49eea524bebea0d2b7dfa1c709a6694de808eb8a
59 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Pavankumar Kondeti
|
0d2621bf90 |
trace/irq: Add trace point to track IRQs disable callers
Add trace point to track IRQs disable callers to isolate issues unrelated to scheduler and improve debug turn around time. Change-Id: Ib1ef45d8bed1fc0e128b5ab2051f0c30e8c50ee7 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org> Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org> |
||
|
Pavankumar Kondeti
|
97f08d47f4 |
sched: Add trace point to track preemption disable callers
Add trace point to track preemption disable callers to isolate issues unrelated to scheduler and improve debug turn around time. Change-Id: If9303b7165167e8f79cd339929daf4afc31a61c4 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org> Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org> |
||
|
Pavankumar Kondeti
|
93b3f0f674 |
Merge remote-tracking branch 'dev/msm-4.9-sched' into msm-4.9
* origin/dev/msm-4.9-sched: ARM: dts: msm: Update silver CPUs's idle cost data for SDM845v2 sched/fair: prevent possible infinite loop in sched_group_energy sched: Update WALT stats also in boosted_cpu_util() sched/fair: Add provision to control the spreading on SMP sched/fair: Add EAS_USE_NEED_IDLE feature sched/fair: Turn off FBT_STRICT_ORDER feature sched/fair: fix incorrect CPU selection for non latency sensitive tasks sched: Add task placement snapshot sched/fair: select the most energy-efficient CPU candidate on wake-up sched/fair: fix array out of bounds access in select_energy_cpu_idx() sched/fair: use min capacity when evaluating active cpus sched/fair: use min capacity when evaluating idle backup cpus sched/fair: use min capacity when evaluating placement energy costs sched/fair: introduce minimum capacity capping sched feature arm/topology: link arch_scale_min_freq_capacity to cpufreq arm64/topology: link arch_scale_min_freq_capacity to cpufreq sched: add arch_scale_min_freq_capacity to track minimum capacity caps cpufreq: add scaled minimum capacity tracking for policy changes arm64: enable max frequency capping arm: enable max frequency capping cpufreq: implement max frequency capping sched/fair: introduce an arch scaling function for max frequency capping cpufreq: remove max frequency capping from scale_freq_capacity() Revert "ANDROID: cpufreq: Max freq invariant scheduler load-tracking and cpu capacity support" Revert "ANDROID: arm: Enable max freq invariant scheduler load-tracking and capacity support" Revert "ANDROID: arm64: Enable max freq invariant scheduler load-tracking and capacity support" sched/fair: reduce rounding errors in energy computations sched/fair: re-factor energy_diff to use a single (extensible) energy_env sched/fair: cleanup select_energy_cpu_brute to be more consistent sched/fair: remove capacity tracking from energy_diff sched/fair: remove energy_diff tracepoint in preparation to re-factoring sched/fair: use *p to reference task_structs sched: EAS: Fix the calculation of group util in group_idle_state() cpufreq: Drop schedfreq governor sched: Sync EAS codebase to android-4.9 sched: Move core_ctl callback from tick to WALT IRQ work sched: fair: Always use energy aware wakeups sched: Introduce new workload differentiation sched: Introduce a different version of freq aggregation sched: report group load to the cpufreq sched: Start reporting top task load to cpufreq sched: Introduce scheduler boost related placement changes sched: Port boost setting mechanisms to EAS sched: integrate core_ctl with EAS sched: EAS: add infrastructure for core_ctl sched: EAS: add core isolation support sched: Introduce an irq_work to report WALT load sched: introduce small wakee task on waker sched: fair: Ignore energy-diff calculations for colocated tasks sched: change default group upmigrate and downmigrate values sched: EAS: colocate related threads sched: cpufreq: Use per_cpu_ptr instead of this_cpu_ptr when reporting load sched: cpufreq: Use sched_clock instead of rq_clock when updating schedutil sched: cpufreq: Update cpufreq once in a WALT window ARM: dts: msm: Add an energy model for the SDM845 CPUs Conflicts: kernel/sched/fair.c Change-Id: Ieaeecb28e57955db3b13d6d9c1d81b204caf0fcf Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Viresh Kumar
|
9255013705 |
cpufreq: Drop schedfreq governor
We all should be using (and improving) the schedutil governor now. Get
rid of the non-upstream governor.
Tested on Hikey.
Change-Id: I2104558b03118b0a9c5f099c23c42cd9a6c2a963
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Git-commit:
|
||
|
Pavankumar Kondeti
|
4e13d11267 |
sched: improve the scheduler
This change is for general scheduler improvement. Change-Id: Ice980dde340bff8362b4f2adc679423d8f54e8e4 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Imran Khan
|
a9788f4c9d |
hung task: check specific tasks for long uninterruptible sleep state
khungtask by default monitors all tasks for long unterruptible sleep. This change introduces a sysctl option, /proc/sys/kernel/ hung_task_selective_monitoring, to enable monitoring selected tasks. If this sysctl option is enabled then only the tasks with /proc/$PID/hang_detection_enabled set are to be monitored, otherwise all tasks are monitored as default case. Some tasks may intentionally moves to uninterruptable sleep state, which shouldn't leads to khungtask panics, as those are recoverable hungs. So to avoid false hung reports, add an option to select tasks to be monitored and report/panic them only. By default, enable the feature always to monitor selected tasks. Change-Id: I48cd8cfe73dbe2b577541fe9607190eac5556bb2 Signed-off-by: Imran Khan <kimran@codeaurora.org> Signed-off-by: Lingutla Chandrasekhar <clingutla@codeaurora.org> |
||
|
Pavankumar Kondeti
|
3066bc4e33 |
sched: move task_util() to sched.h
The same task_util() inline function is defined in fair.c and rt.c. Move it to a common header file. While at it, make sure that this function works for 32 bit also. Previously, util = ((u64)demand)<<10/sched_ravg_window This division of two 64 bit quantities gives a linker error on 32 bit system. The same result can be achieved without type casting the demand to 64 bit and without losing precision by making it util = demand/(sched_ravg_window>>10). Change-Id: Iffc2b6b484fbcdac8d6f9739f96798a8d8e035c5 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Pavankumar Kondeti
|
d4127509d1 |
sched: remove the unused HMP specific code
Remove the unused HMP specific code, mostly wrappers used in load balancer and task placement. The CFS_BANDWIDTH feature is supported for HMP but not for WALT. So remove any remaining references to this feature. It will be reintroduced for WALT in the subsequent patches. Change-Id: Ic166b5e0f2abc0060c5030784c4ff22efc0e3885 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Puja Gupta
|
487dec641c |
sched: Remove HMP related code and config
Since HMP is no longer used, remove related code and config options. Change-Id: I4a30bbdf34c63c9c400e5b826a4758eac26ce607 Signed-off-by: Puja Gupta <pujag@codeaurora.org> Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Joonwoo Park
|
b02fc008c8 |
sched: introduce sched_{up,down}migrate knobs
Introduce knobs sched_upmigrate and sched_downmigrate to control thresholds to up and down migrate tasks between clusters at a runtime for ease of power and performance tuning. Change-Id: I17a464388b404b61d9a9dbe1beee4883c4db8e3f Signed-off-by: Joonwoo Park <joonwoop@codeaurora.org> |
||
|
Joonwoo Park
|
6f18847ea1 |
sched: WALT: introduce sched_group_{up,down}_migrate knobs
Introduce sched_group_upmigrate and sched_group_downmigrate to control these tunables at a runtime. Change-Id: I8d88b0ca15dc1daec99b03de116192acc2660ba7 Signed-off-by: Joonwoo Park <joonwoop@codeaurora.org> |
||
|
Syed Rameez Mustafa
|
20acfe73be |
sched: improve the scheduler
This change is for general scheduler improvement. CRs-Fixed: 2040904 Change-Id: I6efa77cd260228a29a0105146fcdecc4b5ee176e Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org> |
||
|
Joonwoo Park
|
f7d6cd4596 |
sched/walt: factor out WALT from HMP scheduler
At present, hmp.c contains HMP task placement logic as well as WALT. Factor out WALT logic into a separate file walt.c so EAS can use WALT as CPU load tracking and frequency guidance algorithm. Also set sched_io_is_busy = 1 by default. Change-Id: I96c3fd7d637b9bfa283083a1a6ff6d5dd261e0fb Signed-off-by: Joonwoo Park <joonwoop@codeaurora.org> |
||
|
Channagoud Kadabi
|
8810e5fa00 |
Merge remote-tracking branch 'origin/tmp-dcb6110' into 4.8
* origin/tmp-dcb6110:
ANDROID: goldfish_sync: 32 max cmds to save stack
ANDROID: sched/walt: use div_u64 instead of do_div
Linux 4.9.9
drm/i915/execlists: Reset RING registers upon resume
fs: break out of iomap_file_buffered_write on fatal signals
iw_cxgb4: set correct FetchBurstMax for QPs
x86/irq: Make irq activate operations symmetric
irqdomain: Avoid activating interrupts more than once
iio: health: max30100: fixed parenthesis around FIFO count check
iio: dht11: Use usleep_range instead of msleep for start signal
iio: health: afe4403: retrieve a valid iio_dev in suspend/resume
iio: health: afe4404: retrieve a valid iio_dev in suspend/resume
iio: adc: palmas_gpadc: retrieve a valid iio_dev in suspend/resume
staging: greybus: timesync: validate platform state callback
USB: serial: option: add device ID for HP lt2523 (Novatel E371)
usb: gadget: f_fs: Assorted buffer overflow checks.
usb: musb: Fix host mode error -71 regression
USB: Add quirk for WORLDE easykey.25 MIDI keyboard
USB: serial: pl2303: add ATEN device ID
USB: serial: qcserial: add Dell DW5570 QDL
KVM: x86: do not save guest-unsupported XSAVE state
dmaengine: cppi41: Fix oops in cppi41_runtime_resume
dmaengine: cppi41: Fix runtime PM timeouts with USB mass storage
perf/x86/intel/uncore: Clean up hotplug conversion fallout
HID: wacom: Fix poor prox handling in 'wacom_pl_irq'
HID: hid-lg: Fix immediate disconnection of Logitech Rumblepad 2
HID: usbhid: Quirk a AMI virtual mouse and keyboard with ALWAYS_POLL
iwlwifi: mvm: avoid crash on restart w/o reserved queues
iwlwifi: fix double hyphen in MODULE_FIRMWARE for 8000
pinctrl: intel: merrifield: Add missed check in mrfld_config_set()
pinctrl: baytrail: Debounce register is one per community
Revert "vring: Force use of DMA API for ARM-based systems with legacy devices"
Revert "bcma: init serial console directly from ChipCommon code"
percpu-refcount: fix reference leak during percpu-atomic transition
regulator: axp20x: AXP806: Fix dcdcb being set instead of dcdce
vhost: fix initialization for vq->is_le
mmc: sdhci: Ignore unexpected CARD_INT interrupts
cgroup: don't online subsystems before cgroup_name/path() are operational
can: bcm: fix hrtimer/tasklet termination in bcm op removal
tracing: Fix hwlat kthread migration
mm, fs: check for fatal signals in do_generic_file_read()
base/memory, hotplug: fix a kernel oops in show_valid_zones()
mm/memory_hotplug.c: check start_pfn in test_pages_in_a_zone()
cifs: initialize file_info_lock
zswap: disable changing params if init fails
svcrpc: fix oops in absence of krb5 module
NFSD: Fix a null reference case in find_or_create_lock_stateid()
powerpc/mm: Use the correct pointer when setting a 2MB pte
powerpc: Fix build failure with clang due to BUILD_BUG_ON()
powerpc: Add missing error check to prom_find_boot_cpu()
powerpc/eeh: Fix wrong flag passed to eeh_unfreeze_pe()
libata: Fix ATA request sense
libata: apply MAX_SEC_1024 to all CX1-JB*-HP devices
ata: sata_mv:- Handle return value of devm_ioremap.
perf/core: Fix PERF_RECORD_MMAP2 prot/flags for anonymous memory
perf/core: Fix use-after-free bug
crypto: arm64/aes-blk - honour iv_out requirement in CBC and CTR modes
crypto: api - Clear CRYPTO_ALG_DEAD bit before registering an alg
drm/nouveau/nv1a,nv1f/disp: fix memory clock rate retrieval
drm/nouveau/disp/gt215: Fix HDA ELD handling (thus, HDMI audio) on gt215
drm/amdgpu/si: fix crash on headless asics
pinctrl: baytrail: Add missing spinlock usage in byt_gpio_irq_handler
HID: cp2112: fix gpio-callback error handling
HID: cp2112: fix sleep-while-atomic
xtensa: fix noMMU build on cores with MMU
efi/fdt: Avoid FDT manipulation after ExitBootServices()
x86/efi: Always map the first physical page into the EFI pagetables
ext4: validate s_first_meta_bg at mount time
PCI/ASPM: Handle PCI-to-PCIe bridges as roots of PCIe hierarchies
ANDROID: sched: Add Kconfig option DEFAULT_USE_ENERGY_AWARE to set ENERGY_AWARE feature flag
ANDROID: goldfish_sync: Fix sync_file_obj is NULL but dereferenced problem
ANDROID: goldfish_sync: Isolate single module to fix compilation
ANDROID: goldfish_sync: update defconfig for 4.9-compatible version
ANDROID: goldfish_sync: upgrade to new fence sync api
Linux 4.9.8
xfs: fix bmv_count confusion w/ shared extents
xfs: clear _XBF_PAGES from buffers when readahead page
xfs: extsize hints are not unlikely in xfs_bmap_btalloc
xfs: remove racy hasattr check from attr ops
xfs: verify dirblocklog correctly
xfs: fix COW writeback race
xfs: fix xfs_mode_to_ftype() prototype
xfs: don't wrap ID in xfs_dq_get_next_id
xfs: sanity check inode di_mode
xfs: sanity check inode mode when creating new dentry
xfs: replace xfs_mode_to_ftype table with switch statement
xfs: add missing include dependencies to xfs_dir2.h
xfs: sanity check directory inode di_size
xfs: make the ASSERT() condition likely
xfs: don't print warnings when xfs_log_force fails
xfs: don't rely on ->total in xfs_alloc_space_available
xfs: adjust allocation length in xfs_alloc_space_available
xfs: fix bogus minleft manipulations
xfs: bump up reserved blocks in xfs_alloc_set_aside
net: dsa: Bring back device detaching in dsa_slave_suspend()
lwtunnel: Fix oops on state free after encap module unload
net: Specify the owning module for lwtunnel ops
qmi_wwan/cdc_ether: add device ID for HP lt2523 (Novatel E371) WWAN card
af_unix: move unix_mknod() out of bindlock
r8152: don't execute runtime suspend if the tx is not empty
net: mpls: Fix multipath selection for LSR use case
bridge: netlink: call br_changelink() during br_dev_newlink()
net/mlx5e: Do not recycle pages from emergency reserve
tcp: initialize max window for a new fastopen socket
ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock
lwtunnel: fix autoload of lwt modules
net: phy: bcm63xx: Utilize correct config_intr function
net: fix harmonize_features() vs NETIF_F_HIGHDMA
vxlan: fix byte order of vxlan-gpe port number
virtio-net: restore VIRTIO_HDR_F_DATA_VALID on receiving
virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit
net sched actions: fix refcnt when GETing of action after bind
ax25: Fix segfault after sock connection timeout
ip6_tunnel: Account for tunnel header in tunnel MTU
ravb: do not use zero-length alignment DMA descriptor
mlx4: do not call napi_schedule() without care
openvswitch: maintain correct checksum state in conntrack actions
tcp: fix tcp_fastopen unaligned access complaints on sparc
net: systemport: Decouple flow control from __bcm_sysport_tx_reclaim
net: ipv4: fix table id in getroute response
net: lwtunnel: Handle lwtunnel_fill_encap failure
mlxsw: pci: Fix EQE structure definition
mlxsw: switchx2: Fix memory leak at skb reallocation
mlxsw: spectrum: Fix memory leak at skb reallocation
netvsc: add rcu_read locking to netvsc callback
r8152: fix the sw rx checksum is unavailable
FROMLIST: 9p: fix a potential acl leak
ANDROID: sched/walt: use do_div instead of division operator
ANDROID: sched: fix wrong truncation of walt_avg
ANDROID: arm: Fix #if/#ifdef typo in topology.c
ANDROID: arm: Fix build error "conflicting types for 'scale_cpu_capacity'"
ANDROID: net: ipv6: remove unused variable ifindex in
ANDROID: DEBUG: cpufreq: fix cpu_capacity tracing build for non-smp systems
ANDROID: arm: topology: Define TC2 energy and provide it to the scheduler
ANDROID: binder: fix format specifier for type binder_size_t
Linux 4.9.7
drm/i915: Remove WaDisableLSQCROPERFforOCL KBL workaround.
