extension
2015-03-24 15:25:15 -04:00
writeback: implement foreign cgroup inode detection
2015-06-02 08:40:20 -06:00
fs_pin: Allow for the possibility that m_list or s_list go unused.
2015-04-09 11:39:55 -05:00
Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
2015-04-14 14:37:47 -07:00
device property: Introduce firmware node type for platform data
2015-04-03 23:23:51 +02:00
mm: clarify __GFP_NOFAIL deprecation status
2015-04-14 16:49:04 -07:00
Merge branch 'for-4.1/sensor-hub' into for-linus
2015-04-13 23:43:34 +02:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
2015-04-14 09:25:26 -07:00
gpu: host1x: Export host1x_syncpt_read()
2015-04-02 18:46:20 +02:00
mm: hugetlb: cleanup using paeg_huge_active()
2015-04-15 16:35:19 -07:00
hv: remove the per-channel workqueue
2015-04-03 16:18:02 +02:00
Power management and ACPI updates for v4.1-rc1
2015-04-14 20:21:54 -07:00
block: move PM request support to IDE
2015-05-05 13:40:42 -06:00
if_link: Add an additional parameter to ifla_vf_info for RSS querying
2015-04-10 21:57:22 -07:00
net: Move check for multiple vlans to drivers
2015-03-29 13:33:22 -07:00
iommu/vt-d: Add new extended capabilities from v2.3 VT-d specification
2015-03-25 15:43:39 +00:00
genirq: Allow the irqchip state of an IRQ to be save/restored
2015-04-08 23:28:28 +02:00
Char/Misc driver patches for 4.1-rc1
2015-04-21 09:42:58 -07:00
Break up monolithic iommu table/lock into finer graularity pools and lock
2015-04-18 12:32:59 -07:00
iommu: Remove domain_init and domain_free iommu_ops
2015-03-31 15:32:16 +02:00
kernel/resource.c: remove deprecated __check_region() and friends
2015-04-15 16:35:22 -07:00
ipv6: introduce secret_stable to ipv6_devconf
2015-03-23 22:12:08 -04:00
irqchip core change for v4.1 (round 3)
2015-04-11 11:17:28 +02:00
jhash: Update jhash_[321]words functions to use correct initval
2015-04-03 12:52:29 -04:00
jump_label: Allow jump labels to be used in assembly
2015-04-09 09:40:24 +02:00
memory: jz4780-nemc: driver for the NEMC on JZ4780 SoCs
2015-03-26 23:51:36 +01:00
mm/mempool.c: kasan: poison mempool elements
2015-04-15 16:35:20 -07:00
Merge branch 'drm-next-merged' of git://people.freedesktop.org/~airlied/linux into v4l_for_linus
2015-04-21 09:44:55 -03:00
kernel.h: implement DIV_ROUND_CLOSEST_ULL
2015-04-17 09:03:55 -04:00
kexec: allocate the kexec control page with KEXEC_CONTROL_MEMORY_GFP
2015-04-23 16:52:01 +02:00
mm: consolidate all page-flags helpers in <linux/page-flags.h>
2015-04-15 16:35:17 -07:00
Merge branch 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
2015-04-14 13:58:48 -07:00
block: fix blk_stack_limits() regression due to lcm() change
2015-03-31 09:45:50 -06:00
leds: flash: Fix the size of sysfs_groups array
2015-03-30 11:55:48 -07:00
lguest: suppress interrupts for single insn, not range.
2015-03-24 11:52:08 +10:30
Merge branch 'for-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata
2015-04-13 16:42:16 -07:00
memtest: use phys_addr_t for physical addresses
2015-04-14 16:49:06 -07:00
writeback: implement memcg writeback domain based throttling
2015-06-02 08:38:13 -06:00
mm, hotplug: fix concurrent memory hot-add deadlock
2015-04-14 16:49:00 -07:00
mm, mempool: disallow mempools based on slab caches with constructors
2015-04-15 16:35:18 -07:00
mm: numa: remove migrate_ratelimited
2015-04-14 16:49:05 -07:00
misc: Add attribute groups
2015-03-25 11:44:52 +01:00
memcg: add per cgroup dirty page accounting
2015-06-02 08:33:33 -06:00
mm: rcu-protected get_mm_exe_file()
2015-04-17 09:04:07 -04:00
mm: refactor zone_movable_is_highmem()
2015-04-15 16:35:16 -07:00
MIPS: Add CDMM bus support
2015-03-31 12:04:12 +02:00
Some clean ups and small fixes, but the biggest change is the addition
2015-04-14 10:49:03 -07:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace
2015-04-18 11:20:31 -04:00
get rid of the size argument of sock_sendmsg()
2015-04-11 15:27:37 -04:00
bridge/nl: remove wrong use of NLM_F_MULTI
2015-04-29 14:59:16 -04:00
netfilter: Pass socket pointer down through okfn().
