[KERNEL] [Code_Blue r84] [Lollipop 5.0.* & CM12]

Code_Blue r615

• blu_plug: remove powersaver_mode (Lollipop has a builtin similar feature), improve screenoff routines and timer
• qup_i2c: Initialize I2C resource before registering, Fix for unhandled IRQ, Improve implementation of PM callbacks, Handle scenario if runtime PM is disable
• msm: kgsl: Fix Z180 memory leak
• futex: Prevent attaching to kernel threads
• cgroup: fix exit() vs rmdir() race
• slub: fix a memory leak in get_partial_node()
• Bluetooth: fix possible info leak in bt_sock_recvmsg(), Fix invalid length check in l2cap_information_rsp(), Remove unused hci_le_ltk_reply(), Fix missing length checks for L2CAP signalling PDUs, Fix crash in l2cap_build_cmd() with small MTU, Cancel power_on and power_off work, Prevent Bluetooth L2CAP Remote Heap Memory Corruption
• tmpfs: fix mempolicy object leaks
• selinux: fix inode security list corruption, possible NULL deref in context_struct_to_string, use ARRAY_SIZE, Reduce overhead of mls_level_isvalid() function call, Report permissive mode in avc: denied messages.
• ARM: Fix undefined instruction exception handling

Hast Du den Kernel drauf?

Ne, habe den Hells L Kernel drauf

---------- Hinzugefügt um 21:21 ---------- Vorheriger Beitrag war um 21:19 ----------

Und im CM den Racer, weil der letzte CM-Code Blue Kernel bei mir nicht mehr ordentlich lief.

Code_Blue r620

• Merged Lollipop 5.0 (LRX21O)
• defconfig: Disable /dev/diag
• Revert "Revert "Revert "hammerhead: Enable /dev/diag"""
• workqueue: trivial fix for return statement in work_busy(), exit rescuer_thread() as TASK_RUNNING, add missing smp_wmb() in process_one_work(), reimplement work_on_cpu() using system_wq
• jiffies: Fix timeval conversion to jiffies
• cgroup: refactor allow_attach function into common code
• tracing/syscalls: Ignore numbers outside NR_syscalls' range, Fix perf syscall tracing when syscall_nr == -1
• msm: mpm-of: set wakeup interrupt as NO_SUSPEND
• qcom: mpm-of: fix bug in finding the pending interrupts
• msm_bam_rmnet: Fix potential memory corruption
• sched/cpuidle: reduce IPI store. Backport upstream 3.16 scheduler updates
• tick: Dynamically set broadcast irq affinity
• sched: Fix clear NOHZ_BALANCE_KICK, Micro-optimize the smart wake-affine logic, Implement smarter wake-affine logic, Fix select_idle_sibling() bouncing cow syndrome, add wait_for_completion_io[_timeout]
• block: Reserve only one queue tag for sync IO if only 3 tags are available, account iowait time when waiting for completion of IO request
• writeback: Fix occasional slow sync(1), fix race that cause writeback hung, fix writeback cache thrashing
• sync: don't block the flusher thread waiting on IO
• lowmemorykiller.c: convert to use for_each_thread
• introduce for_each_thread() to replace the buggy while_each_thread(), (fix)

Wow, schon mit der neuen Base
Wahnsinn, wie schnell es hier geht.