perf/core: Fix concurrent sys_perf_event_open() vs. 'move_group' race
mm, memcg: do not retry precharge charges
platform/x86: intel_mid_powerbtn: Set IRQ_ONESHOT
platform/x86: mlx-platform: free first dev on error
virtio_mmio: Set DMA masks appropriately
memory_hotplug: make zone_can_shift() return a boolean value
pinctrl: baytrail: Rectify debounce support
pinctrl: uniphier: fix Ethernet (RMII) pin-mux setting for LD20
pinctrl: broxton: Use correct PADCFGLOCK offset
s5k4ecgx: select CRC32 helper
IB/rxe: Prevent from completer to operate on non valid QP
IB/rxe: Fix rxe dev insertion to rxe_dev_list
IB/umem: Release pid in error and ODP flow
drm/i915: Check for NULL atomic state in intel_crtc_disable_noatomic()
drm/i915: Fix calculation of rotated x and y offsets for planar formats
drm/i915: Don't init hpd polling for vlv and chv from runtime_suspend()
drm/i915: Don't leak edid in intel_crt_detect_ddc()
drm/i915: prevent crash with .disable_display parameter
drm/i915: Clear ret before unbinding in i915_gem_evict_something()
v4l: tvp5150: Don't override output pinmuxing at stream on/off time
v4l: tvp5150: Fix comment regarding output pin muxing
v4l: tvp5150: Reset device at probe time, not in get/set format handlers
pctv452e: move buffer to heap, no mutex
iw_cxgb4: free EQ queue memory on last deref
SUNRPC: cleanup ida information when removing sunrpc module
NFSv4.0: always send mode in SETATTR after EXCLUSIVE4
NFSv4.1: Fix a deadlock in layoutget
nfs: Don't increment lock sequence ID after NFS4ERR_MOVED
parisc: Don't use BITS_PER_LONG in userspace-exported swab.h header
ARC: [arcompact] handle unaligned access delay slot corner case
ARC: udelay: fix inline assembler by adding LP_COUNT to clobber list
can: ti_hecc: add missing prepare and unprepare of the clock
can: c_can_pci: fix null-pointer-deref in c_can_start() - set device pointer
IB/srp: fix invalid indirect_sg_entries parameter value
IB/srp: fix mr allocation when the device supports sg gaps
IB/iser: Fix sg_tablesize calculation
IB/cxgb3: fix misspelling in header guard
s390/ptrace: Preserve previous registers for short regset write
s390/mm: Fix cmma unused transfer from pgste into pte
RDMA/cma: Fix unknown symbol when CONFIG_IPV6 is not enabled
Btrfs: remove ->{get, set}_acl() from btrfs_dir_ro_inode_operations
Btrfs: disable xattr operations on subvolume directories
Btrfs: remove old tree_root case in btrfs_read_locked_inode()
ISDN: eicon: silence misleading array-bounds warning
xfs: prevent quotacheck from overloading inode lru
sysctl: fix proc_doulongvec_ms_jiffies_minmax()
userns: Make ucounts lock irq-safe
vring: Force use of DMA API for ARM-based systems with legacy devices
mm, page_alloc: fix premature OOM when racing with cpuset mems update
mm, page_alloc: move cpuset seqcount checking to slowpath
mm, page_alloc: fix fast-path race with cpuset update or removal
mm, page_alloc: fix check for NULL preferred_zone
mm/mempolicy.c: do not put mempolicy before using its nodemask
mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp
drm/atomic: clear out fence when duplicating state
Revert "drm/radeon: always apply pci shutdown callbacks"
drm/vc4: fix a bounds check
drm/vc4: Return -EINVAL on the overflow checks failing.
drm/vc4: Fix an integer overflow in temporary allocation layout.
drm/vc4: Fix memory leak of the CRTC state.
drm/i915: Ignore bogus plane coordinates on SKL when the plane is not visible
drm: Fix broken VT switch with video=1366x768 option
drm: Schedule the output_poll_work with 1s delay if we have delayed event
tile/ptrace: Preserve previous registers for short regset write
fbdev: color map copying bounds checking
ANDROID: sched/walt: Drop arch-specific timer access
ANDROID: sched/walt: include missing header for arm_timer_read_counter()
ANDROID: fs: Export vfs_rmdir2
ANDROID: fs: Export free_fs_struct and set_fs_pwd
ANDROID: cpufreq: interactive: Use idle-end notifiers
FROMLIST: cpufreq: Add android's 'interactive' governor
ANDROID: cpufreq: conservative: fix duplicate 'static' error
ANDROID: sdcardfs: eliminate the offset argument to ->direct_IO
ANDROID: sdcardfs: make it use new .rename i_op
ANDROID: sdcardfs: Propagate dentry down to inode_change_ok()
ANDROID: sdcardfs: get rid of 'parent' argument of ->d_compare()
ANDROID: sdcardfs: add parent pointer into dentry name hash
ANDROID: sdcardfs: use wrappers to access i_mutex
ANDROID: mnt: remount should propagate to slaves of slaves
ANDROID: sdcardfs: Fix locking issue with permision fix up
ANDROID: sdcardfs: Switch ->d_inode to d_inode()
ANDROID: sdcardfs: Change magic value
ANDROID: sdcardfs: Use per mount permissions
ANDROID: sdcardfs: Add gid and mask to private mount data
ANDROID: sdcardfs: User new permission2 functions
ANDROID: vfs: Add permission2 for filesystems with per mount permissions
ANDROID: vfs: Add setattr2 for filesystems with per mount permissions
ANDROID: vfs: Allow filesystems to access their private mount data
ANDROID: mnt: Add filesystem private data to mount points
ANDROID: sdcardfs: Move directory unlock before touch
ANDROID: sdcardfs: fix external storage exporting incorrect uid
ANDROID: sdcardfs: Added top to sdcardfs_inode_info
ANDROID: sdcardfs: Switch package list to RCU
ANDROID: sdcardfs: Fix locking for permission fix up
ANDROID: sdcardfs: Check for other cases on path lookup
ANDROID: sdcardfs: override umask on mkdir and create
ANDROID: sched/debug: Add energy procfs interface
ANDROID: cpufreq: sched: Fix kernel crash on accessing sysfs file
ANDROID: FIXUP: sched/tune: add fixes missing from a previous patch
ANDROID: sched: tune: Fix lacking spinlock initialization
ANDROID: cgroup: Remove leftover instances of allow_attach
ANDROID: FIXUP: sched: scheduler-driven cpu frequency selection
ANDROID: sched/rt: Add Kconfig option to enable panicking for RT throttling
ANDROID: sched/rt: print RT tasks when RT throttling is activated
ANDROID: sched/fair: Favor higher cpus only for boosted tasks
ANDROID: sched/fair: call OPP update when going idle after migration
ANDROID: sched/cpufreq_sched: fix thermal capping events
ANDROID: sched/fair: Picking cpus with low OPPs for tasks that prefer idle CPUs
ANDROID: FIXUP: sched/tune: do initialization as a postcore_initicall
ANDROID: DEBUG: sched: add tracepoint for RD overutilized
ANDROID: sched/tune: Introducing a new schedtune attribute prefer_idle
ANDROID: sched: use util instead of capacity to select busy cpu
ANDROID: arch_timer: add error handling when the MPM global timer is cleared
ANDROID: FIXUP: sched: Fix double-release of spinlock in move_queued_task
ANDROID: FIXUP: sched/fair: Fix hang during suspend in sched_group_energy
ANDROID: FIXUP: sched: fix SchedFreq integration for both PELT and WALT
ANDROID: sched: EAS: Avoid causing spikes to max-freq unnecessarily
ANDROID: FIXUP: sched: fix set_cfs_cpu_capacity when WALT is in use
ANDROID: sched/walt: Accounting for number of irqs pending on each core
ANDROID: sched: Introduce Window Assisted Load Tracking (WALT)
ANDROID: sched/tune: fix PB and PC cuts indexes definition
ANDROID: sched/fair: optimize idle cpu selection for boosted tasks
ANDROID: FIXUP: sched/tune: fix accounting for runnable tasks
ANDROID: sched/tune: use a single initialisation function
ANDROID: FIXUP: sched/tune: fix payoff calculation for boost region
ANDROID: sched/tune: Add support for negative boost values
ANDROID: FIX: sched/tune: move schedtune_nornalize_energy into fair.c
ANDROID: FIX: sched/tune: update usage of boosted task utilisation on CPU selection
ANDROID: sched/fair: add tunable to set initial task load
ANDROID: sched/fair: add tunable to force selection at cpu granularity
ANDROID: sched: EAS: take cstate into account when selecting idle core
ANDROID: sched/cpufreq_sched: Consolidated update
ANDROID: FIXUP: sched: fix build for non-SMP target
ANDROID: DEBUG: sched/tune: add tracepoint on P-E space filtering
ANDROID: DEBUG: sched/tune: add tracepoint for energy_diff() values
ANDROID: DEBUG: sched/tune: add tracepoint for task boost signal
CHROMIUM: sched: update the average of nr_running
ANDROID: DEBUG: schedtune: add tracepoint for schedtune_tasks_update() values
ANDROID: DEBUG: schedtune: add tracepoint for CPU boost signal
ANDROID: DEBUG: schedtune: add tracepoint for SchedTune configuration update
ANDROID: DEBUG: sched,cpufreq: add cpu_capacity change tracepoint
ANDROID: DEBUG: sched: add tracepoint for CPU load/util signals
ANDROID: DEBUG: sched: add tracepoint for task load/util signals
ANDROID: DEBUG: sched: add tracepoint for cpu/freq scale invariance
ANDROID: sched/fair: filter energy_diff() based on energy_payoff value
ANDROID: sched/tune: add support to compute normalized energy
ANDROID: sched/fair: keep track of energy/capacity variations
ANDROID: sched/fair: add boosted task utilization
ANDROID: sched/{fair,tune}: track RUNNABLE tasks impact on per CPU boost value
ANDROID: sched/tune: compute and keep track of per CPU boost value
ANDROID: sched/tune: add initial support for CGroups based boosting
ANDROID: sched/fair: add boosted CPU usage
ANDROID: sched/fair: add function to convert boost value into "margin"
ANDROID: sched/tune: add sysctl interface to define a boost value
ANDROID: sched/tune: add detailed documentation
ANDROID: fixup! sched: scheduler-driven cpu frequency selection
ANDROID: sched: remove call of sched_avg_update from sched_rt_avg_update
ANDROID: sched/cpufreq_sched: add trace events
ANDROID: sched/fair: jump to max OPP when crossing UP threshold
ANDROID: sched/fair: cpufreq_sched triggers for load balancing
ANDROID: sched/{core,fair}: trigger OPP change request on fork()
ANDROID: sched/fair: add triggers for OPP change requests
ANDROID: sched: scheduler-driven cpu frequency selection
ANDROID: cpufreq: introduce cpufreq_driver_is_slow
ANDROID: sched: Add group_misfit_task load-balance type
ANDROID: sched: Add per-cpu max capacity to sched_group_capacity
ANDROID: sched: Do eas idle balance regardless of the rq avg idle value
ANDROID: arm64: Enable max freq invariant scheduler load-tracking and capacity support
ANDROID: arm: Enable max freq invariant scheduler load-tracking and capacity support
ANDROID: sched: Update max cpu capacity in case of max frequency constraints
ANDROID: cpufreq: Max freq invariant scheduler load-tracking and cpu capacity support
ANDROID: sched: Disable energy-unfriendly nohz kicks
ANDROID: sched: Consider a not over-utilized energy-aware system as balanced
ANDROID: sched: Energy-aware wake-up task placement
ANDROID: sched: Determine the current sched_group idle-state
ANDROID: sched, cpuidle: Track cpuidle state index in the scheduler
ANDROID: sched: Add over-utilization/tipping point indicator
ANDROID: sched: Estimate energy impact of scheduling decisions
ANDROID: sched: Extend sched_group_energy to test load-balancing decisions
ANDROID: sched: Calculate energy consumption of sched_group
ANDROID: sched: Highest energy aware balancing sched_domain level pointer
ANDROID: sched: Relocated cpu_util() and change return type
ANDROID: sched: Compute cpu capacity available at current frequency
ANDROID: sched: Support for extracting EAS energy costs from DT
ANDROID: arm64, topology: Updates to use DT bindings for EAS costing data
ANDROID: arm64: Cpu invariant scheduler load-tracking and capacity support
ANDROID: arm: Cpu invariant scheduler load-tracking and capacity support
ANDROID: sched: Introduce SD_SHARE_CAP_STATES sched_domain flag
ANDROID: sched: Initialize energy data structures
ANDROID: sched: Make energy awareness a sched feature
ANDROID: sched: Documentation for scheduler energy cost model
ANDROID: sched: Prevent unnecessary active balance of single task in sched group
ANDROID: sched: Enable idle balance to pull single task towards cpu with higher capacity
ANDROID: sched: Consider spare cpu capacity at task wake-up
ANDROID: sched: Add cpu capacity awareness to wakeup balancing
ANDROID: arm: Update arch_scale_cpu_capacity() to reflect change to define
ANDROID: arm64: Enable frequency invariant scheduler load-tracking support
ANDROID: arm: Enable frequency invariant scheduler load-tracking support
ANDROID: cpufreq: Frequency invariant scheduler load-tracking support
ANDROID: [CPUFREQ] Don't export governors for default governor
ANDROID: kernel/configs: recommended: CONFIG_ARM64_SW_TTBR0_PAN=y
ANDROID: kernel/configs: base: Enable QUOTA related configs
ANDROID: kernel/configs: recommended: Enable MEMORY_STATE_TIME
FROMLIST: config: android-base: enable hardened usercopy and kernel ASLR
FROMLIST: config: android-recommended: disable aio support
ANDROID: kernel/configs: recommended: enable fstack-protector-strong
ANDROID: kernel/configs: base: enable UID_CPUTIME
ANDROID: kernel/configs: base: restrict access to perf events
ANDROID: configs: base: enable configfs gadget functions
ANDROID: configs: merge AOSP config fragments
ANDROID: Implement memory_state_time, used by qcom,cpubw
ANDROID: dm: rebase for 4.9
ANDROID: usb: otg-wakelock: Remove wakelock.h dependencies
ANDROID: gpio_matrix: Remove wakelock.h dependencies
ANDROID: fiq_debugger: Remove wakelock.h dependencies
UPSTREAM: net: socket: don't set sk_uid to garbage value in ->setattr()
ANDROID: trace: net: use %pK for kernel pointers
UPSTREAM: net: ipv4: Don't crash if passing a null sk to ip_rt_update_pmtu.
UPSTREAM: net: inet: Support UID-based routing in IP protocols.
UPSTREAM: net: core: add UID to flows, rules, and routes
UPSTREAM: net: core: Add a UID field to struct sock.
ANDROID: fs: FS tracepoints to track IO.
ANDROID: MMC/UFS IO Latency Histograms.
CHROMIUM: fix warning when releasing active sync point
ANDROID: goldfish_pipe: fix allmodconfig build
ANDROID: goldfish: goldfish_pipe: fix locking errors
ANDROID: goldfish_pipe: fix call_kern.cocci warnings
ANDROID: goldfish_pipe: An implementation of more parallel pipe
ANDROID: goldfish_pipe: bugfixes and performance improvements.
ANDROID: goldfish: disable GOLDFISH_SYNC
ANDROID: goldfish: enable CONFIG_INET_DIAG_DESTROY
ANDROID: build: fix build config kernel_dir
ANDROID: dm verity: add minimum prefetch size
ANDROID: build: add build server configs for goldfish
UPSTREAM: trace: Update documentation for mono, mono_raw and boot clock
UPSTREAM: trace: Add an option for boot clock as trace clock
UPSTREAM: timekeeping: Add a fast and NMI safe boot clock
ANDROID: video: goldfishfb: fix platform_no_drv_owner.cocci warnings
ANDROID: arm64: rename ranchu defconfig to ranchu64
ANDROID: arch: x86: disable pic for Android toolchain
ANDROID: goldfish: Add goldfish sync driver
ANDROID: goldfish: add ranchu defconfigs
ANDROID: goldfish_audio: Clear audio read buffer status after each read
ANDROID: goldfish_events: no extra EV_SYN; register goldfish
ANDROID: goldfish_fb: Set pixclock = 0
ANDROID: goldfish: Enable ACPI-based enumeration for goldfish audio
ANDROID: goldfish: Enable ACPI-based enumeration for goldfish framebuffer
ANDROID: video: goldfishfb: add devicetree bindings
ANDROID: usb: gadget: function: cleanup: Add blank line after declaration
ANDROID: usb: gadget: f_mtp: simplify ptp NULL pointer check
ANDROID: usb: gadget: audio_source: fix comparison of distinct pointer types
ANDROID: binder: support for file-descriptor arrays.
ANDROID: binder: support for scatter-gather.
ANDROID: binder: add extra size to allocator.
ANDROID: binder: refactor binder_transact()
ANDROID: binder: support multiple /dev instances.
ANDROID: binder: deal with contexts in debugfs.
ANDROID: binder: support multiple context managers.
ANDROID: binder: split flat_binder_object.
