[KERNEL] [UNIFIED] Anykernel

[Android-Andi]

@Andi46 let me check again later/tomorrow.
BTW. AnyKernel isn't needed on my latest builds, it's already included by default.

Edit:
I am not sure if safety net checks the fingerprint, maybe we need to add one of the official fingerprints in build.prop.

samsung/espresso10wifixx/espresso10wifi:4.2.2/JDQ39/P5110XXDML1:user/release-keys

(Maybe also ro.build.description from official Samsung rom

espresso10wifixx-user 4.2.2 JDQ39 P5110XXDML1 release-keys

Or
We need to define
ro.build.release_type=release
Or
ro.build.tags=release-keys)

Shouldn't matter which one we use (official build props from p3110, p3113, p5110 or p5113)


If you have time before I have: try latest UA 6.0 build on a clean flash. I know @Ziyan tested it there on tuna.

 

[Andi46]

@Andi46 let me check again later/tomorrow.
BTW. AnyKernel isn't needed on my latest builds, it's already included by default.

Edit:
I am not sure if safety net checks the fingerprint, maybe we need to add one of the official fingerprints in build.prop.

samsung/espresso10wifixx/espresso10wifi:4.2.2/JDQ39/P5110XXDML1:user/release-keys

It's OK, please take your time...
That sounds... From what i can remember the sn check is looking for the fingerprint - i think @hennymcc told us users once in the LOS 14.1 thread of S5mini + i've seen it several times in logs :fingers-crossed:
maybe you can ask him once for the preriquisites - i think he really knows the drill and dreamed timewise about it :laugh: - the CTS profile was false there too. because i'm rooted i have to use Magisk to pass SN... had tested it with/without Magisk + with/without Anykernel

 

[Android-Andi]

Ok, safety net checks for fingerprint and friends, I found a 2nd issue anyway. I'll upload a new build sometimes tomorrow.

 

[Android-Andi]

Some people have been disappointed on Lineage 13 for tuna because interactive governor is different on the omap based kernel. Attached AnyKernel is in sync with previously used interactive governor on tuna (synced with 3.4 kernel).

I've had no time to test, yet. Please let me know if it runs better for you now.

 

[ChuckMichael]

SafetyNet successfully passes on the latest AnyKernel and official LOS. I just changed this in build.prop :

ro.build.description=espresso10wifixx-user 4.2.2 JDQ39 P5110XXDML1 release-keys
ro.build.fingerprint=samsung/espresso10wifixx/espresso10wifi:4.2.2/JDQ39/P5110XXDML1:user/release-keys

Thank you so much!

 

[Android-Andi]

@ChuckMichael will be fixed by default on next Lineage - only don't install the suaddon

 

[agus_ulo]

@ChuckMichael will be fixed by default on next Lineage - only don't install the suaddon

Yes, fixed default on lineage os 13 official, latest build (espresso3g)
Thanks Andi

 

[guest4711]

Some people have been disappointed on Lineage 13 for tuna because interactive governor is different on the omap based kernel. Attached AnyKernel is in sync with previously used interactive governor on tuna (synced with 3.4 kernel).

I've had no time to test, yet. Please let me know if it runs better for you now.

As usually it behaves very well, perhaps a little bit smoother. Using it on last RR 5.7.4. Thanks again, Andi!

 

[Android-Andi]

Following changes on attached AnyKernel:

tuna: defconfig: enable LZ4-backed zRAM
dccp/tcp: do not inherit mc_list from parent (CVE-2017-8890)

(LZ4-backed zRAM for tuna doesn't interest us much here)

 

[Android-Andi]