Ja 😀 habe aber im Moment wieder Helms Kernel, der war schneller

Jetzt auch kompatibel mit der neuen Base und CM 12

Code_Blue r646
-> Compatible with Lollipop stock LRX21O, AOSP 5.0.0_r7 LRX21V & CM12
-> blu_active: merge upstream CM and CAF updates
-> cpufreq: Move get_cpu_idle_time() to cpufreq.c
-> cpufreq_stats: do not remove sysfs files if frequency table is not present
-> sched: Reduce overestimating rq->avg_idle
-> sched/fair: Optimize cgroup pick_next_task_fair(), Optimize find_busiest_queue()
-> ARM: mutex: use generic atomic_dec-based implementation for ARMv6+
-> ARM: 7467/1: mutex: use generic xchg-based implementation for ARMv6+
-> mutex: back out architecture specific check for negative mutex count, Queue mutex spinners with MCS lock to reduce cacheline contention, Make more scalable by doing less atomic operations
-> Simple GPU governor: adapt GPU Simple Governor for Snapdrgaon 800 SOCs, fix a merge derp
-> msm: vidc: Initialize kernel space stack variables
-> msm: camera_v2: Create jpeg debugfs
-> msm: camera2: cpp: Reduce debug prints in cpp driver
-> block: row: Fix crash when adding a new field in bio struct, do not notify urgent request, when flush with data in flight
-> WLAN subsystem: Sysctl support for key TCP/IP parameters
-> msm: Create shadow files under /sys/kernel/ipv4
-> fs/super.c: sync ro remount after blocking writers
-> mmc: block: stop bkops during shutdown, Fix error path of mmc_blk_alloc_req(), flush request requeue after timeout
-> mmc: core: do not reinsert prepeared FUA and FLUSH requests in stop flow, increase timeout for disable cache operation, skip stop flow when current request already done, Refactor bus speed selection code
-> mmc: sdhci: Use max timeout and skip timeout calculation
-> mmc: card: update long_sequential_write_test queue depth
-> sdhci: sdhci_stop_request() returns error when no request
-> Revert "block: Add URGENT request notification support to CFQ scheduler"
-> defconfig: Make noop default I/O scheduler
-> rtc: alarm: have dummy set_power_on_alarm function, init power_on_alarm_lock mutex in alarmtimer_rtc_timer_init, Proper handling of spinlock, fix a bug that prevented the device to reboot into the botloader/recovery
-> RAMdisk updates, merge latest CM12 updates

Kannst Du was zum Kernel hinsichtlich der Akkulaufzeit sagen? MERCI

Gesendet von meinem Nexus 5 mit Tapatalk

Hallo. Kann ich leider nicht, da ich die meiste Zeit Hells Core drauf habe. Versuche hier nur die neuen infos bereit zu stellen. Kenne aber 2 Leute, die den zusammen mit Slim oder Velocity nutzen und sehr zufrieden sind.

okay, Danke
Hells ist auch meine erste Wahl, ausgewogen und der Akku wird auch geschont, perfekt

Gesendet von meinem Nexus 5 mit Tapatalk

das geht ja fix, schon wieder ein neues Update. Viel Spaß beim Testen!

Code_Blue r718

• Milestone 8 release
• blu_active: Revert "Check actual freq before set new freq" and boost logic
• thermal: raise thermal frequency throttle limits
• backlight: Add backlight dimmer option (disabled by default), More gradual levels for backlight dimmer mod, more gradual transitions, fix a bug where the screen would be completely dark or near darkness with backlight_dim, clean the code, add minimum backlight brightness module param, scale max current when the brightness level is below 15 instead of using always 0
• i2c: Drop missing things to 19.2 MHz
• ARM: Use -mcpu=cortex-a15 when targeting MSM Krait CPUs, add cpufreq transiton notifier to adjust loops_per_jiffy for smp
• ARM: 7685/1: delay: use private ticks_per_jiffy field for timer-based delay ops
• ARM: 7449/1: use generic strnlen_user and strncpy_from_user functions
• kthread: Fix the race condition when kthread is parked
• kthread_work: wake up worker only when the worker is idle
• timer: Generalize timer->base flags handling
• tick: Upstream fixes
• ARM: 7653/2: do not scale loops_per_jiffy when using a constant delay clock
• rcu: Yield simpler, Use smp_hotplug_thread facility for RCUs per-CPU kthread, Add a module parameter to force use of expedited RCU primitives, Add RCU_FAST_NO_HZ tracing for idle exit
• softirq: Use hotplug thread infrastructure
• Revert "sched_avg: add run queue averaging"
• net/ping: handle protocol mismatching scenario

Das geht echt total "fix"
Nach jedem Update denk ich mir, da müsste doch mal "etwas" Ruhe einkehren aber das Gegenteil ist der Fall.

Uns kann's freuen, so flotte Jungs zu haben

[KERNEL] [Code_Blue r777] [Lollipop 5.0.2 & CM12 / KK 4.4.4 & CM11 CAF]

Code_Blue r777
-> Merry Christmas edition release
-> Compressed with lz4-r125
-> makefile: refresh default toolchain linaro 2014.12 gcc 4.9.3
-> blu_active: Merged some CM improvements, Remove infinite boost and interactive boost framework, Add independent input boost freq and controls, Add Thread sync freq feature
-> blu_active: Revert "Honor boosted state in several timer calculations", "Consider changed min limit a frequency boost", "Honor changed minimum frequency"
-> msm: mdss: improvements in gamma interface (Protect access to panel command struct by mutex, Remove the broken debug FS)
-> net: ipv4: prevent leakage of uninitialized memory to user in recv syscalls
-> sched: Fix inaccurate accounting for real-time task
-> netfilter: IDLETIMER: fix invalid deference of timer
-> soc: qcom: rq_stats: Pass correct cpu id to update_average_load function

Quelle: XDA


eng.stk: I will not support caf, had enough of that in the past.
And if cm goes caf in official nightlies, I will drop cm support also.