ANDROID: [RFC]cgroup: Change from CAP_SYS_NICE to CAP_SYS_RESOURCE for cgroup migration permissions
CHROMIUM: cgroups: relax permissions on moving tasks between cgroups
ANDROID: dm: android-verity: Remove fec_header location constraint
ANDROID: fiq_debugger: Pass task parameter to unwind_frame()
ANDROID: input: keyreset: switch to orderly_reboot
ANDROID: cpuset: Make cpusets restore on hotplug
ANDROID: Don't show empty tag stats for unprivileged uids
ANDROID: dm: android-verity: Allow android-verity to be compiled as an independent module
ANDROID: dm-verity: adopt changes made to dm callbacks
ANDROID: dm verity fec: pack the fec_header structure
ANDROID: dm: android-verity: Verify header before fetching table
ANDROID: dm: allow adb disable-verity only in userdebug
ANDROID: dm: mount as linear target if eng build
ANDROID: dm: use default verity public key
ANDROID: dm: fix signature verification flag
ANDROID: dm: use name_to_dev_t
ANDROID: dm: rename dm-linear methods for dm-android-verity
ANDROID: dm: Minor cleanup
ANDROID: dm: Mounting root as linear device when verity disabled
ANDROID: dm-android-verity: Rebase on top of 4.1
ANDROID: dm: Add android verity target
ANDROID: dm: fix dm_substitute_devices()
ANDROID: dm: Rebase on top of 4.9
CHROMIUM: dm: boot time specification of dm=
ANDROID: usb: gadget: f_accessory: remove duplicate endpoint alloc
ANDROID: sdcardfs: fix itnull.cocci warnings
ANDROID: sdcardfs: Truncate packages_gid.list on overflow
ANDROID: netfilter: xt_quota2: make quota2_log work well
ANDROID: cpu: send KOBJ_ONLINE event when enabling cpus
ANDROID: dm verity fec: initialize recursion level
ANDROID: dm verity fec: add missing release from fec_ktype
ANDROID: dm verity fec: limit error correction recursion
FROMLIST: security,perf: Allow further restriction of perf_event_open
ANDROID: ARM64: Ignore Image-dtb from git point of view
ANDROID: arm64: add option to build Image-dtb
ANDROID: usb: gadget: f_midi: set fi->f to NULL when free f_midi function
ANDROID: xt_qtaguid: Fix panic caused by processing non-full socket.
ANDROID: fiq_debugger: Add fiq_debugger.disable option
FROMLIST: wlcore: Disable filtering in AP role
ANDROID: fiq_debugger: Add option to apply uart overlay by FIQ_DEBUGGER_UART_OVERLAY
ANDROID: usb: dual-role: make stub functions inline
ANDROID: quick selinux support for tracefs
ANDROID: xt_qtaguid: Fix panic caused by synack processing
ANDROID: fuse: Add support for d_canonical_path
ANDROID: vfs: change d_canonical_path to take two paths
ANDROID: netfilter: xt_qtaguid: seq_printf fixes
ANDROID: mmc: Add CONFIG_MMC_SIMULATE_MAX_SPEED
ANDROID: dm verity fec: add sysfs attribute fec/corrected
ANDROID: mm: Export do_munmap
ANDROID: sdcardfs: remove unneeded __init and __exit
ANDROID: sdcardfs: Remove unused code
ANDROID: fs: Export d_absolute_path
ANDROID: sdcardfs: remove effectless config option
ANDROID: inotify: Fix erroneous update of bit count
ANDROID: fs: sdcardfs: Declare LOOKUP_CASE_INSENSITIVE unconditionally
ANDROID: trace: cpufreq: fix typo in min/max cpufreq
ANDROID: sdcardfs: Add support for d_canonical_path
ANDROID: vfs: add d_canonical_path for stacked filesystem support
ANDROID: sdcardfs: Bring up to date with Android M permissions:
ANDROID: Changed type-casting in packagelist management
ANDROID: Port of sdcardfs to 4.4
ANDROID: Included sdcardfs source code for kernel 3.0
ANDROID: usb: gadget: Add support for MTP OS desc
CHROMIUM: usb: gadget: f_accessory: add .raw_request callback
CHROMIUM: usb: gadget: audio_source: add .free_func callback
CHROMIUM: usb: gadget: f_mtp: fix usb_ss_ep_comp_descriptor
CHROMIUM: usb: gadget: f_mtp: Add SuperSpeed support
ANDROID: dm-crypt: run in a WQ_HIGHPRI workqueue
ANDROID: power: Provide dummy log_suspend_abort_reason() if SUSPEND is disabled
ANDROID: PM / suspend: Add dependency on RTC_LIB
ANDROID: net: pppolac/pppopns: Replace msg.msg_iov with iov_iter_kvec()
ANDROID: mmc: sdio: Disable retuning in sdio_reset_comm()
ANDROID: kernel/watchdog: fix unused variable warning
ANDROID: usb: gadget: f_mtp: don't use le16 for u8 field
ANDROID: lowmemorykiller: fix declaration order warnings
ANDROID: mmc: move to a SCHED_FIFO thread
ANDROID: skip building drivers as modules
ANDROID: wakeup: Add the guard condition for len in pm_get_active_wakeup_sources
ANDROID: goldfish: pipe: fix platform_no_drv_owner.cocci warnings
ANDROID: epoll: use freezable blocking call
ANDROID: Fix for in kernel emergency remount when loop mounts are used
ANDROID: kbuild: Makefile.clean: make Kbuild and Makefile optional
ANDROID: kbuild: make it possible to specify the module output dir
ANDROID: ext4: Add support for FIDTRIM, a best-effort ioctl for deep discard trim
ANDROID: hardlockup: detect hard lockups without NMIs using secondary cpus
ANDROID: rtc-palmas: correct for bcd year
ANDROID: w1: ds2482: Manage SLPZ pin sleep state
ANDROID: fuse: Freeze client on suspend when request sent to userspace
ANDROID: serial_core: Add wake_peer uart operation
ANDROID: mm: add a field to store names for private anonymous memory
ANDROID: pstore/ram: Add ramoops_console_write_buf api
ANDROID: pstore: Update Documentation/android.txt
ANDROID: initramfs: Add skip_initramfs command line option
ANDROID: of: Fix build warnings
ANDROID: of: fix CONFIG_CMDLINE_EXTEND
ANDROID: ARM64: copy CONFIG_CMDLINE_EXTEND from ARM
ANDROID: of: Support CONFIG_CMDLINE_EXTEND config option
ANDROID: ARM: decompressor: Flush tlb before swiching domain 0 to client mode
ANDROID: ARM64: add option to build Image.gz/dtb combo
ANDROID: ARM: convert build of appended dtb zImage to list of dtbs
ANDROID: ARM: add config option to build zImage/dtb combo
ANDROID: ARM: Fix dtb list when DTB_IMAGE_NAMES is empty
ANDROID: arm64: pass return address to dma_common_contiguous_remap
ANDROID: arch: arm64: force -fno-pic
ANDROID: arm64: process: dump memory around registers when displaying regs
ANDROID: arm64: check for upper PAGE_SHIFT bits in pfn_valid()
ANDROID: ARM: fault: assume no context when IRQs are disabled during data abort.
ANDROID: ARM: Fix "Make low-level printk work" to use a separate config option
ANDROID: ARM: add option to flush console before reboot
ANDROID: ARM: Make low-level printk work
ANDROID: Optionally flush entire dcache from v6_dma_flush_range
ANDROID: process: Add display of memory around registers when displaying regs.
ANDROID: security: Add proper checks for Android specific capability checks
ANDROID: uid_cputime: skip power reporting per uid for now
ANDROID: uid_cputime: Check for the range while removing range of UIDs.
ANDROID: uid_cputime: Iterates over all the threads instead of processes.
ANDROID: uid_cputime: fix cputime overflow
ANDROID: uid_cputime: Avoids double accounting of process stime, utime and cpu_power in task exit.
ANDROID: uid_cputime: Extends the cputime functionality to report power per uid
ANDROID: proc: uid_cputime: fix show_uid_stat permission
ANDROID: proc: uid_cputime: create uids from kuids
ANDROID: proc: uid: Adds accounting for the cputimes per uid.
ANDROID: fixup! proc: make oom adjustment files user read-only
ANDROID: proc: make oom adjustment files user read-only
ANDROID: proc: smaps: Allow smaps access for CAP_SYS_RESOURCE
ANDROID: wakeup_reason: use vsnprintf instead of snsprintf for vargs.
ANDROID: power: wakeup_reason: fix suspend time reporting
ANDROID: wakeup: Add last wake up source logging for suspend abort reason.
ANDROID: Power: Report suspend times from last_suspend_time
ANDROID: Make suspend abort reason logging depend on CONFIG_PM_SLEEP
ANDROID: power: Add check_wakeup_reason() to verify wakeup source irq
ANDROID: power: Adds functionality to log the last suspend abort reason.
ANDROID: power: Avoids bogus error messages for the suspend aborts.
ANDROID: power: Add property CHARGE_COUNTER_EXT and 64-bit precision properties
ANDROID: Power: Changes the permission to read only for sysfs file /sys/kernel/wakeup_reasons/last_resume_reason
ANDROID: power: wakeup_reason: rename irq_count to irqcount
ANDROID: Power: Add guard condition for maximum wakeup reasons
ANDROID: POWER: fix compile warnings in log_wakeup_reason
ANDROID: Power: add an API to log wakeup reasons
ANDROID: PM / Suspend: Print wall time at suspend entry and exit
ANDROID: power: power_supply: add POWER_SUPPLY_PROP_CHARGE_ENABLED
ANDROID: power: power_supply: add POWER_SUPPLY_PROP_USB_OTG
ANDROID: power: power_supply: move POWER_SUPPLY_PROP_USB_HC to type 'int' order
ANDROID: power_supply: Add custom property for USB High Current mode
ANDROID: trace: power: add trace_clock_set_parent
ANDROID: trace: cpufreq: Add tracing for min/max cpufreq
ANDROID: trace: fix compilation for 4.1
ANDROID: trace/events: fix gpu event timestamp formatting
ANDROID: trace: add non-hierarchical function_graph option
ANDROID: trace: Add an option to show tgids in trace output
ANDROID: trace/events: add gpu trace events
ANDROID: sync: add Documentation/sync.txt
ANDROID: ARM: Call idle notifiers
ANDROID: Move x86_64 idle notifiers to generic
ANDROID: cpuidle: governor: menu: don't use loadavg
ANDROID: sched: add sched blocked tracepoint which dumps out context of sleep.
ANDROID: sched: Enable might_sleep before initializing drivers.
ANDROID: fiq_debugger: Build fixes for 4.1
ANDROID: fiq_debugger: Add fiq_watchdog_triggered api
ANDROID: fiq_debugger: Call fiq_debugger_printf through a function pointer from cpu specific code
ANDROID: fiq_debugger: add ARM64 support
ANDROID: fiq_debugger: split arm support into fiq_debugger_arm.c
ANDROID: fiq_debugger: use pt_regs for registers
ANDROID: fiq_debugger: allow compiling without CONFIG_FIQ_GLUE
ANDROID: fiq_debugger: rename debug->fiq_debugger
ANDROID: fiq_debugger: move into drivers/staging/android/fiq_debugger/
ANDROID: ARM: fiq_glue: Add custom fiq return handler api.
ANDROID: ARM: kgdb: ignore breakpoint instructions from user mode
ANDROID: ARM: fiq_debugger: Update tty code for 3.9
ANDROID: ARM: fiq_debugger: Use kmsg_dumper to dump kernel logs
ANDROID: ARM: fiq_debugger: Fix to compile on 3.7
ANDROID: ARM: fiq_debugger: fix uninitialised spin_lock.
ANDROID: ARM: fiq_debugger: lock between tty and console writes
ANDROID: ARM: fiq_debugger: add process context reboot command
ANDROID: ARM: fiq_debugger: fix multiple consoles and make it a preferred console
ANDROID: kdb: support new lines without carriage returns
ANDROID: ARM: fiq_debugger: add support for kgdb
ANDROID: ARM: fiq_debugger: add debug_putc
ANDROID: ARM: fiq_debugger: add support for reboot commands
ANDROID: ARM: fiq_debugger: fix compiling for v3.3
ANDROID: ARM: Add generic fiq serial debugger
ANDROID: ARM: Add fiq_glue
ANDROID: fix false disconnect due to a signal sent to the reading process
ANDROID: usb: gadget: cleanup: fix unused variable and function warnings
ANDROID: usb: gadget: build audio_source function only if SND is enabled
ANDROID: usb: gadget: configfs: handle gadget reset request for android
ANDROID: usb: gadget: create F_midi device
ANDROID: usb: gadget: Add device attribute to determine gadget state
ANDROID: usb: phy: fix dual role sysfs build if kernel modules are supported
ANDROID: usb: phy: Dual role sysfs class definition
ANDROID: usb: gadget: fix NULL ptr derefer while symlinking PTP func
ANDROID: usb:gadget:Add "state" attribute to android_device
ANDROID: usb: gadget: Do not disconnect unregistered dev
ANDROID: usb: gadget: Relocate f_accessory
ANDROID: usb: gadget: Accessory:Migrate to USB_FUNCTION API
ANDROID: usb: gadget: Move gadget functions code
ANDROID: usb:gadget:audio_source: Move to USB_FUNCTION API
ANDROID: usb: gadget: Add function devices to the parent
ANDROID: usb: gadget: f_audio_source:replace deprecated API
ANDROID: usb: gadget: check for accessory device before disconnecting HIDs
ANDROID: usb: gadget: Add Uevent to notify userspace
ANDROID: usb: gadget: configfs: Add usb_function ptr to fi struct
ANDROID: usb: gadget: mtp/ptp: Migrate functions to the USB_FUNCTION interface
Linux 4.9.6
libceph: stop allocating a new cipher on every crypto request
libceph: uninline ceph_crypto_key_destroy()
tools/virtio/ringtest: fix run-on-all.sh for offline cpus
selftest/powerpc: Wrong PMC initialized in pmc56_overflow test
soc: ti: wkup_m3_ipc: Fix error return code in wkup_m3_ipc_probe()
spi: pxa2xx: add missed break
dmaengine: pl330: Fix runtime PM support for terminated transfers
dmaengine: rcar-dmac: unmap slave resource when channel is freed
s5p-mfc: Fix clock management in s5p_mfc_release() function
s5p-cec: mark PM functions as __maybe_unused again
st-hva: fix some error handling in hva_hw_probe()
ite-cir: initialize use_demodulator before using it
gs1662: drop kfree for memory allocated with devm_kzalloc
platform: pxa_camera: add VIDEO_V4L2 dependency
blackfin: check devm_pinctrl_get() for errors
rpmsg: virtio_rpmsg_bus: fix channel creation
mtd: spi-nor: Fix some error codes in cqspi_setup_flash()
mtd: spi-nor: Off by one in cqspi_setup_flash()
PM / devfreq: Fix the bug of devfreq_add_device when governor is NULL
PM / devfreq: exynos-bus: Fix the wrong return value
scsi: mpt3sas: fix hang on ata passthrough commands
scsi: ses: Fix SAS device detection in enclosure
swiotlb: Add swiotlb=noforce debug option
swiotlb: Convert swiotlb_force from int to enum
arm64: Fix swiotlb fallback allocation
arm64: mm: avoid name clash in __page_to_voff()
xprtrdma: Squelch "max send, max recv" messages at connect time
xprtrdma: Make FRWR send queue entry accounting more accurate
libceph: make sure ceph_aes_crypt() IV is aligned
ceph: fix endianness bug in frag_tree_split_cmp
ceph: fix endianness of getattr mask in ceph_d_revalidate
ceph: fix ceph_get_caps() interruption
ceph: fix scheduler warning due to nested blocking
ARM: 8613/1: Fix the uaccess crash on PB11MPCore
ARM: ux500: fix prcmu_is_cpu_in_wfi() calculation
ARM: dts: omap3: Fix Card Detect and Write Protect on Logic PD SOM-LV
ARM: dts: imx6qdl-nitrogen6_max: fix sgtl5000 pinctrl init
ARM: dts: omap2: Add an empty chosen node to top level DTSI
ARM: dts: omap3: Add an empty chosen node to top level DTSI
ARM: dts: am4372: Add an empty chosen node to top level DTSI
ARM: dts: omap5: Add an empty chosen node to top level DTSI
ARM: dts: omap4: Add an empty chosen node to top level DTSI
ARM: dts: am33xx: Add an empty chosen node to top level DTSI
ARM: dts: dm814x: Add an empty chosen node to top level DTSI
ARM: dts: dm816x: Add an empty chosen node to top level DTSI
ARM: dts: dra7: Add an empty chosen node to top level DTSI
libceph: remove now unused ceph_*{en,de}crypt*() functions
libceph: switch ceph_x_decrypt() to ceph_crypt()
libceph: switch ceph_x_encrypt() to ceph_crypt()
libceph: tweak calcu_signature() a little
libceph: rename and align ceph_x_authorizer::reply_buf
libceph: introduce ceph_crypt() for in-place en/decryption
libceph: introduce ceph_x_encrypt_offset()
libceph: old_key in process_one_ticket() is redundant
libceph: ceph_x_encrypt_buflen() takes in_len
Input: ALPS - fix TrackStick support for SS5 hardware
arm64/ptrace: Reject attempts to set incomplete hardware breakpoint fields
arm64/ptrace: Avoid uninitialised struct padding in fpr_set()
arm64/ptrace: Preserve previous registers for short regset write - 3
arm64/ptrace: Preserve previous registers for short regset write - 2
arm64/ptrace: Preserve previous registers for short regset write
arm64: avoid returning from bad_mode
ARM: dts: da850-evm: fix read access to SPI flash
ARM: dts: OMAP5 / DRA7: indicate that SATA port 0 is available.