Attached kernel going more inline with old tuna kernel

eeedd14 block, bfq: add Early Queue Merge (EQM) to BFQ-v7r8 for 3.3.0
05827bd block: introduce the BFQ-v7r8 I/O sched for 3.3
6c31895 block: cgroups, kconfig, build bits for BFQ-v7r8-3.3
71424cc cfq-iosched: fix the setting of IOPS mode on SSDs
387f925 block: Make CFQ default to IOPS mode on SSDs
ed42654 cgroup: introduce cgroup_taskset and use it in subsys->can_attach(), cancel_attach() and attach()
5ec0469 cgroup: improve old cgroup handling in cgroup_attach_proc()
d399f07 cgroups: more safe tasklist locking in cgroup_attach_proc
0b464858 cgroup: always lock threadgroup during migration
a1e2cc5 threadgroup: extend threadgroup_lock() to cover exit and exec
89c1b57 threadgroup: rename signal->threadgroup_fork_lock to ->group_rwsem
fd4c7fb cgroup: change locking order in attach_task_by_pid()
2ef610a iosched: prevent aliased requests from starving other I/O
61a6e14 mm: pass readahead info down to the i/o scheduler
6960352 block: add REQ_URGENT to request flags
4fc3a5e block: urgent request: remove unnecessary urgent marking
042e773 block: Add API for urgent request handling
b0fb175 block: Add support for reinsert a dispatched req
8af24df block: blk-core: add printk_ratelimit
38a2ba8 block: Expose kblock_schedule_delayed_work()
e6679f5 block: Add BLKROTATIONAL ioctl
f6c2c00 block: Introduce blk_set_stacking_limits function
cb4d5a8 make register_disk() static
3213e3e block/cfq-iosched: fix merge error
cf91019 block: fix lockdep warning on io_context release put_io_context()
c152385 block: don't call elevator callbacks for plug merges
7d7beac block: separate out blk_rq_merge_ok() and blk_try_merge() from elevator functions
eab94f7 block: strip out locking optimization in put_io_context()
5cd8658 block: fix ioc locking warning
1b31b19 block: fix NULL icq_cache reference
7daadfa block,cfq: change code order
b361f8b cfq-iosched: fix use-after-free of cfqq
bc8ad3a9 vfs: fix up ENOIOCTLCMD error handling
6d233b2 block: fix blk_queue_end_tag()
b605e6b block: remove WARN_ON_ONCE() in exit_io_context()
86b2b73 block: an exiting task should be allowed to create io_context
5fd36c6 block: re-use existing 'reading' variable instead of checking direction again
31237a2 block: ioc_cgroup_changed() needs to be exported
67d045a block: don't kick empty queue in blk_drain_queue()
f8af385 block: fix merge error from "block, cfq: move icq creation and rq->elv.icq association to block core"
4f0f8de block, cfq: move icq creation and rq->elv.icq association to  block core
8540252 block, cfq: restructure io_cq creation path for io_context  interface cleanup
2e88f8c block, cfq: move io_cq exit/release to blk-ioc.c
d5174cb block, cfq: move icq cache management to block core
6f7370a block, cfq: move io_cq lookup to blk-ioc.c
f4cb3ca block, cfq: move cfqd->icq_list to request_queue and add request->elv.icq
2ab770a block, cfq: reorganize cfq_io_context into generic and cfq specific parts
891853b block: remove elevator_queue->ops
3a3000f block: reorder elevator switch sequence
06bdbd3 block, cfq: replace current_io_context() with create_io_context()
fb63122 block, cfq: kill cic->key
df6359f block, cfq: kill ioc_gone
3f288cd block, cfq: remove delayed unlink
b70c0a9 block, cfq: unlink cfq_io_context's immediately
b20c193 block: fix merge error from "block, cfq: move ioc ioprio/cgroup changed handling to cic"
61f368b block, cfq: fix cic lookup locking
3c72605 block, cfq: fix race condition in cic creation path and tighten locking
caeab1d block, cfq: move ioc ioprio/cgroup changed handling to cic
ed4ef52 block, cfq: misc updates to cfq_io_context
ac0e891 block: misc updates to blk_get_queue()
2375c55 block: make ioc get/put interface more conventional and fix race on alloction
78601f4 block: rename the return of two functions
11850d3 block: misc ioc cleanups
b4f3c94 block, cfq: move cfqd->cic_index to q->id
95160e0 ida: make ida_simple_get/put() IRQ safe
d640e44 lib/idr.c: fix comment for ida_get_new_above()
7560324 Fix kernel-doc comment typo '@id'
28c5bed ida: simplified functions for id allocation
988d77f block: add missing blk_queue_dead() checks
f9199f4 block: fix drain_all condition in blk_drain_queue()
9e32543 block: add blk_queue_dead()
62490ec block: add missed trace_block_plug
7070ed9 block: avoid unnecessary plug list flush
c31aac2 block: don't call blk_drain_queue() if elevator is not up
e4ba957 blk-throttle: use queue_is_locked() instead of lockdep_is_held()
d0f3e7c blk-throttle: Take blkcg->lock while traversing blkcg->policy_list
413028b blk-throttle: Free up policy node associated with deleted rule
71cd45c block: warn if tag is greater than real_max_depth.
9565bd9 blk-flush: move the queue kick into
ebe8338 blk-flush: fix invalid BUG_ON in blk_insert_flush
898cc09 block: fix flush machinery for stacking drivers with differring flush flags
4a99464 block: Remove the control of complete cpu from bio.
59559fb block: fix a typo in the blk-cgroup.h file
613edf7 block: fix request_queue lifetime handling by making blk_queue_cleanup() properly shutdown
bd8c8b3 block: drop @tsk from attempt_plug_merge() and explain sync rules
bb587b6 block: make get_request[_wait]() fail if queue is dead
f7a45a2 block: reorganize throtl_get_tg() and blk_throtl_bio()
a1faf47 block: reorganize queue draining
1d76861 block: drop unnecessary blk_get/put_queue() in scsi_cmd_ioctl() and blk_get_tg()
cb0f4fc block: pass around REQ_* flags instead of broken down booleans during request alloc/free
03127dd block: move blk_throtl prototypes to block/blk.h
35ec081 block: fix genhd refcounting in blkio_policy_parse_and_set()
d07b9d0 block/blk-sysfs.c: fix kerneldoc references
03d6291 block: document blk-plug
e8e1e6f block: reorder request_queue to remove 64 bit alignment padding
3ad9e79 block: document blk_plug list access
01807d5 block:fix the comment error in blkdev.h
32f1c68 block: refactor generic_make_request
41e11b1 block: Don't check QUEUE_FLAG_SAME_COMP in __blk_complete_request
2d21d3b block: remove support for bio remapping from ->make_request
87644f3 fail_make_request: cleanup should_fail_request
dfe5fc7 block: fix warning with calling smp_processor_id() in preemptible section
4ae4c0c block: rename __make_request() to blk_queue_bio()
a9815d0 block: export __make_request
e86d006 elevator: use ELV_NAME_MAX instead of magic number 16 for chosen_elevator
58053ca block: simplify force plug flush code a little bit
5823dbb block: change force plug flush call order
71c4984 block: Fix queue_flag update when rq_affinity goes from 2 to 1
622e0ed block: strict rq_affinity
1146436 cfq-iosched: Reduce linked group count upon group destruction
494a037 CFQ: add think time check for group
1ba2e1d CFQ: add think time check for service tree
e761ab1 CFQ: move think time check variables to a separate struct
46d9271 fixlet: Remove fs_excl from struct task.
814809c block: avoid unnecessary plug list flush
93db18c block, sx8: kill blk_insert_request()
7050472 block: simplify force plug flush code a little bit
f2485ce block: avoid building too big plug list
f610351 make default readahead size a kernel parameter From: Nikanth Karthikesan <[email protected]>
1e947e9 Revert "BFQ-v7r8 for 3.0.0"