2015-04-07 15:25:55 -04:00
netfilter: bridge: fix NULL deref in physin/out ifindex helpers
2015-04-24 20:51:40 +02:00
netlink: Fix portid type in netlink_notify
2015-04-13 16:35:16 -04:00
uapi: Remove kernel internal declaration
2015-04-21 16:16:04 -04:00
NFS client updates for Linux 4.1
2015-04-26 17:33:59 -07:00
NFS: Don't zap caches on fallocate()
2015-04-23 14:36:28 -04:00
watchdog: introduce the hardlockup_detector_disable() function
2015-04-14 16:48:59 -07:00
NVMe: Meta data handling through submit io ioctl
2015-04-07 19:11:06 -06:00
Devicetree updates for 4.1:
2015-04-24 08:46:18 -07:00
of/fdt: Add endianness helper function for early init code
2015-04-14 19:35:44 -05:00
of: Explicitly include linux/types.h in of_graph.h
2015-03-26 12:14:56 -05:00
of: Add dummy of_irq_to_resource_table() for IRQ_OF=n
2015-04-14 19:23:24 -05:00
memcg: print cgroup information when system panics due to panic_on_oom
2015-04-14 16:49:05 -07:00
mm: hugetlb: cleanup using paeg_huge_active()
2015-04-15 16:35:19 -07:00
memcg: add per cgroup dirty page accounting
2015-06-02 08:33:33 -06:00
PCI/ACPI: Optimize device state transition delays
2015-04-08 16:25:25 -05:00
PCI: Don't clear ASPM bits when the FADT declares it's unsupported
2015-04-09 14:20:11 -05:00
powerpc updates for 4.1
2015-04-16 13:53:32 -05:00
perf: Add ITRACE_START record to indicate that tracing has started
2015-04-02 17:14:17 +02:00
Remove rest of exec domains.
2015-04-12 21:03:31 +02:00
add fixed_phy_update_state() - update state of fixed_phy
2015-04-03 15:08:20 -04:00
printk: comment pr_cont() stating it is only to continue a line
2015-04-15 16:35:23 -07:00
device property: Introduce firmware node type for platform data
2015-04-03 23:23:51 +02:00
ptp: remove 32 bit get/set methods.
2015-03-31 12:01:19 -04:00
kernel/reboot.c: add orderly_reboot for graceful reboot
2015-04-15 16:35:23 -07:00
rhashtable: don't attempt to grow when at max_size
2015-04-24 11:38:12 -04:00
mm: uninline and cleanup page-mapping related helpers
2015-04-15 16:35:19 -07:00
drivers/rtc: Provide y2038 safe rtc_class_ops.set_mmss() replacement
2015-04-03 08:18:24 +02:00
bridge/nl: remove wrong use of NLM_F_MULTI
2015-04-29 14:59:16 -04:00
x86: pvclock: Really remove the sched notifier for cross-cpu migrations
2015-04-27 15:49:30 +02:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
2015-04-15 09:00:47 -07:00
serial: 8250: Decouple RSA probe
2015-03-26 18:18:20 +01:00
earlycon: Fix __earlycon_table stride
2015-04-10 14:39:53 +02:00
net: fix crash in build_skb()
2015-04-25 15:49:49 -04:00
mm, slab: correct config option in comment
2015-04-14 16:48:59 -07:00
smp: don't use 16-bit words for atomic accesses
2015-04-20 09:08:49 -07:00
new helper: msg_data_left()
2015-04-11 15:53:35 -04:00
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial
2015-04-14 09:50:27 -07:00
stmmac: Read tx-fifo-depth and rx-fifo-depth from the devicetree
2015-04-16 13:58:42 -04:00
SCSI misc on 20150416
2015-04-16 19:02:04 -04:00
suspend: simplify block I/O handling
2015-05-19 09:19:59 -06:00
kernel/sysctl.c: threads-max observe limits
2015-04-17 09:04:07 -04:00
sysfs: Document struct attribute_group
2015-03-25 13:28:58 +01:00
tcp: RFC7413 option support for Fast Open client
2015-04-07 18:36:39 -04:00
clockevents: Cleanup dead cpu explicitely
2015-04-03 08:44:37 +02:00
time: Add timerkeeper::tkr_raw
2015-03-27 09:45:07 +01:00
time, drivers/rtc: Don't bother with rtc_resume() for the nonstop clocksource
2015-04-03 08:18:34 +02:00
tracing: Add TRACE_DEFINE_ENUM() macro to map enums to their values
2015-04-08 09:39:56 -04:00
tty: Re-add external interface for tty_set_termios()
2015-04-28 14:26:20 +02:00
include/linux: remove empty conditionals
2015-04-15 16:35:22 -07:00
kernel: conditionally support non-root users, groups and capabilities
2015-04-15 16:35:22 -07:00
new helper: iov_iter_rw()
2015-04-11 22:29:44 -04:00
usb: define a generic USB_RESUME_TIMEOUT macro
2015-04-07 12:58:35 -05:00
uas: Add US_FL_MAX_SECTORS_240 flag
2015-04-28 12:48:57 +02:00
util_macros.h: add find_closest() macro
2015-04-17 09:03:54 -04:00
VFIO updates for v4.1
2015-04-14 18:06:47 -07:00
vgaarb: Stub vga_set_legacy_decoding()
2015-04-07 11:14:38 -06:00
virtio: drop virtio_device_is_legacy_only
2015-04-15 12:41:14 +09:30
virtio: drop a useless config read
2015-04-01 14:37:14 +10:30
virtio_ring: Update weak barriers to use dma_wmb/rmb
2015-04-13 21:04:16 +09:30
watchdog: Add watchdog enable/disable all functions
2015-04-02 17:33:15 +02:00
wlcore: remove wl12xx_platform_data
2015-03-24 09:48:14 -07:00
writeback: implement foreign cgroup inode detection
2015-06-02 08:40:20 -06:00
zsmalloc: support compaction
2015-04-15 16:35:20 -07:00