ceph: fix bad endianness handling in parse_reply_info_extra
ibmvscsis: Fix max transfer length
ibmvscsis: Fix sleeping in interrupt context
ARM: 8634/1: hw_breakpoint: blacklist Scorpion CPUs
svcrdma: avoid duplicate dma unmapping during error recovery
clocksource/exynos_mct: Clear interrupt when cpu is shut down
ubifs: Fix journal replay wrt. xattr nodes
mac80211: implement multicast forwarding on fast-RX path
qla2xxx: Fix crash due to null pointer access
x86/ioapic: Restore IO-APIC irq_chip retrigger callback
powerpc: Ignore reserved field in DCSR and PVR reads and writes
powerpc/ptrace: Preserve previous TM fprs/vsrs on short regset write
powerpc/ptrace: Preserve previous fprs/vsrs on short regset write
powerpc/perf: Fix PM_BRU_CMPL event code for power9
powerpc/icp-opal: Fix missing KVM case and harden replay
KVM: arm/arm64: vgic: Fix deadlock on error handling
KVM: s390: do not expose random data via facility bitmap
mtd: nand: xway: fix build because of module functions
mtd: nand: xway: disable module support
mtd: nand: lpc32xx: fix invalid error handling of a requested irq
ieee802154: atusb: do not use the stack for buffers to make them DMA able
mmc: mxs-mmc: Fix additional cycles after transmission stop
mmc: sdhci-acpi: Only powered up enabled acpi child devices
HID: corsair: fix control-transfer error handling
HID: corsair: fix DMA buffers on stack
PCI: Enumerate switches below PCI-to-PCIe bridges
PCI: designware: Check for iATU unroll only on platforms that use ATU
fuse: fix time_to_jiffies nsec sanity check
fuse: clear FR_PENDING flag when moving requests out of pending queue
ARC: module: Fix !CONFIG_ARC_DW2_UNWIND builds
libnvdimm, namespace: fix pmem namespace leak, delete when size set to zero
svcrpc: don't leak contexts on PROC_DESTROY
sunrpc: don't call sleeping functions from the notifier block callbacks
rcu: Narrow early boot window of illegal synchronous grace periods
rcu: Remove cond_resched() from Tiny synchronize_sched()
x86/PCI: Ignore _CRS on Supermicro X8DTH-i/6/iF/6F
tmpfs: clear S_ISGID when setting posix ACLs
ARM: dts: omap3: Add DTS for Logic PD SOM-LV 37xx Dev Kit
ARM: dts: imx31: fix AVIC base address
ARM: dts: imx31: move CCM device node to AIPS2 bus devices
ARM: dts: imx31: fix clock control module interrupts description
ARM: dts: imx6q-cm-fx6: fix fec pinctrl
ARM: dts: r8a7794: remove Z clock
ARM: dts: r8a7794: Use SYSC "always-on" PM Domain for sound
ARM: dts: bcm283x: fix typo in mailbox address
perf jit: Enable jitdump support without dwarf
perf scripting: Avoid leaking the scripting_context variable
perf callchain: Fixup help/config for no-unwinding
perf diff: Do not overwrite valid build id
perf trace: Check if MAP_32BIT is defined (again)
perf mem: Fix --all-user/--all-kernel options
perf trace: Use the syscall raw_syscalls:sys_enter timestamp
IB/IPoIB: Remove can't use GFP_NOIO warning
IB/mlx4: Check if GRH is available before using it
IB/mlx4: When no DMFS for IPoIB, don't allow NET_IF QPs
IB/mlx4: Fix port query for 56Gb Ethernet links
IB/mlx4: Handle well-known-gid in mad_demux processing
IB/mlx4: Fix out-of-range array index in destroy qp flow
IB/mlx4: Set traffic class in AH
IB/mlx5: Wait for all async command completions to complete
IB/mlx5: Assign SRQ type earlier
IB/mlx5: Fix reported max SGE calculation
IB/mlx5: Avoid system crash when enabling many VFs
IB/rxe: avoid putting a large struct rxe_qp on stack
IB/rxe: Increase max number of completions to 32k
IB/core: Release allocated memory in cache setup failure
Linux 4.9.5
ANDROID: usb: gadget: f_audio_source: Fixed USB Audio Class Interface Descriptor
ANDROID: usb: gadget: f_audio_source: change max ISO packet size
ANDROID: usb: gadget: f_accessory: Enabled Zero Length Packet (ZLP) for acc_write
ANDROID: drivers: usb: gadget: 64-bit related type fixes
ANDROID: usb: gadget: f_accessory: move userspace interface to uapi
ANDROID: usb: gadget: f_mtp: move userspace interface to uapi
ANDROID: USB: remove duplicate out endpoint creation in MTP mode
ANDROID: usb: gadget: Fix android gadget driver build
ANDROID: usb: gadget: Fixes and hacks to make android usb gadget compile on 3.8
ANDROID: usb: otg: otg-wakelock: Fix build for 3.7
ANDROID: usb: gadget: accessory: Fix section mismatch (again)
ANDROID: USB: gadget: f_audio_source: New gadget driver for audio output
ANDROID: USB: gadget: f_accessory: Add support for HID input devices
ANDROID: USB: gadget: Add ACCESSORY_SET_AUDIO_MODE control request and ioctl
ANDROID: usb: gadget: accessory: Fix section mismatch
ANDROID: usb: otg: otg-wakelock: Fix build for 3.4
ANDROID: usb: gadget: adb: Only enable the gadget when adbd is ready
ANDROID: usb: gadget: adb: do not set error flag when dequeuing req
ANDROID: usb: gadget: adb: allow freezing in adb_read
ANDROID: usb: gadget: accessory: Add Android Accessory function
ANDROID: usb: gadget: adb: Add ADB function
ANDROID: usb: gadget: mtp: Add MTP/PTP function
ANDROID: usb: otg: otg-wakelock: fix build for 3.3
ANDROID: usb: otg: Temporarily grab wakelock on charger and disconnect events
ANDROID: USB: OTG: Take wakelock when VBUS present
ANDROID: wlan: Add get_wake_irq functionality
ANDROID: Add flags parameter to get_country_code template
ANDROID: net: wireless: Add get_country_code functionality to platform
ANDROID: network: wireless: Add get_mac_addr functionality to platform
ANDROID: wlan: Create generic wlan platform data header
ANDROID: net: wireless: Decrease scan entry expiration to avoid stall results
ANDROID: bridge: Have tx_bytes count headers like rx_bytes.
ANDROID: rfkill: Introduce CONFIG_RFKILL_PM and use instead of CONFIG_PM to power down
ANDROID: net: ipv6: fix virtual tunneling build
ANDROID: tcp: fix tcp_default_init_rwnd() for 4.1
ANDROID: net: support marking accepting TCP sockets
ANDROID: tcp: add a sysctl to config the tcp_default_init_rwnd
ANDROID: xt_qtaguid: fix a race condition in if_tag_stat_update
ANDROID: netfilter: xt_qtaguid/socket: build fixes for 4.4
ANDROID: netfilter: xt_qtaguid: xt_socket: build fixes
ANDROID: net: xt_qtaguid/xt_socket: fix refcount underflow and crash
ANDROID: xt_qtaguid: use sock_gen_put() instead of xt_socket_put_sk()
ANDROID: xt_qtaguid: Use sk_callback_lock read locks before reading sk->sk_socket
ANDROID: xt_qtaguid: fix broken uid/gid range check
ANDROID: netfilter: Build fixups - kuid/kguid changes & xt_socket_get/put_sk
ANDROID: net: ipv6: autoconf routes into per-device tables
ANDROID: nf: IDLETIMER: Fix broken uid field in the msg
ANDROID: nf: IDLETIMER: Adds the uid field in the msg
ANDROID: netfilter: fix seq_printf type mismatch warning
ANDROID: nf: Remove compilation error caused by e8430cbed3ef15fdb1ac26cfd020e010aa5f1c35
ANDROID: nf: IDLETIMER: time-stamp and suspend/resume handling.
ANDROID: xt_qtaguid: Fix boot panic
ANDROID: net: kuid/kguid build fixes
ANDROID: netfilter: ipv6: fix crash caused by ipv6_find_hdr()
ANDROID: netfilter: xt_qtaguid: 64-bit warning fixes
ANDROID: netfilter: xt_qtaguid: fix memory leak in seq_file handlers
ANDROID: netfilter: xt_qtaguid: fix bad tcp_time_wait sock handling
ANDROID: netfilter: xt_qtaguid: 3.10 fixes
ANDROID: netfilter: xt_quota2: 3.10 fixes.
ANDROID: netfilter: qtaguid: rate limit some of the printks
ANDROID: netfilter: xt_qtaguid: Allow tracking loopback
ANDROID: netfilter: xt_qtaguid: extend iface stat to report protocols
ANDROID: netfilter: xt_qtaguid: remove AID_* dependency for access control
ANDROID: netfilter: qtaguid: Don't BUG_ON if create_if_tag_stat fails
ANDROID: netfilter: xt_qtaguid: fix error exit that would keep a spinlock.
ANDROID: netfilter: xt_qtaguid: report only uid tags to non-privileged processes
ANDROID: netfilter: xt_qtaguid: start tracking iface rx/tx at low level
ANDROID: netfilter: xt_IDLETIMER: Add new netlink msg type
ANDROID: netfilter: xt_qtaguid: fix ipv6 protocol lookup
ANDROID: netfilter: qtaguid: initialize a local var to keep compiler happy.
ANDROID: netfilter: fixup the quota2, and enable.
ANDROID: netfilter: adding the original quota2 from xtables-addons
ANDROID: netfilter: add xt_qtaguid matching module
ANDROID: net: PPPoPNS and PPPoLAC build fixes for 4.4
ANDROID: Hack: net: PPPoPNS and PPPoLAC build fixes for 4.1
ANDROID: net: pppopns: pppolac: fix sendmsg function calls
ANDROID: net: PPPoPNS: Remove length argument from data_ready
ANDROID: net: move PPPoLAC and PPPoPNS headers to uapi
ANDROID: Include if_pppolac.h and if_pppopns.h into header-y target
ANDROID: net: PPPoPNS and PPPoLAC update to use PPP_MRU instead of PPP_MRU
ANDROID: net: Reorder incoming packets in PPPoLAC and PPPoPNS.
ANDROID: net: PPPoPNS and PPPoLAC fixes.
ANDROID: net: add PPP on PPTP Network Server (PPPoPNS) driver.
ANDROID: net: add PPP on L2TP Access Concentrator (PPPoLAC) driver.
ANDROID: sysfs_net_ipv4: Add sysfs-based knobs for controlling TCP window size
ANDROID: net: Only NET_ADMIN is allowed to fully control TUN interfaces.
ANDROID: net: Replace AID_NET_RAW checks with capable(CAP_NET_RAW).
ANDROID: security: Add AID_NET_RAW and AID_NET_ADMIN capability check in cap_capable().
ANDROID: Paranoid network.
ANDROID: Add android_aid.h
ANDROID: fs: block_dump: Don't display inode changes if block_dump < 2
ANDROID: mmc: core: Remove stray CONFIG_EXPERIMENTAL dependencies
ANDROID: mmc: Add "ignore mmc pm notify" functionality
ANDROID: mmc: sdio: Fix sdio_reset_comm for sync
ANDROID: mmc: sdio: fix sdio_reset_comm() voltage selection
ANDROID: mmc: sdio: Fix enable_hs and enable_wide in sdio_reset_comm()
ANDROID: mmc: sdio: Add high speed support to sdio_reset_comm()
ANDROID: mmc: sdio: Claim host in sdio_reset_comm()
ANDROID: mmc: Add new API call 'sdio_reset_comm' for resetting communication with an SDIO device
ANDROID: mmc: Add sdio_readb_ext() function
ANDROID: mmc: Add concept of an 'embedded' SDIO device.
ANDROID: mmc: sd: Add retries in re-detection
ANDROID: mmc: sd: When resuming, try a little harder to init the card
ANDROID: mmc: sd: Add new CONFIG_MMC_PARANOID_SD_INIT for enabling retries during SD detection
ANDROID: mtd: nand: Allow NAND chip ids to be included standalone.
ANDROID: gpio_input: convert from wakelocks to wakeup sources
ANDROID: input: Made keyreset more robust
ANDROID: input: Changed keyreset to act as a wrapper for keycombo.
ANDROID: input: add keycombo, a general key combo driver.
ANDROID: input: misc: keychord: move header to uapi
ANDROID: input: misc: keychord: log when keychord triggered
ANDROID: input: keychord: Add keychord driver
ANDROID: input: Add keyreset driver.
ANDROID: input: misc: gpio_event: remove early suspend
ANDROID: Input: Generic GPIO Input device.
ANDROID: lowmemorykiller: use module_param_cb instead of __module_param_call
ANDROID: lowmemorykiller: trace kill events.
ANDROID: staging: lowmemorykiller: Add config option to support oom_adj values
pinctrl: sh-pfc: Do not unconditionally support PIN_CONFIG_BIAS_DISABLE
arm64: hugetlb: fix the wrong return value for huge_ptep_set_access_flags
arm64: hugetlb: remove the wrong pmd check in find_num_contig()
arm64: hugetlb: fix the wrong address for several functions
powerpc/powernv: Don't warn on PE init if unfreeze is unsupported
powerpc/ibmebus: Fix device reference leaks in sysfs interface
powerpc/ibmebus: Fix further device reference leaks
powerpc/mm: Correct process and partition table max size
bus: vexpress-config: fix device reference leak
blk-mq: Always schedule hctx->next_cpu
power: supply: bq27xxx_battery: Fix register map for BQ27510 and BQ27520
bq24190_charger: Fix PM runtime use for bq24190_battery_set_property
iw_cxgb4: Fix error return code in c4iw_rdev_open()
powercap/intel_rapl: fix and tidy up error handling
ACPI / APEI: Fix NMI notification handling
block: cfq_cpd_alloc() should use @gfp
block: Change extern inline to static inline
ACPI / CPPC: set an error code on probe error path
regulators: helpers: Fix handling of bypass_val_on in get_bypass_regmap
cpufreq: powernv: Disable preemption while checking CPU throttling state
powerpc/64: Simplify adaptation to new ISA v3.00 HPTE format
remoteproc: st: Fix error return code in st_rproc_probe()
remoteproc: qcom_wcnss: Fix circular module dependency
drm: Initialise drm_mm.head_node.allocated
drm/i915: Move the min_pixclk[] handling to the end of readout
drm/panel: simple: Check against num_timings when setting preferred for timing
drm: avoid uninitialized timestamp use in wait_vblank
drm/i915/gen9: Fix PCODE polling during SAGV disabling
i2c: mux: pca954x: fix i2c mux selection caching
NFSv4.1: nfs4_fl_prepare_ds must be careful about reporting success.