 

[kevinarol]

Attached kernel going more inline with old tuna kernel

eeedd14 block, bfq: add Early Queue Merge (EQM) to BFQ-v7r8 for 3.3.0
05827bd block: introduce the BFQ-v7r8 I/O sched for 3.3
6c31895 block: cgroups, kconfig, build bits for BFQ-v7r8-3.3
71424cc cfq-iosched: fix the setting of IOPS mode on SSDs
387f925 block: Make CFQ default to IOPS mode on SSDs
ed42654 cgroup: introduce cgroup_taskset and use it in subsys->can_attach(), cancel_attach() and attach()
5ec0469 cgroup: improve old cgroup handling in cgroup_attach_proc()
d399f07 cgroups: more safe tasklist locking in cgroup_attach_proc
0b464858 cgroup: always lock threadgroup during migration
a1e2cc5 threadgroup: extend threadgroup_lock() to cover exit and exec
89c1b57 threadgroup: rename signal->threadgroup_fork_lock to ->group_rwsem
fd4c7fb cgroup: change locking order in attach_task_by_pid()
2ef610a iosched: prevent aliased requests from starving other I/O
61a6e14 mm: pass readahead info down to the i/o scheduler
6960352 block: add REQ_URGENT to request flags
4fc3a5e block: urgent request: remove unnecessary urgent marking
042e773 block: Add API for urgent request handling
b0fb175 block: Add support for reinsert a dispatched req
8af24df block: blk-core: add printk_ratelimit
38a2ba8 block: Expose kblock_schedule_delayed_work()
e6679f5 block: Add BLKROTATIONAL ioctl
f6c2c00 block: Introduce blk_set_stacking_limits function
cb4d5a8 make register_disk() static
3213e3e block/cfq-iosched: fix merge error
cf91019 block: fix lockdep warning on io_context release put_io_context()
c152385 block: don't call elevator callbacks for plug merges
7d7beac block: separate out blk_rq_merge_ok() and blk_try_merge() from elevator functions
eab94f7 block: strip out locking optimization in put_io_context()
5cd8658 block: fix ioc locking warning
1b31b19 block: fix NULL icq_cache reference
7daadfa block,cfq: change code order
b361f8b cfq-iosched: fix use-after-free of cfqq
bc8ad3a9 vfs: fix up ENOIOCTLCMD error handling
6d233b2 block: fix blk_queue_end_tag()
b605e6b block: remove WARN_ON_ONCE() in exit_io_context()
86b2b73 block: an exiting task should be allowed to create io_context
5fd36c6 block: re-use existing 'reading' variable instead of checking direction again
31237a2 block: ioc_cgroup_changed() needs to be exported
67d045a block: don't kick empty queue in blk_drain_queue()
f8af385 block: fix merge error from "block, cfq: move icq creation and rq->elv.icq association to block core"
4f0f8de block, cfq: move icq creation and rq->elv.icq association to  block core
8540252 block, cfq: restructure io_cq creation path for io_context  interface cleanup
2e88f8c block, cfq: move io_cq exit/release to blk-ioc.c
d5174cb block, cfq: move icq cache management to block core
6f7370a block, cfq: move io_cq lookup to blk-ioc.c
f4cb3ca block, cfq: move cfqd->icq_list to request_queue and add request->elv.icq
2ab770a block, cfq: reorganize cfq_io_context into generic and cfq specific parts
891853b block: remove elevator_queue->ops
3a3000f block: reorder elevator switch sequence
06bdbd3 block, cfq: replace current_io_context() with create_io_context()
fb63122 block, cfq: kill cic->key
df6359f block, cfq: kill ioc_gone
3f288cd block, cfq: remove delayed unlink
b70c0a9 block, cfq: unlink cfq_io_context's immediately
b20c193 block: fix merge error from "block, cfq: move ioc ioprio/cgroup changed handling to cic"
61f368b block, cfq: fix cic lookup locking
3c72605 block, cfq: fix race condition in cic creation path and tighten locking
caeab1d block, cfq: move ioc ioprio/cgroup changed handling to cic