Code_Blue r793


- blu_active: Record hispeed_validated_time in timer function
- cpufreq: Return directly in __cpufreq_get if policy is NULL
- Revamp spinlock ASM code
- regulator: core: Release regulator-regulator supplies on error
- msm: rq_stats: Remove useless iowait function
- of: Fix missing memory initialization on FDT unflattening, fix memory initialization for expanded DT
- module_param: stop double-calling parameters.
- Fix blocking allocations called very early during bootup
- af_key: initialize satype in key_notify_policy_flush(), more info leaks in pfkey messages, fix info leaks in notify messages
- net: Fix references to out-of-scope variables in put_cmsg_compat(), socket: error on a negative msg_namelen, clamp ->msg_namelen instead of returning an error, add BUG_ON if kernel advertises msg_namelen > sizeof(struct sockaddr_storage), heap overflow in __audit_sockaddr(), Block MSG_CMSG_COMPAT in send(m)msg and, fix info leak in compat dev_ifconf()recv(m)msg
- Fix order of arguments to compat_put_time[spec|val]
- pktgen: clean up ktime_t helpers
- iopoll: Use ktime_get() instead of jiffies for timeout calculations - fix, change uS to us in comments, Use ktime_get() instead of jiffies for timeout calculations
- x86, x32: Correct invalid use of user timespec in the kernel
- Add power.hammerhead. so to installer (should fix battery saver on unsupported ROMs)
- RAMdisk fixes and CM12 sync


Quelle: Code Blue Changelog

Danke für's aktualisieren
Schön, wenn der Thread hier aktuell bleibt. Ist ja auch ein Top Kernel

komm leider im Moment nicht so recht dazu, den Thread immer aktuell zu halten, v.a. da ich den Kernel nicht mehr nutze; versuche aber mein Bestes.

Code_Blue r841
-> blu_active: improve freq calc with new CPUFREQ_RELATION
-> cpufreq: Introduce new relation for freq selection, Initial CPU utilization should be 0
-> ARM: 7479/1: mm: avoid NULL dereference when flushing gate_vma with VIVT caches
-> ARM: 8051/1: put_user: fix possible data corruption in put_user
-> ARM: 7836/1: add __get_user_unaligned/__put_user_unaligned
-> ARM: warnings in arch/arm/include/asm/uaccess.h
-> ARM: 7528/1: uaccess: annotate [__]{get,put}_user functions with might_fault()
-> qseecom: Sync with CM12 (lot's of commits, check github)
-> kernel/timeconst.h: Fix this warning by removing a unneeded use of defined
-> fbcmap: prevent memory overflow
-> cpufreq: conservative: Set some usable defaults, Improve support for micro idle accounting
-> introduce kref_put_mutex()
-> msm: ADSPRPC: Fix offset for input and output buffers, Always map buffers into IOMMU if one is present, Save and restore contexts when interrupted, Add mappings to IOMMU for remote heap, Remove scatterlist validation, Close SMD channel after RPC session ends
-> defconfig: disable selinux avc stats
-> dtc: add integer overflow checks in fdt header
-> scripts: only set CROSS_COMPILE when unset to begin with
-> tracing: Have tracing_off() actually turn tracing off
-> mm: fix prctl_set_vma_anon_name
-> mmc: msm_sdccc: Remove unncessary debug
-> msm: ultrasound: ignore kmemleak false detection
-> soc: qcom: bam_dmux: Reset SSR flags in bam_init
-> RAMdisk CM12 updates

Quelle: code_blue

Perfekt, besten Dank für Dein Engagement

Kommt morgen wieder was ,ist schon wieder ein update draußen. Bin nur gerade nicht zu Hause

super, nur her damit

Gesendet von meinem Nexus 5 mit Tapatalk

Freut mich das ich euch das Update präsentieren kann und viel mehr freut mich das es der Code_Blue Kernel nun auch zum Nexus 6 geschafft hat!!!!

All right crew!
Back to the heavyweight jam...
Just push Code_Blue r852 for hammerhead N5 and my new project for shamu N6 Code_Blue r28
Check out website.
Keep it real, keep it true, keep it#Code_Blue

https://plus.google.com/115240575995...ts/Dpb3Y9pUEgX

Gesendet von meinem Nexus 6