NFS: Fix a performance regression in readdir
pNFS: Fix race in pnfs_wait_on_layoutreturn
NFS: fix typo in parameter description
pinctrl: meson: fix gpio request disabling other modes
btrfs: fix error handling when run_delayed_extent_op fails
btrfs: fix locking when we put back a delayed ref that's too new
nvme: apply DELAY_BEFORE_CHK_RDY quirk at probe time too
x86/cpu: Fix bootup crashes by sanitizing the argument of the 'clearcpuid=' command-line option
i2c: piix4: Avoid race conditions with IMC
net/mlx5: Only cancel recovery work when cleaning up device
USB: serial: ch341: fix modem-control and B0 handling
drm/amdgpu: drop verde dpm quirks
drm/amdgpu: update si kicker smc firmware
drm/radeon: drop verde dpm quirks
drm/radeon: update smc firmware selection for SI
drm: Clean up planes in atomic commit helper failure path
drm/i915/gen9: Fix PCODE polling timeout in stable backport
net/af_iucv: don't use paged skbs for TX on HiperSockets
sysctl: Drop reference added by grab_header in proc_sys_readdir
Clearing FIFOs in RS485 emulation mode causes subsequent transmits to break
extcon: return error code on failure
sysrq: attach sysrq handler correctly for 32-bit kernel
orinoco: Use shash instead of ahash for MIC calculations
ibmvscsis: Fix srp_transfer_data fail return code
tty/serial: atmel_serial: BUG: stop DMA from transmitting in stop_tx
tty/serial: atmel: RS485 half duplex w/DMA: enable RX after TX is done
virtio_blk: avoid DMA to stack for the sense buffer
dmaengine: omap-dma: Fix dynamic lch_map allocation
drivers: char: mem: Fix thinkos in kmem address checks
mnt: Protect the mountpoint hashtable with mount_lock
pid: fix lockdep deadlock warning due to ucount_lock
vme: Fix wrong pointer utilization in ca91cx42_slave_get
Revert "tty: serial: 8250: add CON_CONSDEV to flags"
ASoC: hdmi-codec: use unsigned type to structure members with bit-field
btrfs: fix crash when tracepoint arguments are freed by wq callbacks
xhci: fix deadlock at host remove by running watchdog correctly
fix a fencepost error in pipe_advance()
i2c: fix kernel memory disclosure in dev interface
i2c: print correct device invalid address
Input: elants_i2c - avoid divide by 0 errors on bad touchscreen data
USB: serial: ch341: fix open and resume after B0
USB: serial: ch341: fix control-message error handling
USB: serial: ch341: fix open error handling
USB: serial: ch341: fix resume after reset
USB: serial: ch341: fix initial modem-control state
USB: serial: kl5kusb105: fix line-state error handling
usb: musb: fix runtime PM in debugfs
wusbcore: Fix one more crypto-on-the-stack bug
x86/CPU/AMD: Fix Bulldozer topology
x86/bugs: Separate AMD E400 erratum and C1E bug
x86/cpu/AMD: Clean up cpu_llc_id assignment per topology feature
bridge: netfilter: Fix dropping packets that moving through bridge interface
xfs: Timely free truncated dirty pages
gpio: Move freeing of GPIO hogs before numbing of the device
nl80211: fix sched scan netlink socket owner destruction
x86/efi: Don't allocate memmap through memblock after mm_init()
efi/x86: Prune invalid memory map entries and fix boot regression
efi/libstub/arm*: Pass latest memory map to the kernel
KVM: x86: Introduce segmented_write_std
KVM: x86: emulate FXSAVE and FXRSTOR
KVM: x86: add asm_safe wrapper
KVM: x86: add Align16 instruction flag
KVM: x86: fix NULL deref in vcpu_scan_ioapic
KVM: x86: flush pending lapic jump label updates on module unload
jump_labels: API for flushing deferred jump label updates
KVM: eventfd: fix NULL deref irqbypass consumer
KVM: x86: fix emulation of "MOV SS, null selector"
mm/hugetlb.c: fix reservation race when freeing surplus pages
mm/slab.c: fix SLAB freelist randomization duplicate entries
mm: support anonymous stable page
mm, memcg: fix the active list aging for lowmem requests when memcg is enabled
ocfs2: fix crash caused by stale lvb with fsdlm plugin
mm: fix devm_memremap_pages crash, use mem_hotplug_{begin, done}
mm: pmd dirty emulation in page fault handler
dax: fix deadlock with DAX 4k holes
zram: support BDI_CAP_STABLE_WRITES
zram: revalidate disk under init_lock
selftests: do not require bash for the generated test
selftests: do not require bash to run netsocktests testcase
drm/savage: dereferencing an error pointer
drm/vc4: Fix a couple error codes in vc4_cl_lookup_bos()
drm/tegra: dpaux: Fix error handling
regulator: axp20x: Fix axp809 ldo_io registration error on cold boot
regulator: tps65086: Fix 25mV ranges for BUCK regulators
pinctrl: sh-pfc: Add helper to handle bias lookup table
pinctrl: sh-pfc: r8a7795: Use lookup function for bias data
pinctrl: imx: fix imx_pinctrl_desc initialization
Input: i8042 - add Pegatron touchpad to noloop table
Input: xpad - use correct product id for x360w controllers
Linux 4.9.4
rtlwifi: rtl_usb: Fix missing entry in USB driver's private data
rtlwifi: Fix enter/exit power_save
drm/i915/gen9: Fix PCODE polling during CDCLK change notification
ALSA: usb-audio: Add a quirk for Plantronics BT600
spi: mvebu: fix baudrate calculation for armada variant
ARM: omap2+: am437x: rollback to use omap3_gptimer_timer_init()
ARM: 8631/1: clkdev: Detect errors in clk_hw_register_clkdev() for mass registration
ARM: OMAP4+: Fix bad fallthrough for cpuidle
ARM: OMAP5: Fix build for PM code
ARM: OMAP5: Fix mpuss_early_init
bus: arm-ccn: Prevent hotplug callback leak
svcrdma: Clear xpt_bc_xps in xprt_setup_rdma_bc() error exit arm
ARM: qcom_defconfig: Fix MDM9515 LCC and GCC config
ARM: zynq: Reserve correct amount of non-DMA RAM
ARM: pxa: fix pxa25x interrupt init
ARM64: dts: bcm2835: Fix bcm2837 compatible string
ARM64: dts: bcm2837-rpi-3-b: remove incorrect pwr LED
arm64: dts: mt8173: Fix auxadc node
tools/virtio: fix READ_ONCE()
powerpc: Fix build warning on 32-bit PPC
ALSA: firewire-tascam: Fix to handle error from initialization of stream data
HID: hid-cypress: validate length of report
net: vrf: do not allow table id 0
net: ipv4: Fix multipath selection with vrf
net/mlx5e: Remove WARN_ONCE from adaptive moderation code
gro: Disable frag0 optimization on IPv6 ext headers
gro: use min_t() in skb_gro_reset_offset()
gro: Enter slow-path if there is no tailroom
net: add the AF_QIPCRTR entries to family name tables
net: dsa: Ensure validity of dst->ds[0]
r8152: fix rx issue for runtime suspend
r8152: split rtl8152_suspend function
net: dsa: bcm_sf2: Utilize nested MDIO read/write
net: dsa: bcm_sf2: Do not clobber b53_switch_ops
bpf: change back to orig prog on too many passes
net: vrf: Add missing Rx counters
ipv4: Do not allow MAIN to be alias for new LOCAL w/ custom rules
igmp: Make igmp group member RFC 3376 compliant
flow_dissector: Update pptp handling to avoid null pointer deref.
drop_monitor: consider inserted data in genlmsg_end
drop_monitor: add missing call to genlmsg_end
net: ipv4: dst for local input routes should use l3mdev if relevant
net: fix incorrect original ingress device index in PKTINFO
rtnl: stats - add missing netlink message size checks
net/mlx5e: Disable netdev after close
net/mlx5e: Don't sync netdev state when not registered
net/mlx5: Prevent setting multicast macs for VFs
net/mlx5: Mask destination mac value in ethtool steering rules
net/mlx5: Avoid shadowing numa_node
net/mlx5: Cancel recovery work in remove flow
net/mlx5: Check FW limitations on log_max_qp before setting it
net/sched: cls_flower: Fix missing addr_type in classify
net: stmmac: Fix race between stmmac_drv_probe and stmmac_open
net, sched: fix soft lockup in tc_classify
ipv6: handle -EFAULT from skb_copy_bits
inet: fix IP(V6)_RECVORIGDSTADDR for udp sockets
sctp: sctp_transport_lookup_process should rcu_read_unlock when transport is null
net: vrf: Drop conntrack data after pass through VRF device on Tx
net: vrf: Fix NAT within a VRF
Linux 4.9.3
usb: gadget: composite: always set ep->mult to a sensible value
Revert "usb: gadget: composite: always set ep->mult to a sensible value"
Revert "rtlwifi: Fix enter/exit power_save"
tick/broadcast: Prevent NULL pointer dereference
clocksource/dummy_timer: Move hotplug callback after the real timers
xfs: fix max_retries _show and _store functions
xfs: fix crash and data corruption due to removal of busy COW extents
xfs: use the actual AG length when reserving blocks
xfs: fix double-cleanup when CUI recovery fails
xfs: use GPF_NOFS when allocating btree cursors
xfs: ignore leaf attr ichdr.count in verifier during log replay
xfs: don't cap maximum dedupe request length
xfs: don't allow di_size with high bit set
xfs: error out if trying to add attrs and anextents > 0
xfs: don't crash if reading a directory results in an unexpected hole
xfs: complain if we don't get nextents bmap records
xfs: check for bogus values in btree block headers
xfs: forbid AG btrees with level == 0
xfs: handle cow fork in xfs_bmap_trace_exlist
xfs: pass state not whichfork to trace_xfs_extlist
xfs: Move AGI buffer type setting to xfs_read_agi
xfs: pass post-eof speculative prealloc blocks to bmapi
xfs: use new extent lookup helpers xfs_file_iomap_begin_delay
xfs: clean up cow fork reservation and tag inodes correctly
xfs: use new extent lookup helpers in __xfs_reflink_reserve_cow
xfs: track preallocation separately in xfs_bmapi_reserve_delalloc()
xfs: remove prev argument to xfs_bmapi_reserve_delalloc
xfs: always succeed when deduping zero bytes
xfs: factor rmap btree size into the indlen calculations
xfs: new inode extent list lookup helpers
xfs: fix unbalanced inode reclaim flush locking
xfs: check minimum block size for CRC filesystems
xfs: provide helper for counting extents from if_bytes
xfs: don't BUG() on mixed direct and mapped I/O
xfs: don't skip cow forks w/ delalloc blocks in cowblocks scan
xfs: check return value of _trans_reserve_quota_nblks
xfs: don't call xfs_sb_quota_from_disk twice
tpm_tis: Check return values from get_burstcount.
drm/i915/gen9: fix the WM memory bandwidth WA for Y tiling cases
drm/i915/gen9: unconditionally apply the memory bandwidth WA
drm/i915: disable PSR by default on HSW/BDW
drm/radeon: Always store CRTC relative radeon_crtc->cursor_x/y values
s390/pci: fix dma address calculation in map_sg
s390/topology: always use s390 specific sched_domain_topology_level
powerpc/pci/rpadlpar: Fix device reference leaks
PCI: Enable access to non-standard VPD for Chelsio devices (cxgb3)
PCI: Support INTx masking on ConnectX-4 with firmware x.14.1100+
PCI: Convert Mellanox broken INTx quirks to be for listed devices only
PCI: Convert broken INTx masking quirks from HEADER to FINAL
PCI: Add Mellanox device IDs
PCI: rockchip: Correct the use of FTS mask
PCI: rockchip: Fix negotiated lanes calculation
staging: media: davinci_vpfe: unlock on error in vpfe_reqbufs()
f2fs: hide a maybe-uninitialized warning
f2fs: remove percpu_count due to performance regression
md: fix refcount problem on mddev when stopping array.
md: MD_RECOVERY_NEEDED is set for mddev->recovery
crypto: arm64/aes-ce - fix for big endian
crypto: arm64/aes-xts-ce: fix for big endian
crypto: arm64/sha1-ce - fix for big endian
crypto: arm64/aes-neon - fix for big endian
crypto: arm64/aes-ccm-ce: fix for big endian
crypto: arm/aes-ce - fix for big endian
crypto: arm64/ghash-ce - fix for big endian
crypto: arm64/sha2-ce - fix for big endian
s390/crypto: unlock on error in prng_tdes_read()
mm, compaction: fix NR_ISOLATED_* stats for pfn based migration
mm: khugepaged: fix radix tree node leak in shmem collapse error path
mm: khugepaged: close use-after-free race during shmem collapsing
docs-rst: fix LaTeX \DURole renewcommand with Sphinx 1.3+
mm/hugetlb.c: use the right pte val for compare in hugetlb_cow
rpmsg: qcom_smd: Correct return value for O_NONBLOCK
mmc: mmc_test: Uninitialized return value
genirq/affinity: Fix node generation from cpumask
PM / wakeirq: Fix dedicated wakeirq for drivers not using autosuspend
irqchip/bcm7038-l1: Implement irq_cpu_offline() callback
PCI/MSI: Check for NULL affinity mask in pci_irq_get_affinity()
ima: fix memory leak in ima_release_policy
relay: check array offset before using it
sbp-target: Fix second argument of percpu_ida_alloc()
target/iscsi: Fix double free in lio_target_tiqn_addtpg()
scsi: mvsas: fix command_active typo
scsi: g_NCR5380: Fix release_region in error handling
ASoC: samsung: i2s: Fixup last IRQ unsafe spin lock call
ASoC: Intel: Skylake: Fix a shift wrapping bug
ASoC: cht_bsw_rt5645: Fix leftover kmalloc
ASoC: lpass-platform: initialize dma channel number
iommu/vt-d: Flush old iommu caches for kdump when the device gets context mapped
iommu/vt-d: Fix pasid table size encoding
iommu/amd: Fix the left value check of cmd buffer
iommu/amd: Missing error code in amd_iommu_init_device()
clk: renesas: mstp: Support 8-bit registers for r7s72100
clk: imx31: fix rewritten input argument of mx31_clocks_init()
clk: sunxi-ng: sun8i-h3: Set CLK_SET_RATE_PARENT for audio module clocks
clk: sunxi-ng: sun8i-a23: Set CLK_SET_RATE_PARENT for audio module clocks
clk: ti: dra7: fix "failed to lookup clock node gmac_gmii_ref_clk_div" boot message
clk: clk-wm831x: fix a logic error
clk: qcom: ipq806x: Fix board clk rates
Input: synaptics-rmi4 - unlock on error
hwmon: (lm90) fix temp1_max_alarm attribute
hwmon: (g762) Fix overflows and crash seen when writing limit attributes
hwmon: (nct7802) Fix overflows seen when writing into limit attributes
hwmon: (ds620) Fix overflows seen when writing temperature limits
hwmon: (amc6821) sign extension temperature
hwmon: (scpi) Fix module autoload
platform/x86: fujitsu-laptop: use brightness_set_blocking for LED-setting callbacks
x86/cpu: Probe CPUID leaf 6 even when cpuid_level == 6
x86/prctl/uapi: Remove #ifdef for CHECKPOINT_RESTORE
debugfs: improve DEFINE_DEBUGFS_ATTRIBUTE for !CONFIG_DEBUG_FS
clk: renesas: cpg-mssr: Fix inverted debug check
efi/efivar_ssdt_load: Don't return success on allocation failure
cris: Only build flash rescue image if CONFIG_ETRAX_AXISFLASHMAP is selected
ath10k: use the right length of "background"
mfd: tps65217: Fix page fault on unloading modules
ath10k: fix failure to send NULL func frame for 10.4
nl80211: Use different attrs for BSSID and random MAC addr in scan req
mac80211: fix tid_agg_rx NULL dereference
drm/i915: tune down the fast link training vs boot fail
drm/i915/dp: add lane_count check in intel_dp_check_link_status
usb: dwc3: gadget: always unmap EP0 requests
usb: dwc3: ep0: explicitly call dwc3_ep0_prepare_one_trb()
usb: dwc3: ep0: add dwc3_ep0_prepare_one_trb()
iio: accel: st_accel: fix LIS3LV02 reading and scaling
staging: iio: ad7606: fix improper setting of oversampling pins
mei: move write cb to completion on credentials failures
mei: bus: fix mei_cldev_enable KDoc
mei: fix parameter rename KDoc
USB: serial: io_ti: bind to interface after fw download
dibusb: fix possible memory leak in dibusb_rc_query()
ARM: dts: sun7i: bananapi-m1-plus: Enable USB PHY for USB host support
arm64: dts: hip06: Correct hardware pin number of usb node
USB: phy: am335x-control: fix device and of_node leaks
ARM: dts: r8a7794: Correct hsusb parent clock
usb: gadget: fix request length error for isoc transfer
usb: gadget: Fix second argument of percpu_ida_alloc()
USB: serial: kl5kusb105: abort on open exception path
ALSA: usb-audio: Fix bogus error return in snd_usb_create_stream()
usb: musb: blackfin: add bfin_fifo_offset in bfin_ops
usb: gadget: udc: core: fix return code of usb_gadget_probe_driver()
usb: hub: Move hub_port_disable() to fix warning if PM is disabled
usb: musb: Fix trying to free already-free IRQ 4
usb: dwc3: gadget: Fix full speed mode
usb: dwc3: pci: Fix dr_mode misspelling
usb: dwc3: pci: add Intel Gemini Lake PCI ID
xhci: Fix race related to abort operation
xhci: Use delayed_work instead of timer for command timeout
usb: xhci-mem: use passed in GFP flags instead of GFP_KERNEL
USB: serial: mos7720: fix parallel probe
USB: serial: mos7720: fix parport use-after-free on probe errors
USB: serial: mos7720: fix use-after-free on probe errors
USB: serial: mos7720: fix NULL-deref at open
USB: serial: mos7840: fix NULL-deref at open
USB: serial: kobil_sct: fix NULL-deref in write
USB: serial: cyberjack: fix NULL-deref at open
USB: serial: oti6858: fix NULL-deref at open
USB: serial: io_edgeport: fix NULL-deref at open
USB: serial: ti_usb_3410_5052: fix NULL-deref at open
USB: serial: garmin_gps: fix memory leak on failed URB submit
USB: serial: iuu_phoenix: fix NULL-deref at open
USB: serial: io_ti: fix I/O after disconnect
USB: serial: io_ti: fix another NULL-deref at open
USB: serial: io_ti: fix NULL-deref at open
USB: serial: spcp8x5: fix NULL-deref at open
USB: serial: keyspan_pda: verify endpoints at probe
USB: serial: pl2303: fix NULL-deref at open
USB: serial: quatech2: fix sleep-while-atomic in close
USB: serial: omninet: fix NULL-derefs at open and disconnect
usb: return error code when platform_get_irq fails
usb: xhci: hold lock over xhci_abort_cmd_ring()
xhci: Handle command completion and timeout race
usb: host: xhci: Fix possible wild pointer when handling abort command
usb: xhci: fix return value of xhci_setup_device()
xhci: free xhci virtual devices with leaf nodes first
usb: xhci: apply XHCI_PME_STUCK_QUIRK to Intel Apollo Lake
usb: xhci: fix possible wild pointer
usb: dwc3: core: avoid Overflow events
usb: gadget: composite: Test get_alt() presence instead of set_alt()
USB: dummy-hcd: fix bug in stop_activity (handle ep0)
USB: fix problems with duplicate endpoint addresses
USB: gadgetfs: fix checks of wTotalLength in config descriptors
USB: gadgetfs: fix use-after-free bug
USB: gadgetfs: fix unbounded memory allocation bug
usb: gadgetfs: restrict upper bound on device configuration size
usb: storage: unusual_uas: Add JMicron JMS56x to unusual device
usb: musb: dsps: implement clear_ep_rxintr() callback
usb: musb: core: add clear_ep_rxintr() to musb_platform_ops
KVM: MIPS: Flush KVM entry code from icache globally
KVM: MIPS: Don't clobber CP0_Status.UX
KVM: x86: reset MMU on KVM_SET_VCPU_EVENTS
drm/i915: Initialize overlay->last_flip properly
drm/i915: Force VDD off on the new power seqeuencer before starting to use it
drm/i915: Fix oops in overlay due to frontbuffer tracking
drm/i915: Fix oopses in the overlay code due to i915_gem_active stuff
gcc-plugins: update gcc-common.h for gcc-7
asm-prototypes: Clear any CPP defines before declaring the functions
mac80211: initialize fast-xmit 'info' later
pinctrl/amd: Set the level based on ACPI tables
ARM: davinci: da850: don't add emac clock to lookup table twice
HID: sensor-hub: Move the memset to sensor_hub_get_feature()
parisc: Mark cr16 clocksource unstable on SMP systems
parisc: Add line-break when printing segfault info
fscrypt: fix renaming and linking special files
ALSA: usb-audio: Fix irq/process data synchronization
ALSA: hda - Apply asus-mode8 fixup to ASUS X71SL
ALSA: hda - Fix up GPIO for ASUS ROG Ranger
staging: octeon: Call SET_NETDEV_DEV()
iio: bmi160: Fix time needed to sleep after command execution
iio: max44000: correct value in illuminance_integration_time_available
iio: common: st_sensors: fix channel data parsing
Linux 4.9.2
drm/i915: Fix setting of boost freq tunable
drm/i915: skip the first 4k of stolen memory on everything >= gen8
drm/i915: Initialize dev_priv->atomic_cdclk_freq at init time
drm/i915: Fix cdclk vs. dev_cdclk mess when not recomputing things
drm/i915/dsi: Do not clear DPOUNIT_CLOCK_GATE_DISABLE from vlv_init_display_clock_gating
drm/i915/dsi: Fix chv_exec_gpio disabling the GPIOs it is setting
net: mvpp2: fix dma unmapping of TX buffers for fragments
sg_write()/bsg_write() is not fit to be called under KERNEL_DS
kconfig/nconf: Fix hang when editing symbol with a long prompt
target/user: Fix use-after-free of tcmu_cmds if they are expired
libnvdimm, pfn: fix align attribute
of, numa: Return NUMA_NO_NODE from disable of_node_to_nid() if nid not possible.