ed4ef52 block, cfq: misc updates to cfq_io_context
ac0e891 block: misc updates to blk_get_queue()
2375c55 block: make ioc get/put interface more conventional and fix race on alloction
78601f4 block: rename the return of two functions
11850d3 block: misc ioc cleanups
b4f3c94 block, cfq: move cfqd->cic_index to q->id
95160e0 ida: make ida_simple_get/put() IRQ safe
d640e44 lib/idr.c: fix comment for ida_get_new_above()
7560324 Fix kernel-doc comment typo '@id'
28c5bed ida: simplified functions for id allocation
988d77f block: add missing blk_queue_dead() checks
f9199f4 block: fix drain_all condition in blk_drain_queue()
9e32543 block: add blk_queue_dead()
62490ec block: add missed trace_block_plug
7070ed9 block: avoid unnecessary plug list flush
c31aac2 block: don't call blk_drain_queue() if elevator is not up
e4ba957 blk-throttle: use queue_is_locked() instead of lockdep_is_held()
d0f3e7c blk-throttle: Take blkcg->lock while traversing blkcg->policy_list
413028b blk-throttle: Free up policy node associated with deleted rule
71cd45c block: warn if tag is greater than real_max_depth.
9565bd9 blk-flush: move the queue kick into
ebe8338 blk-flush: fix invalid BUG_ON in blk_insert_flush
898cc09 block: fix flush machinery for stacking drivers with differring flush flags
4a99464 block: Remove the control of complete cpu from bio.
59559fb block: fix a typo in the blk-cgroup.h file
613edf7 block: fix request_queue lifetime handling by making blk_queue_cleanup() properly shutdown
bd8c8b3 block: drop @tsk from attempt_plug_merge() and explain sync rules
bb587b6 block: make get_request[_wait]() fail if queue is dead
f7a45a2 block: reorganize throtl_get_tg() and blk_throtl_bio()
a1faf47 block: reorganize queue draining
1d76861 block: drop unnecessary blk_get/put_queue() in scsi_cmd_ioctl() and blk_get_tg()
cb0f4fc block: pass around REQ_* flags instead of broken down booleans during request alloc/free
03127dd block: move blk_throtl prototypes to block/blk.h
35ec081 block: fix genhd refcounting in blkio_policy_parse_and_set()
d07b9d0 block/blk-sysfs.c: fix kerneldoc references
03d6291 block: document blk-plug
e8e1e6f block: reorder request_queue to remove 64 bit alignment padding
3ad9e79 block: document blk_plug list access
01807d5 block:fix the comment error in blkdev.h
32f1c68 block: refactor generic_make_request
41e11b1 block: Don't check QUEUE_FLAG_SAME_COMP in __blk_complete_request
2d21d3b block: remove support for bio remapping from ->make_request
87644f3 fail_make_request: cleanup should_fail_request
dfe5fc7 block: fix warning with calling smp_processor_id() in preemptible section
4ae4c0c block: rename __make_request() to blk_queue_bio()
a9815d0 block: export __make_request
e86d006 elevator: use ELV_NAME_MAX instead of magic number 16 for chosen_elevator
58053ca block: simplify force plug flush code a little bit
5823dbb block: change force plug flush call order
71c4984 block: Fix queue_flag update when rq_affinity goes from 2 to 1
622e0ed block: strict rq_affinity
1146436 cfq-iosched: Reduce linked group count upon group destruction
494a037 CFQ: add think time check for group
1ba2e1d CFQ: add think time check for service tree
e761ab1 CFQ: move think time check variables to a separate struct
46d9271 fixlet: Remove fs_excl from struct task.
814809c block: avoid unnecessary plug list flush
93db18c block, sx8: kill blk_insert_request()
7050472 block: simplify force plug flush code a little bit
f2485ce block: avoid building too big plug list
f610351 make default readahead size a kernel parameter From: Nikanth Karthikesan <[email protected]>
1e947e9 Revert "BFQ-v7r8 for 3.0.0"