powerpc/boot: Request no dynamic linker for boot wrapper
powerpc/ps3: Fix system hang with GCC 5 builds
powerpc/64e: Convert cmpi to cmpwi in head_64.S
SUNRPC: fix refcounting problems with auth_gss messages.
pNFS: Fix a deadlock between read resends and layoutreturn
pNFS: Clear NFS_LAYOUT_RETURN_REQUESTED when invalidating the layout stateid
pNFS: Don't clear the layout stateid if a layout return is outstanding
pNFS: On error, do not send LAYOUTGET until the LAYOUTRETURN has completed
nfs_write_end(): fix handling of short copies
libceph: verify authorize reply on connect
PCI: Check for PME in targeted sleep state
i40iw: Use correct src address in memcpy to rdma stats counters
bad_inode: add missing i_op initializers
Input: drv260x - fix input device's parent assignment
v4l: tvp5150: Add missing break in set control handler
media: solo6x10: fix lockup by avoiding delayed register write
s5p-mfc: fix failure path of s5p_mfc_alloc_memdev()
mn88473: fix chip id check on probe
mn88472: fix chip id check on probe
IB/cma: Fix a race condition in iboe_addr_get_sgid()
IB/rxe: Fix a memory leak in rxe_qp_cleanup()
IB/multicast: Check ib_find_pkey() return value
IPoIB: Avoid reading an uninitialized member variable
IB/mad: Fix an array index check
fgraph: Handle a case where a tracer ignores set_graph_notrace
x86/smpboot: Make logical package management more robust
platform/x86: asus-nb-wmi.c: Add X45U quirk
ftrace/x86_32: Set ftrace_stub to weak to prevent gcc from using short jumps to it
vsock/virtio: fix src/dst cid format
fsnotify: Fix possible use-after-free in inode iteration on umount
kvm: nVMX: Allow L1 to intercept software exceptions (#BP and #OF)
KVM: PPC: Book3S HV: Don't lose hardware R/C bit updates in H_PROTECT
KVM: PPC: Book3S HV: Save/restore XER in checkpointed register state
scsi: aacraid: remove wildcard for series 9 controllers
md/raid5: limit request size according to implementation limits
sc16is7xx: Drop bogus use of IRQF_ONESHOT
latent_entropy: fix ARM build error on earlier gcc
arm64: KVM: pmu: Reset PMSELR_EL0.SEL to a sane value before entering the guest
s390/kexec: use node 0 when re-adding crash kernel memory
s390/vmlogrdr: fix IUCV buffer allocation
firmware: fix usermode helper fallback loading
ARC: mm: arc700: Don't assume 2 colours for aliasing VIPT dcache
scsi: avoid a permanent stop of the scsi device's request queue
scsi: zfcp: fix rport unblock race with LUN recovery
scsi: zfcp: do not trace pure benign residual HBA responses at default level
scsi: zfcp: fix use-after-"free" in FC ingress path after TMF
iscsi-target: Return error if unable to add network portal
scsi: megaraid_sas: Do not set MPI2_TYPE_CUDA for JBOD FP path for FW which does not support JBOD sequence map
scsi: megaraid_sas: For SRIOV enabled firmware, ensure VF driver waits for 30secs before reset
stm class: Fix device leak in open error path
vt: fix Scroll Lock LED trigger name
block: protect iterate_bdevs() against concurrent close
mei: me: add lewisburg device ids
mei: request async autosuspend at the end of enumeration
drivers/gpu/drm/ast: Fix infinite loop if read fails
drm/amdgpu: fix init save/restore list in gfx_v8.0
drm/amdgpu: fix enable_cp_power_gating in gfx_v8.0.
drm/amd/powerplay: bypass fan table setup if no fan connected
drm/gma500: Add compat ioctl
drm/radeon/si: load the proper firmware on 0x87 oland boards
drm/radeon: add additional pci revision to dpm workaround
drm/radeon: Hide the HW cursor while it's out of bounds
drm/radeon: Also call cursor_move_locked when the cursor size changes
drm/nouveau/fifo/gf100-: protect channel preempt with subdev mutex
drm/nouveau/i2c/gk110b,gm10x: use the correct implementation
drm/nouveau/ttm: wait for bo fence to signal before unmapping vmas
drm/nouveau/ltc: protect clearing of comptags with mutex
drm/nouveau/bios: require checksum to match for fast acpi shadow method
drm/nouveau/kms: lvds panel strap moved again on maxwell
drm/nouveau/gr: fallback to legacy paths during firmware lookup
drm/amd/amdgpu: enable GUI idle INT after enabling CGCG
drm/amdgpu: Also call cursor_move_locked when the cursor size changes
drm/amdgpu: Store CRTC relative amdgpu_crtc->cursor_x/y values
drm/amdgpu: add additional pci revision to dpm workaround
drm/amdgpu/si: load the proper firmware on 0x87 oland boards
ACPI / video: Add force_native quirk for HP Pavilion dv6
ACPI / video: Add force_native quirk for Dell XPS 17 L702X
staging: comedi: ni_mio_common: fix E series ni_ai_insn_read() data
staging: comedi: ni_mio_common: fix M Series ni_ai_insn_read() data mask
staging: lustre: ldlm: pl_recalc time handling is wrong
staging/lustre/osc: Revert erroneous list_for_each_entry_safe use
hv: acquire vmbus_connection.channel_mutex in vmbus_free_channels()
docs: sphinx-extensions: make rstFlatTable work with docutils 0.13
thermal: hwmon: Properly report critical temperature in sysfs
clk: bcm2835: Avoid overwriting the div info when disabling a pll_div clk
arm64: tegra: Add VDD_GPU regulator to Jetson TX1
gpio: chardev: Return error for seek operations
gpio: stmpe: fix interrupt handling bug
timekeeping_Force_unsigned_clocksource_to_nanoseconds_conversion
mmc: sd: Meet alignment requirements for raw_ssr DMA
regulator: stw481x-vmmc: fix ages old enable error
mmc: sdhci: Fix recovery from tuning timeout
Revert "mmc: sdhci: Reset cmd and data circuits after tuning failure"
ath9k: do not return early to fix rcu unlocking
ath9k: Really fix LED polarity for some Mini PCI AR9220 MB92 cards.
ath9k: fix ath9k_hw_gpio_get() to return 0 or 1 on success
cfg80211/mac80211: fix BSS leaks when abandoning assoc attempts
rtl8xxxu: Work around issue with 8192eu and 8723bu devices not reconnecting
perf/x86/intel/cstate: Prevent hotplug callback leak
perf annotate: Don't throw error for zero length symbols
perf/x86: Fix exclusion of BTS and LBR for Goldmont
rtlwifi: Fix enter/exit power_save
ath10k: fix soft lockup during firmware crash/hw-restart
ssb: Fix error routine when fallback SPROM fails
Linux 4.9.1
x86/kbuild: enable modversions for symbols exported from asm
builddeb: fix cross-building to arm64 producing host-arch debs
xfs: set AGI buffer type in xlog_recover_clear_agi_bucket
xfs: fix up xfs_swap_extent_forks inline extent handling
arm/xen: Use alloc_percpu rather than __alloc_percpu
xen/gntdev: Use VM_MIXEDMAP instead of VM_IO to avoid NUMA balancing
tpm xen: Remove bogus tpm_chip_unregister
kernel/debug/debug_core.c: more properly delay for secondary CPUs
watchdog: qcom: fix kernel panic due to external abort on non-linefetch
watchdog: mei_wdt: request stop on reboot to prevent false positive event
kernel/watchdog: use nmi registers snapshot in hardlockup handler
CIFS: Fix a possible memory corruption in push locks
CIFS: Decrease verbosity of ioctl call
CIFS: Fix a possible double locking of mutex during reconnect
CIFS: Fix missing nls unload in smb2_reconnect()
CIFS: Fix a possible memory corruption during reconnect
cifs: Fix smbencrypt() to stop pointing a scatterlist at the stack
ASoC: intel: Fix crash at suspend/resume without card registration
dm space map metadata: fix 'struct sm_metadata' leak on failed create
dm raid: fix discard support regression
dm rq: fix a race condition in rq_completed()
dm crypt: mark key as invalid until properly loaded
dm flakey: return -EINVAL on interval bounds error in flakey_ctr()
dm table: an 'all_blk_mq' table must be loaded for a blk-mq DM device
dm table: fix 'all_blk_mq' inconsistency when an empty table is loaded
blk-mq: Do not invoke .queue_rq() for a stopped queue
PM / OPP: Don't use OPP structure outside of rcu protected section
PM / OPP: Pass opp_table to dev_pm_opp_put_regulator()
usb: gadget: composite: always set ep->mult to a sensible value
mm, page_alloc: keep pcp count and list contents in sync if struct page is corrupted
mm/vmscan.c: set correct defer count for shrinker
nvmet: Fix possible infinite loop triggered on hot namespace removal
loop: return proper error from loop_queue_rq()
f2fs: fix to determine start_cp_addr by sbi->cur_cp_pack
f2fs: fix overflow due to condition check order
f2fs: set ->owner for debugfs status file's file_operations
Revert "f2fs: use percpu_counter for # of dirty pages in inode"
ext4: do not perform data journaling when data is encrypted
ext4: return -ENOMEM instead of success
ext4: reject inodes with negative size
ext4: add sanity checking to count_overhead()
ext4: fix in-superblock mount options processing
ext4: use more strict checks for inodes_per_block on mount
ext4: fix stack memory corruption with 64k block size
ext4: fix mballoc breakage with 64k block size
ext4: don't lock buffer in ext4_commit_super if holding spinlock
crypto: caam - fix AEAD givenc descriptors
ptrace: Don't allow accessing an undumpable mm
ptrace: Capture the ptracer's creds not PT_PTRACE_CAP
vfs,mm: fix return value of read() at s_maxbytes
mm: Add a user_ns owner to mm_struct and fix ptrace permission checks
block_dev: don't test bdev->bd_contains when it is not stable
splice: reinstate SIGPIPE/EPIPE handling
fs: exec: apply CLOEXEC before changing dumpable task flags
exec: Ensure mm->user_ns contains the execed files
clk: ti: omap36xx: Work around sprz319 advisory 2.1
ALSA: hda: when comparing pin configurations, ignore assoc in addition to seq
ALSA: hda - Gate the mic jack on HP Z1 Gen3 AiO
ALSA: hda - fix headset-mic problem on a Dell laptop
ALSA: hda - ignore the assoc and seq when comparing pin configurations
ALSA: hda/ca0132 - Add quirk for Alienware 15 R2 2016
ALSA: hiface: Fix M2Tech hiFace driver sampling rate change
ALSA: usb-audio: Add QuickCam Communicate Deluxe/S7500 to volume_control_quirks
usbip: vudc: fix: Clear already_seen flag also for ep0
USB: UHCI: report non-PME wakeup signalling for Intel hardware
usb: gadget: composite: correctly initialize ep->maxpacket
usb: gadget: f_uac2: fix error handling at afunc_bind
usb: core: usbport: Use proper LED API to fix potential crash
usb: hub: Fix auto-remount of safely removed or ejected USB-3 devices
usb: dwc3: gadget: set PCM1 field of isochronous-first TRBs
USB: cdc-acm: add device id for GW Instek AFG-125
USB: serial: kl5kusb105: fix open error path
USB: serial: option: add dlink dwm-158
USB: serial: option: add support for Telit LE922A PIDs 0x1040, 0x1041
Btrfs: fix qgroup rescan worker initialization
Btrfs: fix emptiness check for dirtied extent buffers at check_leaf()
btrfs: store and load values of stripes_min/stripes_max in balance status item
Btrfs: fix relocation incorrectly dropping data references
Btrfs: fix tree search logic when replaying directory entry deletes
Btrfs: fix deadlock caused by fsync when logging directory entries
Btrfs: fix BUG_ON in btrfs_mark_buffer_dirty
btrfs: limit async_work allocation and worker func duration
hotplug: Make register and unregister notifier API symmetric
ANDROID: Shrink ashmem directly through shmem_fallocate
ANDROID: ashmem: Add shmem_set_file to mm/shmem.c
ANDROID: Add android config documentation to boot framework.
Conflicts:
Documentation/android.txt
Documentation/cpu-freq/governors.txt
Documentation/networking/ip-sysctl.txt
arch/arm64/boot/Makefile
arch/arm64/kernel/process.c
arch/arm64/kernel/topology.c
arch/arm64/mm/dma-mapping.c
drivers/cpufreq/cpufreq_interactive.c
drivers/input/misc/gpio_matrix.c
drivers/md/Kconfig
drivers/md/dm-android-verity.c
drivers/md/dm-android-verity.h
drivers/md/dm-linear.c
drivers/md/dm-verity-fec.h
drivers/mmc/core/host.c
drivers/platform/goldfish/goldfish_pipe.c
drivers/power/supply/power_supply_sysfs.c
drivers/scsi/ufs/ufshcd.c
drivers/scsi/ufs/ufshcd.h
drivers/staging/android/fiq_debugger/fiq_debugger.c
drivers/staging/android/lowmemorykiller.c
drivers/usb/dwc3/core.h
drivers/usb/gadget/Kconfig
drivers/usb/gadget/configfs.c
drivers/usb/gadget/function/Makefile
drivers/usb/gadget/function/f_accessory.c
drivers/usb/gadget/function/f_audio_source.c
drivers/usb/gadget/function/f_fs.c
drivers/usb/gadget/function/f_mtp.c
drivers/usb/phy/Kconfig
drivers/usb/phy/otg-wakelock.c
fs/ext4/inode.c
fs/ext4/ioctl.c
fs/f2fs/data.c
fs/proc/task_mmu.c
fs/sdcardfs/derived_perm.c
fs/sdcardfs/file.c
fs/sdcardfs/inode.c
fs/sdcardfs/lookup.c
fs/sdcardfs/main.c
fs/sdcardfs/packagelist.c
fs/sdcardfs/sdcardfs.h
fs/sdcardfs/super.c
include/linux/mmc/core.h
include/linux/power_supply.h
include/linux/sched.h
include/linux/usb/f_mtp.h
include/net/fib_rules.h
include/net/route.h
include/trace/events/cpufreq_interactive.h
include/trace/events/power.h
include/trace/events/sched.h
include/uapi/linux/magic.h
include/uapi/linux/prctl.h
init/Kconfig
kernel/cgroup.c
kernel/configs/android-base.config
kernel/power/process.c
kernel/sched/Makefile
kernel/sched/core.c
kernel/sched/cputime.c
kernel/sched/fair.c
kernel/sched/rt.c
kernel/sched/sched.h
kernel/sched/stop_task.c
kernel/sched/tune.c
lib/Kconfig.debug
net/core/fib_rules.c
net/ipv4/inet_connection_sock.c
net/ipv4/ping.c
net/ipv4/raw.c
net/ipv4/route.c
net/ipv4/syncookies.c
net/ipv4/udp.c
net/ipv6/af_inet6.c
net/ipv6/ah6.c
net/ipv6/datagram.c
net/ipv6/esp6.c
net/ipv6/icmp.c
net/ipv6/inet6_connection_sock.c
net/ipv6/ip6_vti.c
net/ipv6/ipcomp6.c
net/ipv6/ping.c
net/ipv6/raw.c
net/ipv6/route.c
net/ipv6/syncookies.c
net/ipv6/tcp_ipv6.c
net/ipv6/udp.c
Change-Id: I82455dc7b564fc5d3ad2b784a0eb8f96c6b05d4c
Signed-off-by: Channagoud Kadabi <ckadabi@codeaurora.org>
|
||
|
Pavankumar Kondeti
|
a4908e9b5f |
sysctl: define upper limit for sched_freq_reporting_policy
Setting sched_freq_reporting_policy tunable to an unsupported values results in a warning from the scheduler. The previous policy setting is also lost. As sched_freq_reporting_policy can not be set to an incorrect value now, remove the WARN_ON_ONCE from the scheduler. Change-Id: I58d7e5dfefb7d11d2309bc05a1dd66acdc11b766 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> [rameezmustafa@codeaurora.org: Port to msm-4.8.] Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org> |
||
|
Vikram Mulukutla
|
40c30b2591 |
sched: hmp: Remove the global sysctl_sched_enable_colocation tunable
Colocation in HMP includes a tunable that turns on or off the feature globally across all colocation groups. Supporting this tunable correctly would result in complexity that would outweigh any foreseeable benefits. For example, disabling the feature globally would involve deleting all colocation groups one by one while ensuring no placement decisions are made during the process. Remove the tunable. Adding or removing a task from a colocation group is still possible and so we're not losing functionality. Change-Id: I4cb8bcdbee98d3bdd168baacbac345eca9ea8879 Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org> |
||
|
Syed Rameez Mustafa
|
e343972d8e |
sched: Convert the global wake_up_idle flag to a per cluster flag
Since clusters can vary significantly in the power and performance characteristics, there may be a need to have different CPU selection policies based on which cluster a task is being placed on. For example the placement policy can be more aggressive in using idle CPUs on cluster that are power efficient and less aggressive on clusters that are geared towards performance. Add support for per cluster wake_up_idle flag to allow greater flexibility in placement policies. Change-Id: I18cd3d907cd965db03a13f4655870dc10c07acfe Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org> |
||
|
Srivatsa Vaddagiri
|
b8b701353b |
sched: Avoid packing tasks with low sleep time
Low sleep time can be an indication that waking tasks will not receive any vruntime bonus and hence would suffer from latency when packed. short-burst tasks sleeping on an average more than sched_short_sleep_ns are not eligible for packing. This policy covers the case where a task runs in short bursts and sleeping for smaller duration in between. Change-Id: Ib81fa37809b85c267949cd433bc6115dd89f100e Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org> Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Srivatsa Vaddagiri
|
b36e661e52 |
sched: Avoid waking idle cpu for short-burst tasks
Introduce sched_short_burst tunable to classify "short-burst" tasks. These tasks are eligible for packing to avoid overhead associated with waking up an idle CPU. select_best_cpu() ignores power-cost and selects the CPU with least wakeup latency which is not loaded with IRQs and can accommodate this task without exceeding spill limits. The ties are broken with load followed by previous CPU. This policy does not affect cluster selection but only CPU selection in the selected cluster. The tasks eligible for "wakeup-up-idle" and "boost" are not considered for packing. This policy is applied for both "fair" and "rt" scheduling class tasks. Change-Id: I2a05493fde93f58636725f18d0ce8dbce4418a30 Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org> Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Syed Rameez Mustafa
|
084075baab |
sched/hmp: Enhance co-location and scheduler boost features
The recent introduction of the schedtune cgroup controller has provided
the scheduler with added flexibility in terms of some of it's placement
features. In particular each cgroup under the schedtune controller can
now specify:
1) Whether it needs co-location along with other cgroups
2) Whether it is eligible for scheduler boost (sched_boost_enabled)
3) Whether the kernel can override the boost eligibility when necessary
(sched_boost_no_override)
The scheduler now creates a reserved co-location group at boot. This
group is used to co-locate all tasks that form part of any one of the
cgroups that have co-location enabled. This reserved group can neither
be destroyed nor reused for other purposes. Furthermore, cgroups are
only allowed to indicate their co-location preference once at boot.
Further updates are disallowed.
Since we are now creating co-location groups for an extended period of
time, there are a few other factors to consider when determining the
preferred cluster for the group. We first exclude any tasks in the
group that have not been observed to be running for a significant
amount of time. Secondly we introduce the notion of group up and down
migrate tunables to allow different migration policies than individual
tasks. Lastly we break co-location if a single task in a group exceeds
up-migrate but the total load of the group does not exceed group
up-migrate.
In terms of sched_boost, the scheduler now supports multiple types of
boost. These are:
1) FULL_THROTTLE : Force up-migrate tasks belonging any cgroup that
has the sched_boost_enabled flag turned on. Little
CPUs will only be used when big CPUs can no longer
accommodate tasks. Also up-migrate all RT tasks.
2) CONSERVATIVE : Override the sched_boost_enabled flag for all cgroups
except those that have the sched_boost_no_override
flag set. Force up-migrate all tasks belonging to only
those cgroups that still remain eligible for boost.
RT tasks do not get force up migrated.
3) RESTRAINED : Start frequency aggregation for co-located tasks. This
type of boost does not force up-migrate any task.
Finally the boost API removes ref-counting. This means that there can
only be a single entity using boost at any given time. If multiple
entities are managing boost, they are required to be well behaved so
that they don't interfere with one another. Even for a single client,
it is not possible to switch directly from one boost type to another.
Boost must be first turned off before switching over to a new type.
Change-Id: I8d224a70cbef162f27078b62b73acaa22670861d
Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org>
|
||
|
Olav Haugan
|
651132f4ad |
sched/core_ctl: Move header file to global location
Move the header file of core control to the standard linux include directory to allow other entities to include this file. Change-Id: I2ddb8b3b96063be3c6a6cb6bc333998e007f9de7 Signed-off-by: Olav Haugan <ohaugan@codeaurora.org> |
||
|
Syed Rameez Mustafa
|
cbf0804d8a |
sched: Add multiple load reporting policies for cpu frequency
The previous patches in this series introduce the mechanics of CPU load tracking without fixups for intra cluster migration and top task load tracking. Add a tunable that dictates what of the above needs to be considered when reporting load to the governor. The default policy is to take the maximum of the CPU load and top task load. Change-Id: Ie585a11ed774b929910d04c41471db3a2a102ec5 Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org> |
||
|
Patrick Bellasi
|
62c1c06921 |
sched/tune: add sysctl interface to define a boost value
The current (CFS) scheduler implementation does not allow "to boost"
tasks performance by running them at a higher OPP compared to the
minimum required to meet their workload demands.
To support tasks performance boosting the scheduler should provide a
"knob" which allows to tune how much the system is going to be optimised
for energy efficiency vs performance.
This patch is the first of a series which provides a simple interface to
define a tuning knob. One system-wide "boost" tunable is exposed via:
/proc/sys/kernel/sched_cfs_boost
which can be configured in the range [0..100], to define a percentage
where:
- 0% boost requires to operate in "standard" mode by scheduling
tasks at the minimum capacities required by the workload demand
- 100% boost requires to push at maximum the task performances,
"regardless" of the incurred energy consumption
A boost value in between these two boundaries is used to bias the
power/performance trade-off, the higher the boost value the more the
scheduler is biased toward performance boosting instead of energy
efficiency.
Change-Id: I59a41725e2d8f9238a61dfb0c909071b53560fc0
cc: Ingo Molnar <mingo@redhat.com>
cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Git-commit: 63c8fad2b06805ef88f1220551289f0a3c3529f1
Git-repo: https://source.codeaurora.org/quic/la/kernel/msm-4.4
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
|
||
|
Pavankumar Kondeti
|
72b49a3988 |
sched: add a knob to prefer the waker CPU for sync wakeups
The current policy has a preference to select an idle CPU in the waker cluster compared to the waker CPU running only 1 task. By selecting an idle CPU, it eliminates the chance of waker migrating to a different CPU after the wakee preempts it. This policy is also not susceptible to the incorrect "sync" usage i.e the waker does not goto sleep after waking up the wakee. However LPM exit latency associated with an idle CPU outweigh the above benefits on some targets. So add a knob to prefer the waker CPU having only 1 runnable task over idle CPUs in the waker cluster. Change-Id: Id974748c07625c1b19112235f426a5d204dfdb33 Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org> |
||
|
Srinath Sridharan
|
3a73c96a28 |
ANDROID: sched/walt: Accounting for number of irqs pending on each core
Schedules on a core whose irq count is less than a threshold. Improves I/O performance of EAS. Change-Id: I08ff7dd0d22502a0106fc636b1af2e6fe9e758b5 Signed-off-by: Andres Oportus <andresoportus@google.com> |
||
|
Srivatsa Vaddagiri
|
26c2154816 |
ANDROID: sched: Introduce Window Assisted Load Tracking (WALT)
use a window based view of time in order to track task
demand and CPU utilization in the scheduler.
Window Assisted Load Tracking (WALT) implementation credits:
Srivatsa Vaddagiri, Steve Muckle, Syed Rameez Mustafa, Joonwoo Park,
Pavan Kumar Kondeti, Olav Haugan
2016-03-06: Integration with EAS/refactoring by Vikram Mulukutla
and Todd Kjos
Change-Id: I21408236836625d4e7d7de1843d20ed5ff36c708
Includes fixes for issues:
eas/walt: Use walt_ktime_clock() instead of ktime_get_ns() to avoid a
race resulting in watchdog resets
BUG: 29353986
Change-Id: Ic1820e22a136f7c7ebd6f42e15f14d470f6bbbdb
Handle walt accounting anomoly during resume
During resume, there is a corner case where on wakeup, a task's
prev_runnable_sum can go negative. This is a workaround that
fixes the condition and warns (instead of crashing).
BUG: 29464099
Change-Id: I173e7874324b31a3584435530281708145773508
Signed-off-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Srinath Sridharan <srinathsr@google.com>
Signed-off-by: Juri Lelli <juri.lelli@arm.com>
[jstultz: fwdported to 4.4]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Andres Oportus <andresoportus@google.com>
|
||
|
Todd Kjos
|
c6a6f3bfa0 |
ANDROID: sched/fair: add tunable to set initial task load
The choice of initial task load upon fork has a large influence on CPU and OPP selection when scheduler-driven DVFS is in use. Make this tuneable by adding a new sysctl "sched_initial_task_util". If the sched governor is not used, the default remains at SCHED_LOAD_SCALE Otherwise, the value from the sysctl is used. This defaults to 0. Signed-off-by: "Todd Kjos <tkjos@google.com>" Signed-off-by: Andres Oportus <andresoportus@google.com> |
||
|
Juri Lelli
|
1931b93dba |
ANDROID: sched/fair: add tunable to force selection at cpu granularity
EAS assumes that clusters with smaller capacity cores are more energy-efficient. This may not be true on non-big-little devices, so EAS can make incorrect cluster selections when finding a CPU to wake. The "sched_is_big_little" hint can be used to cause a cpu-based selection instead of cluster-based selection. This change incorporates the addition of the sync hint enable patch EAS did not honour synchronous wakeup hints, a new sysctl is created to ask EAS to use this information when selecting a CPU. The control is called "sched_sync_hint_enable". Also contains: EAS: sched/fair: for SMP bias toward idle core with capacity For SMP devices, on wakeup bias towards idle cores that have capacity vs busy devices that need a higher OPP eas: favor idle cpus for boosted tasks BUG: 29533997 BUG: 29512132 Change-Id: I0cc9a1b1b88fb52916f18bf2d25715bdc3634f9c Signed-off-by: Juri Lelli <juri.lelli@arm.com> Signed-off-by: Srinath Sridharan <srinathsr@google.com> eas/sched/fair: Favoring busy cpus with low OPPs BUG: 29533997 BUG: 29512132 Change-Id: I9305b3239698d64278db715a2e277ea0bb4ece79 Signed-off-by: Juri Lelli <juri.lelli@arm.com> Signed-off-by: Andres Oportus <andresoportus@google.com> |
||
|
Srinath Sridharan
|
bf47bdd180 |
ANDROID: sched: EAS: take cstate into account when selecting idle core
Introduce a new sysctl for this option, 'sched_cstate_aware'. When this is enabled, select_idle_sibling in CFS is modified to choose the idle CPU in the sibling group which has the lowest idle state index - idle state indexes are assumed to increase as sleep depth and hence wakeup latency increase. In this way, we attempt to minimise wakeup latency when an idle CPU is required. Signed-off-by: Srinath Sridharan <srinathsr@google.com> Includes: sched: EAS: fix select_idle_sibling when sysctl_sched_cstate_aware is enabled, best_idle cpu will not be chosen in the original flow because it will goto done directly Bug: 30107557 Change-Id: Ie09c2e3960cafbb976f8d472747faefab3b4d6ac Signed-off-by: martin_liu <martin_liu@htc.com> Signed-off-by: Andres Oportus <andresoportus@google.com> |
||
|
Patrick Bellasi
|
69fa4c768a |
ANDROID: sched/tune: add sysctl interface to define a boost value
The current (CFS) scheduler implementation does not allow "to boost"
tasks performance by running them at a higher OPP compared to the
minimum required to meet their workload demands.
To support tasks performance boosting the scheduler should provide a
"knob" which allows to tune how much the system is going to be optimised
for energy efficiency vs performance.
This patch is the first of a series which provides a simple interface to
define a tuning knob. One system-wide "boost" tunable is exposed via:
/proc/sys/kernel/sched_cfs_boost
which can be configured in the range [0..100], to define a percentage
where:
- 0% boost requires to operate in "standard" mode by scheduling
tasks at the minimum capacities required by the workload demand
- 100% boost requires to push at maximum the task performances,
"regardless" of the incurred energy consumption
A boost value in between these two boundaries is used to bias the
power/performance trade-off, the higher the boost value the more the
scheduler is biased toward performance boosting instead of energy
efficiency.
cc: Ingo Molnar <mingo@redhat.com>
cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Andres Oportus <andresoportus@google.com>
|
||
|
Syed Rameez Mustafa
|
dddcab78bd |
sched: Introduce HMP scheduling
Add a snapshot of all changes associated with the HMP scheduler ported over to the latest kernel version. This will allow proper big Little scheduling for targets that will use this kernel version. The msm-4-4 snapshot is taken at commitcc60f0790f27 (sched: constrain HMP scheduler tunable range with in better way). Please note that this is not an exact snapshot; certain HMP related code has been removed. This includes redundant code, useless CONFIG options, rarely taken error code paths, runtime enable/disable support for the HMP scheduler. Change-Id: I923f35904008f1fb74ac9c1a39481810cdea5790 Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org> |
||
|
Andrey Ryabinin
|
ea606cf5d8 |
mm: move max_map_count bits into mm.h
max_map_count sysctl unrelated to scheduler. Move its bits from include/linux/sched/sysctl.h to include/linux/mm.h. Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
|
Mel Gorman
|
cb2517653f |
sched/debug: Make schedstats a runtime tunable that is disabled by default
schedstats is very useful during debugging and performance tuning but it
incurs overhead to calculate the stats. As such, even though it can be
disabled at build time, it is often enabled as the information is useful.
This patch adds a kernel command-line and sysctl tunable to enable or
disable schedstats on demand (when it's built in). It is disabled
by default as someone who knows they need it can also learn to enable
it when necessary.
The benefits are dependent on how scheduler-intensive the workload is.
If it is then the patch reduces the number of cycles spent calculating
the stats with a small benefit from reducing the cache footprint of the
scheduler.
These measurements were taken from a 48-core 2-socket
machine with Xeon(R) E5-2670 v3 cpus although they were also tested on a
single socket machine 8-core machine with Intel i7-3770 processors.
netperf-tcp
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Hmean 64 560.45 ( 0.00%) 575.98 ( 2.77%)
Hmean 128 766.66 ( 0.00%) 795.79 ( 3.80%)
Hmean 256 950.51 ( 0.00%) 981.50 ( 3.26%)
Hmean 1024 1433.25 ( 0.00%) 1466.51 ( 2.32%)
Hmean 2048 2810.54 ( 0.00%) 2879.75 ( 2.46%)
Hmean 3312 4618.18 ( 0.00%) 4682.09 ( 1.38%)
Hmean 4096 5306.42 ( 0.00%) 5346.39 ( 0.75%)
Hmean 8192 10581.44 ( 0.00%) 10698.15 ( 1.10%)
Hmean 16384 18857.70 ( 0.00%) 18937.61 ( 0.42%)
Small gains here, UDP_STREAM showed nothing intresting and neither did
the TCP_RR tests. The gains on the 8-core machine were very similar.
tbench4
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Hmean mb/sec-1 500.85 ( 0.00%) 522.43 ( 4.31%)
Hmean mb/sec-2 984.66 ( 0.00%) 1018.19 ( 3.41%)
Hmean mb/sec-4 1827.91 ( 0.00%) 1847.78 ( 1.09%)
Hmean mb/sec-8 3561.36 ( 0.00%) 3611.28 ( 1.40%)
Hmean mb/sec-16 5824.52 ( 0.00%) 5929.03 ( 1.79%)
Hmean mb/sec-32 10943.10 ( 0.00%) 10802.83 ( -1.28%)
Hmean mb/sec-64 15950.81 ( 0.00%) 16211.31 ( 1.63%)
Hmean mb/sec-128 15302.17 ( 0.00%) 15445.11 ( 0.93%)
Hmean mb/sec-256 14866.18 ( 0.00%) 15088.73 ( 1.50%)
Hmean mb/sec-512 15223.31 ( 0.00%) 15373.69 ( 0.99%)
Hmean mb/sec-1024 14574.25 ( 0.00%) 14598.02 ( 0.16%)
Hmean mb/sec-2048 13569.02 ( 0.00%) 13733.86 ( 1.21%)
Hmean mb/sec-3072 12865.98 ( 0.00%) 13209.23 ( 2.67%)
Small gains of 2-4% at low thread counts and otherwise flat. The
gains on the 8-core machine were slightly different
tbench4 on 8-core i7-3770 single socket machine
Hmean mb/sec-1 442.59 ( 0.00%) 448.73 ( 1.39%)
Hmean mb/sec-2 796.68 ( 0.00%) 794.39 ( -0.29%)
Hmean mb/sec-4 1322.52 ( 0.00%) 1343.66 ( 1.60%)
Hmean mb/sec-8 2611.65 ( 0.00%) 2694.86 ( 3.19%)
Hmean mb/sec-16 2537.07 ( 0.00%) 2609.34 ( 2.85%)
Hmean mb/sec-32 2506.02 ( 0.00%) 2578.18 ( 2.88%)
Hmean mb/sec-64 2511.06 ( 0.00%) 2569.16 ( 2.31%)
Hmean mb/sec-128 2313.38 ( 0.00%) 2395.50 ( 3.55%)
Hmean mb/sec-256 2110.04 ( 0.00%) 2177.45 ( 3.19%)
Hmean mb/sec-512 2072.51 ( 0.00%) 2053.97 ( -0.89%)
In constract, this shows a relatively steady 2-3% gain at higher thread
counts. Due to the nature of the patch and the type of workload, it's
not a surprise that the result will depend on the CPU used.