Working great with latest Los13 build :highfive:

GTab2 P3113

 

[alfa21]

hi Andi,
can you please add to your kernels the option to include the /proc/config.gz directly accessible from the live?
you can find it in make menuconfig / general

kernel .config support -> enable access to .config through /proc/config.gz

kind regards

 

[dogukans3131]

AndroidAndi,I've found a bug in your kernel

GPU Frequency not visible and Gpu Load always %0

 

[guest4711]

AndroidAndi,I've found a bug in your kernel

GPU Frequency not visible and Gpu Load always %0

It's not a problem of @Android-Andi's anykernel. GPU settings and display work fine for me (i use Kernel Adiutor (Manage kernel parameters) - https://f-droid.org/app/com.nhellfire.kerneladiutor ) for it.

 

[Android-Andi]

New AnyKernel attached.

Changes:

dd27c97 defconfig: tuna: Modular OMAPLFB
4398e15 defconfig: espresos: Modular OMAPLFB
81ac9e4 omaplfb: Prepare for modular build (omaplfb.ko)
8b08733 mfd: twl6040: Set i2c mode correctly.
7d306af ASoC: tuna/sdp4430: fix mismatched mcbsp rx/tx threshold set.
39225d8 ASoC: tuna: simplify some logic.

 

[guest4711]

New AnyKernel attached.

Changes:

Thank you, Andi. I'm just downloading and will install on RR as usual...
I'll update this posting after testing. Have a nice weekend!

Edit: Installed without problems. Tab2 runs smooth as far as I can see (did not try any games).
Edit2: Not sure what's causing it (have to check) but no more deep sleeping ...