hackbench-pipes
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Amean 1 0.0637 ( 0.00%) 0.0660 ( -3.59%)
Amean 4 0.1229 ( 0.00%) 0.1181 ( 3.84%)
Amean 7 0.1921 ( 0.00%) 0.1911 ( 0.52%)
Amean 12 0.3117 ( 0.00%) 0.2923 ( 6.23%)
Amean 21 0.4050 ( 0.00%) 0.3899 ( 3.74%)
Amean 30 0.4586 ( 0.00%) 0.4433 ( 3.33%)
Amean 48 0.5910 ( 0.00%) 0.5694 ( 3.65%)
Amean 79 0.8663 ( 0.00%) 0.8626 ( 0.43%)
Amean 110 1.1543 ( 0.00%) 1.1517 ( 0.22%)
Amean 141 1.4457 ( 0.00%) 1.4290 ( 1.16%)
Amean 172 1.7090 ( 0.00%) 1.6924 ( 0.97%)
Amean 192 1.9126 ( 0.00%) 1.9089 ( 0.19%)
Some small gains and losses and while the variance data is not included,
it's close to the noise. The UMA machine did not show anything particularly
different
pipetest
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v2r2
Min Time 4.13 ( 0.00%) 3.99 ( 3.39%)
1st-qrtle Time 4.38 ( 0.00%) 4.27 ( 2.51%)
2nd-qrtle Time 4.46 ( 0.00%) 4.39 ( 1.57%)
3rd-qrtle Time 4.56 ( 0.00%) 4.51 ( 1.10%)
Max-90% Time 4.67 ( 0.00%) 4.60 ( 1.50%)
Max-93% Time 4.71 ( 0.00%) 4.65 ( 1.27%)
Max-95% Time 4.74 ( 0.00%) 4.71 ( 0.63%)
Max-99% Time 4.88 ( 0.00%) 4.79 ( 1.84%)
Max Time 4.93 ( 0.00%) 4.83 ( 2.03%)
Mean Time 4.48 ( 0.00%) 4.39 ( 1.91%)
Best99%Mean Time 4.47 ( 0.00%) 4.39 ( 1.91%)
Best95%Mean Time 4.46 ( 0.00%) 4.38 ( 1.93%)
Best90%Mean Time 4.45 ( 0.00%) 4.36 ( 1.98%)
Best50%Mean Time 4.36 ( 0.00%) 4.25 ( 2.49%)
Best10%Mean Time 4.23 ( 0.00%) 4.10 ( 3.13%)
Best5%Mean Time 4.19 ( 0.00%) 4.06 ( 3.20%)
Best1%Mean Time 4.13 ( 0.00%) 4.00 ( 3.39%)
Small improvement and similar gains were seen on the UMA machine.
The gain is small but it stands to reason that doing less work in the
scheduler is a good thing. The downside is that the lack of schedstats and
tracepoints may be surprising to experts doing performance analysis until
they find the existence of the schedstats= parameter or schedstats sysctl.
It will be automatically activated for latencytop and sleep profiling to
alleviate the problem. For tracepoints, there is a simple warning as it's
not safe to activate schedstats in the context when it's known the tracepoint
may be wanted but is unavailable.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Matt Fleming <matt@codeblueprint.co.uk>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <mgalbraith@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1454663316-22048-1-git-send-email-mgorman@techsingularity.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
||
|
Juri Lelli
|
2726d6ce38 |
sched/deadline: Unify dl_time_before() usage
Move dl_time_before() static definition in include/linux/sched/deadline.h so that it can be used by different parties without being re-defined. Reported-by: Luca Abeni <luca.abeni@unitn.it> Signed-off-by: Juri Lelli <juri.lelli@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1441188096-23021-3-git-send-email-juri.lelli@arm.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Thomas Gleixner
|
bc7a34b8b9 |
timer: Reduce timer migration overhead if disabled
Eric reported that the timer_migration sysctl is not really nice performance wise as it needs to check at every timer insertion whether the feature is enabled or not. Further the check does not live in the timer code, so we have an extra function call which checks an extra cache line to figure out that it is disabled. We can do better and store that information in the per cpu (hr)timer bases. I pondered to use a static key, but that's a nightmare to update from the nohz code and the timer base cache line is hot anyway when we select a timer base. The old logic enabled the timer migration unconditionally if CONFIG_NO_HZ was set even if nohz was disabled on the kernel command line. With this modification, we start off with migration disabled. The user visible sysctl is still set to enabled. If the kernel switches to NOHZ migration is enabled, if the user did not disable it via the sysctl prior to the switch. If nohz=off is on the kernel command line, migration stays disabled no matter what. Before: 47.76% hog [.] main 14.84% [kernel] [k] _raw_spin_lock_irqsave 9.55% [kernel] [k] _raw_spin_unlock_irqrestore 6.71% [kernel] [k] mod_timer 6.24% [kernel] [k] lock_timer_base.isra.38 3.76% [kernel] [k] detach_if_pending 3.71% [kernel] [k] del_timer 2.50% [kernel] [k] internal_add_timer 1.51% [kernel] [k] get_nohz_timer_target 1.28% [kernel] [k] __internal_add_timer 0.78% [kernel] [k] timerfn 0.48% [kernel] [k] wake_up_nohz_cpu After: 48.10% hog [.] main 15.25% [kernel] [k] _raw_spin_lock_irqsave 9.76% [kernel] [k] _raw_spin_unlock_irqrestore 6.50% [kernel] [k] mod_timer 6.44% [kernel] [k] lock_timer_base.isra.38 3.87% [kernel] [k] detach_if_pending 3.80% [kernel] [k] del_timer 2.67% [kernel] [k] internal_add_timer 1.33% [kernel] [k] __internal_add_timer 0.73% [kernel] [k] timerfn 0.54% [kernel] [k] wake_up_nohz_cpu Reported-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul McKenney <paulmck@linux.vnet.ibm.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Viresh Kumar <viresh.kumar@linaro.org> Cc: John Stultz <john.stultz@linaro.org> Cc: Joonwoo Park <joonwoop@codeaurora.org> Cc: Wenbo Wang <wenbo.wang@memblaze.com> Link: http://lkml.kernel.org/r/20150526224512.127050787@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de> |
||
|
Thomas Gleixner
|
0782e63bc6 |
sched: Handle priority boosted tasks proper in setscheduler()
Ronny reported that the following scenario is not handled correctly: T1 (prio = 10) lock(rtmutex); T2 (prio = 20) lock(rtmutex) boost T1 T1 (prio = 20) sys_set_scheduler(prio = 30) T1 prio = 30 .... sys_set_scheduler(prio = 10) T1 prio = 30 The last step is wrong as T1 should now be back at prio 20. Commit |
||
|
Kirill A. Shutemov
|
3fb1c8dcfc |
mm: update comment for DEFAULT_MAX_MAP_COUNT
With ELF extended numbering 16-bit bound is not hard limit any more. [akpm@linux-foundation.org: fix typo] Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
|
Dongsheng Yang
|
7aa2c016db |
sched: Consolidate open coded implementations of nice level frobbing into nice_to_rlimit() and rlimit_to_nice()
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/a568a1e3cc8e78648f41b5035fa5e381d36274da.1399532322.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dongsheng Yang
|
3ee237dddc |
sched/prio: Add 3 macros of MAX_NICE, MIN_NICE and NICE_WIDTH in prio.h
Currently there is lots of hard coding to 19 and -20, to represent maximum and minimum of nice values. This patch add three macros in prio.h for maximum, minimum and width of nice value, and uses it to remove hardcoded values in prio.h. Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/3994e89327b2b15f992277cdf9f409c516f87d1b.1392103744.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de> [ Collapsed two small patches. ] Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dongsheng Yang
|
7e298d60f7 |
sched/prio: Use DEFAULT_PRIO to define NICE_TO_PRIO() and PRIO_TO_NICE()
There is already a macro named DEFAULT_PRIO in prio.h, we can use it to define NICE_TO_PRIO and PRIO_TO_NICE rather than use hard coding of (MAX_RT_PRIO + 20). Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/4e28ec36fb49e8906027cbbdd900ab26a149905e.1392103744.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Thomas Gleixner
|
c365c292d0 |
sched: Consider pi boosting in setscheduler()
If a PI boosted task policy/priority is modified by a setscheduler() call we unconditionally dequeue and requeue the task if it is on the runqueue even if the new priority is lower than the current effective boosted priority. This can result in undesired reordering of the priority bucket list. If the new priority is less or equal than the current effective we just store the new parameters in the task struct and leave the scheduler class and the runqueue untouched. This is handled when the task deboosts itself. Only if the new priority is higher than the effective boosted priority we apply the change immediately. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> [ Rebase ontop of v3.14-rc1. ] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Dario Faggioli <raistlin@linux.it> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1391803122-4425-7-git-send-email-bigeasy@linutronix.de Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dongsheng Yang
|
d0ea026808 |
sched: Implement task_nice() as static inline function
As patch "sched: Move the priority specific bits into a new header file" exposes the priority related macros in linux/sched/prio.h, we don't have to implement task_nice() in kernel/sched/core.c any more. This patch implements it in linux/sched/sched.h as static inline function, saving the kernel stack and enhancing performance a bit. Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Cc: clark.williams@gmail.com Cc: rostedt@goodmis.org Cc: raistlin@linux.it Cc: juri.lelli@gmail.com Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1390878045-7096-1-git-send-email-yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dongsheng Yang
|
6b6350f155 |
sched: Expose some macros related to priority
Some macros in kernel/sched/sched.h about priority are private to kernel/sched. But they are useful to other parts of the core kernel. This patch moves these macros from kernel/sched/sched.h to include/linux/sched/prio.h so that they are available to other subsystems. Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Cc: raistlin@linux.it Cc: juri.lelli@gmail.com Cc: clark.williams@gmail.com Cc: rostedt@goodmis.org Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/2b022810905b52d13238466807f4b2a691577180.1390859827.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dongsheng Yang
|
5c228079ce |
sched: Move the priority specific bits into a new header file
Some bits about priority are defined in linux/sched/rt.h, but some of them are not only for rt scheduler, such as MAX_PRIO. This patch move them all into a new header file, linux/sched/prio.h. Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Cc: clark.williams@gmail.com Cc: rostedt@goodmis.org Cc: raistlin@linux.it Cc: juri.lelli@gmail.com Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/f7549508a1588da2c613d601748ca9de30fa5dcf.1390859827.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Linus Torvalds
|
ab5318788c |
Merge branch 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull core debug changes from Ingo Molnar: "This contains mostly kernel debugging related updates: - make hung_task detection more configurable to distros - add final bits for x86 UV NMI debugging, with related KGDB changes - update the mailing-list of MAINTAINERS entries I'm involved with" * 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: hung_task: Display every hung task warning sysctl: Add neg_one as a standard constraint x86/uv/nmi, kgdb/kdb: Fix UV NMI handler when KDB not configured x86/uv/nmi: Fix Sparse warnings kgdb/kdb: Fix no KDB config problem MAINTAINERS: Restore "L: linux-kernel@vger.kernel.org" entries |
||
|
Aaron Tomlin
|
270750dbc1 |
hung_task: Display every hung task warning
When khungtaskd detects hung tasks, it prints out backtraces from a number of those tasks. Limiting the number of backtraces being printed out can result in the user not seeing the information necessary to debug the issue. The hung_task_warnings sysctl controls this feature. This patch makes it possible for hung_task_warnings to accept a special value to print an unlimited number of backtraces when khungtaskd detects hung tasks. The special value is -1. To use this value it is necessary to change types from ulong to int. Signed-off-by: Aaron Tomlin <atomlin@redhat.com> Reviewed-by: Rik van Riel <riel@redhat.com> Acked-by: David Rientjes <rientjes@google.com> Cc: oleg@redhat.com Link: http://lkml.kernel.org/r/1390239253-24030-3-git-send-email-atomlin@redhat.com [ Build warning fix. ] Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Andi Kleen
|
54a43d5498 |
numa: add a sysctl for numa_balancing
Add a working sysctl to enable/disable automatic numa memory balancing at runtime. This allows us to track down performance problems with this feature and is generally a good idea. This was possible earlier through debugfs, but only with special debugging options set. Also fix the boot message. [akpm@linux-foundation.org: s/sched_numa_balancing/sysctl_numa_balancing/] Signed-off-by: Andi Kleen <ak@linux.intel.com> Acked-by: Mel Gorman <mgorman@suse.de> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
|
Peter Zijlstra
|
1724813d9f |
sched/deadline: Remove the sysctl_sched_dl knobs
Remove the deadline specific sysctls for now. The problem with them is that the interaction with the exisiting rt knobs is nearly impossible to get right. The current (as per before this patch) situation is that the rt and dl bandwidth is completely separate and we enforce rt+dl < 100%. This is undesirable because this means that the rt default of 95% leaves us hardly any room, even though dl tasks are saver than rt tasks. Another proposed solution was (a discarted patch) to have the dl bandwidth be a fraction of the rt bandwidth. This is highly confusing imo. Furthermore neither proposal is consistent with the situation we actually want; which is rt tasks ran from a dl server. In which case the rt bandwidth is a direct subset of dl. So whichever way we go, the introduction of dl controls at this point is painful. Therefore remove them and instead share the rt budget. This means that for now the rt knobs are used for dl admission control and the dl runtime is accounted against the rt runtime. I realise that this isn't entirely desirable either; but whatever we do we appear to need to change the interface later, so better have a small interface for now. Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/n/tip-zpyqbqds1r0vyxtxza1e7rdc@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org> |
||
|
Dario Faggioli
|
332ac17ef5 |
sched/deadline: Add bandwidth management for SCHED_DEADLINE tasks
In order of deadline scheduling to be effective and useful, it is
important that some method of having the allocation of the available
CPU bandwidth to tasks and task groups under control.
This is usually called "admission control" and if it is not performed
at all, no guarantee can be given on the actual scheduling of the
-deadline tasks.
Since when RT-throttling has been introduced each task group have a
bandwidth associated to itself, calculated as a certain amount of
runtime over a period. Moreover, to make it possible to manipulate
such bandwidth, readable/writable controls have been added to both
procfs (for system wide settings) and cgroupfs (for per-group
settings).
Therefore, the same interface is being used for controlling the
bandwidth distrubution to -deadline tasks and task groups, i.e.,
new controls but with similar names, equivalent meaning and with
the same usage paradigm are added.
However, more discussion is needed in order to figure out how
we want to manage SCHED_DEADLINE bandwidth at the task group level.
Therefore, this patch adds a less sophisticated, but actually
very sensible, mechanism to ensure that a certain utilization
cap is not overcome per each root_domain (the single rq for !SMP
configurations).
Another main difference between deadline bandwidth management and
RT-throttling is that -deadline tasks have bandwidth on their own
(while -rt ones doesn't!), and thus we don't need an higher level
throttling mechanism to enforce the desired bandwidth.
This patch, therefore:
- adds system wide deadline bandwidth management by means of:
* /proc/sys/kernel/sched_dl_runtime_us,
* /proc/sys/kernel/sched_dl_period_us,
that determine (i.e., runtime / period) the total bandwidth
available on each CPU of each root_domain for -deadline tasks;
- couples the RT and deadline bandwidth management, i.e., enforces
that the sum of how much bandwidth is being devoted to -rt
-deadline tasks to stay below 100%.
This means that, for a root_domain comprising M CPUs, -deadline tasks
can be created until the sum of their bandwidths stay below:
M * (sched_dl_runtime_us / sched_dl_period_us)
It is also possible to disable this bandwidth management logic, and
be thus free of oversubscribing the system up to any arbitrary level.
Signed-off-by: Dario Faggioli <raistlin@linux.it>
Signed-off-by: Juri Lelli <juri.lelli@gmail.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1383831828-15501-12-git-send-email-juri.lelli@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
||
|
Dario Faggioli
|
2d3d891d33 |
sched/deadline: Add SCHED_DEADLINE inheritance logic
Some method to deal with rt-mutexes and make sched_dl interact with the current PI-coded is needed, raising all but trivial issues, that needs (according to us) to be solved with some restructuring of the pi-code (i.e., going toward a proxy execution-ish implementation). This is under development, in the meanwhile, as a temporary solution, what this commits does is: - ensure a pi-lock owner with waiters is never throttled down. Instead, when it runs out of runtime, it immediately gets replenished and it's deadline is postponed; - the scheduling parameters (relative deadline and default runtime) used for that replenishments --during the whole period it holds the pi-lock-- are the ones of the waiting task with earliest deadline. Acting this way, we provide some kind of boosting to the lock-owner, still by using the existing (actually, slightly modified by the previous commit) pi-architecture. We would stress the fact that this is only a surely needed, all but clean solution to the problem. In the end it's only a way to re-start discussion within the community. So, as always, comments, ideas, rants, etc.. are welcome! :-) Signed-off-by: Dario Faggioli <raistlin@linux.it> Signed-off-by: Juri Lelli <juri.lelli@gmail.com> [ Added !RT_MUTEXES build fix. ] Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1383831828-15501-11-git-send-email-juri.lelli@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org> |