 

[Oldnobby]

Hi @Android-Andi
I also flash your anykernel on RRmix and it works fine! Thank you

Sent from my Galaxy Tab 2 using XDA Labs

 

[Android-Andi]

Another one, previous changes plus the following:

d2414c7 SELinux: do not allocate stack space for AVC data unless needed
78c76be selinux: don't transliterate MAY_NOT_BLOCK to IPERM_FLAG_RCU
1a07dd6 lsm_audit: don't specify the audit pre/post callbacks in 'struct common_audit_data'
3e52d03 Yama: add PR_SET_PTRACER_ANY
575c5a9 security: Yama LSM
63ee3e8 capabilities: initialize has_cap
e28294b capabilities: do not grant full privs for setuid w/ file caps + no effective caps
64d62ad selinux: fix off-by-one in setprocattr
288c0a4 inet_diag: Partly rename inet_ to sock_
484951f rcu: treewide: Do not use rcu_read_lock_held when calling rcu_dereference_check
995867f selinux: Fix Kernel panic due to NULL pointer dereference
9c23e17 selinux: remove unused|deprecated includes
364f5cc4 SELinux: check OPEN on truncate calls
9e2bea0 SELinux: include definition of new capabilities
81ef3d4 SELinux: remove avd from selinux_audit_data
2e477cf LSM: shrink the common_audit_data data union
2efa0b5 get rid of pointless includes of ext2_fs.h
393b3e3 selinuxfs: merge dentry allocation into sel_make_dir()
5940c71 SELinux: selinux/xfrm.h needs net/flow.h
42ef646 selinux: Casting (void *) value returned by kmalloc is useless
550e224 selinuxfs: remove custom hex_to_bin()
169c48d selinux: sparse fix: fix several warnings in the security server code
8dde0ae selinux: sparse fix: include selinux.h in exports.c
6c2c392 selinux: sparse fix: fix warnings in netlink code
0100b82 selinux: sparse fix: eliminate warnings for selinuxfs
928f5e0 selinux: sparse fix: declare selinux_disable() in security.h
0da4356 selinux: sparse fix: move selinux_complete_init
5b31a53 selinux: sparse fix: make selinux_secmark_refcount static
a19256d doc: Update the email address for Paul Moore in various source files
2e90e97 security,rcu: Convert call_rcu(sel_netport_free) to kfree_rcu()
9b59fb3 security,rcu: Convert call_rcu(sel_netnode_free) to kfree_rcu()
1f5b1f9 ->permission() sanitizing: don't pass flags to ->inode_permission()
381ff1a audit: treat s_id as an untrusted string

Edit:
there might be some roms (e.g. old KitKat Roms or CM/LineageOS 11) not using init.omap4.rc , those roms would need a different patch.

 

[guest4711]

Another one, previous changes plus the following:

No new problems detected during boot and first checks. Deep sleep seems to work, too. Thanks again, Andi.

 

[Android-Andi]

0d38d70 defconfig: tuna: Modular OMAPLFB
040e1f7 defconfig: espresso: Modular OMAPLFB
24b05a1 omaplfb: Prepare for modular build (omaplfb.ko)
6e375c1 kbuild: make it possible to specify the module output dir
6abd06a diffconfig: Update script to support python versions 2.5 through 3.3
19cda27 diffconfig: Gracefully exit if the default config files are not present
9e33b69 scripts/Kbuild.include: Fix portability problem of "echo -e"
7d1be38 radix-tree: fix contiguous iterator
db7fc69 radix-tree: rewrite gang lookup using iterator
cbaec1c radix-tree: introduce bit-optimized iterator
66ed07f radix_tree: take radix_tree_path off stack
ebc511c radix_tree: clean away saw_unset_tag leftovers
361da27 tmpfs radix_tree: locate_item to speed up swapoff
4487c2a Add support for mempool stats
8305589 ARM: support XZ compressed kernels
d640d40 kernel: avoid adding non-thread-group task to LMK rbtree
faf1c9f lmk: wrap the lmk code with config
f0f75af slub: zero page to fix boot crashes
21d6dcb slub: refactoring unfreeze_partials()
a13a3ea slub: change cmpxchg_double_slab in unfreeze_partials to __cmpxchg_double_slab
1207b54 slub: use __SetPageSlab function to set PG_slab flag
f511a6a slub: change cmpxchg_double_slab in get_freelist() to __cmpxchg_double_slab
6054c55 slub: remove unused argument of init_kmem_cache_node()
23d7e53 slub: fix incorrect return type of get_any_partial()
5068915 slub: fix a possible memory leak
0a76a71 slub: Extract get_freelist from __slab_alloc
0911e1a slub: Switch per cpu partial page support off for debugging
7dbcf55 slub: add missed accounting
061f22b slub: avoid potential NULL dereference or corruption
81174bd slub: use irqsafe_cpu_cmpxchg for put_cpu_partial
01cabee slub: add taint flag outputting to debug paths
94f084f slub: move discard_slab out of node lock
c7ba793 slub: use correct parameter to add a page to partial list tail
f540b8a lib/string.c: introduce memchr_inv()
8b33dfc slub: Discard slab page when node partial > minimum partial number
b5778fb slub: correct comments error for per cpu partial
789a4c3 mm: restrict access to slab files under procfs and sysfs
dc9754d slub: Code optimization in get_partial_node()
c6392fd slub: explicitly document position of inserting slab to partial list
fdea042 slub: add slab with one free object to partial list tail
c355e1c slub: per cpu cache for partial pages
230a97b Avoid duplicate _count variables in page_struct
2b3e75a mm: Rearrange struct page
c899c74 slub: return object pointer from get_partial() / new_slab().
d38ff13 slub: pass kmem_cache_cpu pointer to get_partial()
e7afa12 slub: Prepare inuse field in new_slab()
5a77812 slub: Remove useless statements in __slab_alloc
5f8f5ee slub: free slabs without holding locks
37a2681 slub: Fix partial count comparison confusion
e13d64d slub: fix check_bytes() for slub debugging
72760ae slub: Fix full list corruption if debugging is on
0fa186f slub: use print_hex_dump
855f7d4 slub: When allocating a new slab also prep the first object
9630a81 slub: disable interrupts in cmpxchg_double_slab when falling back to pagelock
db2e1c1 SLUB: Fix missing <linux/stacktrace.h> include
194099f slub: reduce overhead of slub_debug
4c4e206 slub: Add method to verify memory is not freed
f6d7e63 slub: Enable backtrace for create/delete points
5be85f9 slub: Not necessary to check for empty slab on load_freelist
6eaa4ea slub: fast release on full slab
cd3bfac slub: Add statistics for the case that the current slab does not match the node
428bb56 slub: Get rid of the another_slab label
5061f38 slub: Avoid disabling interrupts in free slowpath
64ca2b0 slub: Disable interrupts in free_debug processing
0456c2d slub: Invert locking and avoid slab lock
f263743 slub: Rework allocator fastpaths
8d096f4 slub: Pass kmem_cache struct to lock and freeze slab
c82b9ae slub: explicit list_lock taking
7952aa9 slub: Add cmpxchg_double_slab()
56676af slub: Move page->frozen handling near where the page->freelist handling occurs
d8df6fb slub: Do not use frozen page flag but a bit in the page counters
c43e9ec slub: Push irq disable into allocate_slab()
194cfd0 mm:ashmem: avoid ashmem deadlock
29ba7d7 mm: page_alloc: retry direct compaction
bbebcfc radix_tree: exceptional entries and indices
bea98b3 lowmemorykiller: Check tgid with rcu lock held
a41ef08 lowmemorykiller: maintain LMK rbtree with signal->adj_node
46b8db4 staging: android: lmk: add rcu lock while test task flag
6b64f76 staging: android: lmk: check TIF directly
0d21d0b staging: android: lmk: skip if killed by lmk
32d73b2 drivers:lmk: Fix null pointer issue
93ae827 drivers:lmk: Fix double delete issue
4a1a004 staging:android:lmk: read rb tree root with spinlock
e1badf0 staging: android: lowmemorykiller: select a new task to kill
0a70bbe oom_adj_rbtree: delete task before changing oom score
3c59bde staging: android: lowmemorykiller: implement task's adj rbtree
31e8d6a oom: fix race while temporarily setting current's oom_score_adj
1cecf47 lowmemorykiller: Use asynchronous compaction
0cf3410 lowmemorykiller: Compact memory when killing processes
74e0daa android/lowmemorykiller: Ignore tasks with freed mm
861376b android/lowmemorykiller: Wait for memory to be freed
49b650a android: lowmemorykiller: adjust module param for kernel 3.0 interface
b49a0a6 Revert "staging: android: lowmemorykiller: fix build breakage on kernel 3.0"
e05d466 Revert "mm: compaction: fix echo 1 > compact_memory return error issue"
207962a ksm: check and skip page, if it is already scanned
93780f4 mm/ksm: engage deferred timer by default
0c0b236 ksm: Provide support to use deferred timers for scanner thread
1c4cd2b neigh: Better handling of transition to NUD_PROBE state
cc95d3d sock_diag: Move the sock_ code to net/core/
2b0fa41 inet_diag: Cleanup type2proto last user
4c66304 inet_diag: Introduce socket family checks
acf55ad inet_diag: Switch the _dump to work with new header
23997c1 inet_diag: Switch the _get_exact to work with new header
fad685a inet_diag: Introduce new inet_diag_req header
b564b21 sock_diag: Initial skeleton
8218b1a inet_diag: Switch from _GETSOCK to IPPROTO_ numbers
aeb0ea1 inet_diag: Move byte-code finding up the call-stack
3c3baf6 sock_diag: Introduce new message type
ee1b3b0 tcp_cubic: do not set epoch_start in the future
4bc10ac tcp_cubic: better follow cubic curve after idle period
727e0ed net: support marking accepting TCP sockets
2f7dcb7 net: add a sysctl to reflect the fwmark on replies
bf8323b treewide: fix potentially dangerous trailing ';' in #defined values/expressions
f8bb8ed cfg80211: Fix race in bss timeout
032170b SELinux: do not allocate stack space for AVC data unless needed
afb4497 selinux: don't transliterate MAY_NOT_BLOCK to IPERM_FLAG_RCU
2ca91f5 lsm_audit: don't specify the audit pre/post callbacks in 'struct common_audit_data'
3ba7c64 Yama: add PR_SET_PTRACER_ANY
d6e37fe security: Yama LSM
a4d734a capabilities: initialize has_cap
295daf1 capabilities: do not grant full privs for setuid w/ file caps + no effective caps
55234ca selinux: fix off-by-one in setprocattr
6cc1a3c inet_diag: Partly rename inet_ to sock_
995a269 rcu: treewide: Do not use rcu_read_lock_held when calling rcu_dereference_check
87d213c selinux: Fix Kernel panic due to NULL pointer dereference
a7ba115 selinux: remove unused|deprecated includes
52942a6 SELinux: check OPEN on truncate calls
54acd40 SELinux: include definition of new capabilities
9109c7c SELinux: remove avd from selinux_audit_data
8c24fa1 LSM: shrink the common_audit_data data union
b82674a get rid of pointless includes of ext2_fs.h
05ddd75 selinuxfs: merge dentry allocation into sel_make_dir()
7368704 SELinux: selinux/xfrm.h needs net/flow.h
76336bc selinux: Casting (void *) value returned by kmalloc is useless
db26ef5 selinuxfs: remove custom hex_to_bin()
9ee848b selinux: sparse fix: fix several warnings in the security server code
a752c2d selinux: sparse fix: include selinux.h in exports.c
915e80e selinux: sparse fix: fix warnings in netlink code
9e46243 selinux: sparse fix: eliminate warnings for selinuxfs
6f99d78 selinux: sparse fix: declare selinux_disable() in security.h
fd34239 selinux: sparse fix: move selinux_complete_init
ff939bf selinux: sparse fix: make selinux_secmark_refcount static
c5bd62b doc: Update the email address for Paul Moore in various source files
4e89813 security,rcu: Convert call_rcu(sel_netport_free) to kfree_rcu()
a36f9f2 security,rcu: Convert call_rcu(sel_netnode_free) to kfree_rcu()
90c7f29 ->permission() sanitizing: don't pass flags to ->inode_permission()
9c7250b audit: treat s_id as an untrusted string
8b08733 mfd: twl6040: Set i2c mode correctly.
7d306af ASoC: tuna/sdp4430: fix mismatched mcbsp rx/tx threshold set.
39225d8 ASoC: tuna: simplify some logic.

On previous AnyKernel it seems like it didn't replaced the modules because of the supported android version check.