mirror of
https://github.com/hardkernel/linux.git
synced 2026-04-02 03:03:00 +09:00
16ce487b86fe055ee2efb84ab53cd12802dc509e
24689 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Luan Yuan
|
695cede0cc |
Amlogic: sync the code from mainline. [1/1]
PD#SWPL-17246 Problem: sync the code from mainline. Solution: sync the code from mainline. 7c03859983c2 OSS vulnerability found in [boot.img]:[linux_kernel] (CVE-2018-12232) Risk:[] [1/1] ba89a3d9c791 OSS vulnerability found in [boot.img]:[linux_kernel] (CVE-2019-8912) Risk:[] [1/1] c434d0530610 Android Security Bulletin - November 2019-11 - Kernel components binder driver - CVE-2019-2214 [1/1] ff8d9012fbd4 Android Security Bulletin - November 2019-11 - Kernel components ext4 filesystem - CVE-2019-11833 [1/1] 3c52e964495e cec: store msg after bootup from st [1/2] 94198a56ee10 lcd: support tcon vac and demura data [2/2] 1add1a008a03 vout: spi: porting lcd driver and SPI to Linux [1/1] 3e8d7b0e5f97 hdmirx: add hpd recovery logic when input clk is unstable [1/1] f92e7ba21c62 ppmgr: Add 10bit, dolby and HDR video rotation. [1/1] dab2cc37cd95 dvb: fix dmx2 interrupt bug [1/1] 9d31efae4a55 dv: add dv target output mode [1/1] e86eb9d1b5c5 hdmirx: add rx phy tdr enable control [1/1] 8ea66f645bf6 dts: enable spi for gva [1/1] baf6e74528ef drm: add drm support for tm2 [1/1] Verify: verify by newton Change-Id: I9415060a4b39895b5d624117271a72fc6a1fd187 Signed-off-by: Luan Yuan <luan.yuan@amlogic.com> |
||
|
Hui Zhang
|
826ef2cb51 |
19af5c5 dtv_demod: add AGC control for board t309 [1/1]
a9eb626 dtv_demod: Prevent NULL pointer crash caused by tuner attach failure [1/1] d27b145 dtv_demod: TL1 dtmb suspend hangup issue [1/1] 27c05dd dtv_demod: tl1,dvbc, new method for fast channel searching [1/1] fa80091 dtv_demod: txlx atsc-t loses lock when play video for a long time [1/1] 43db0c5 build: fix build err [1/1] d4abd67 dtv_demod: TXLX ISDB-T can't search channel [1/1] 70f2c39 dtv_demod: TL1 dvbc search time more than 3 min [1/1] 16d4727 dtv_demod: TL1 dtmb cma allocate fail after dtmb->dvbc->dtmb by cmd [1/1] b5aa9de media_module: h264/mpeg2/h265 [2/2] 3d6a5ed ppmgr: use light reg for video decoder local reset [1/1] ppmgr: use light reg for video decoder local reset [1/1] PD#SWPL-5014 Problem: ppmgr provider error sometimes in long time run under poor signal Solution: use light reg for video decoder local reset Verify: X301 Change-Id: I1500c9ddfdce76b3e9cb2b24a35cdee765f82d5f Signed-off-by: Hui Zhang <hui.zhang@amlogic.com> media_module: h264/mpeg2/h265 [2/2] PD#SWPL-3654 Problem: provide aspect_ratio information by AMSTREAM_IOC_VDECSTAT Solution: add ration_control information in vdec_status function for h264/mh264/vh265/mpeg12/mmpeg12, and normalized it for aspect_ratio in AMSTREAM_IOC_VDECSTAT message. Verify: Verified U212 Change-Id: Icd9270eb8f2ce2f6f7455ec28780d26ac6c56348 Signed-off-by: Rong Zhang <rong.zhang@amlogic.com> dtv_demod: TL1 dtmb cma allocate fail after dtmb->dvbc->dtmb by cmd [1/1] PD#TV-1748 Problem: [Hisense-T962X2-P-customer-DTV]:Failed to lock frequency in dtmb system Solution: leave the last mode before enter current mode to release the resources Verify: verified by t962x2_x301 Change-Id: Ibd7e1d4c67f5a5ad2191e0031fdd0b81812c8606 Signed-off-by: Zhiwei Yuan <zhiwei.yuan@amlogic.com> dtv_demod: TL1 dvbc search time more than 3 min [1/1] PD#SWPL-3555 Problem: {DVB-C}Hisense needs less than 3 minutes for dvb-c search.(5/5，None) Solution: 1.use demod internal state machine to optimize time 2.remove j.83b filter setting in dvbc mode for tl1 & txlx Verify: verified by t962x2_x301&t962x_r311 Change-Id: Ic922e3da727179e2e35a0fc80e52ac7242c54129 Signed-off-by: Zhiwei Yuan <zhiwei.yuan@amlogic.com> dtv_demod: TXLX ISDB-T can't search channel [1/1] PD#SWPL-3987 Problem: [Einstein]use ISDB-T cannot search any channel Solution: release cma memory at leave_mode for ISDB-t&DVB-T Verify: verified by t962x_r311 Change-Id: I9b79687a4d5270932ab30afaa3e7d2817e900149 Signed-off-by: Zhiwei Yuan <zhiwei.yuan@amlogic.com> build: fix build err [1/1] PD#SWPL-4150 Problem: When switch to the toolchain shipped with android P, multipile "-Werror=maybe-uninitialized" error occurs, and the code fail to be compiled Solution: Initialize variables if necessary Verify: Locally on Ampere Change-Id: I280de648914565656831e211539bf41a7dee2b4a Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com> dtv_demod: txlx atsc-t loses lock when play video for a long time [1/1] PD#SWPL-3418 Problem: [Einstein]when AC off/on,enter live TV , TV display blank screen and no signal about 5S Solution: use post-eq to track(0x912=0x50) Verify: verified by t962x_r311 Change-Id: I924c55a8f1c92328e4371cb730e1b03bd720457f Signed-off-by: Zhiwei Yuan <zhiwei.yuan@amlogic.com> dtv_demod: tl1,dvbc, new method for fast channel searching [1/1] PD#TV-2154 Problem: [Hisense-T962X2-P-customer-DTV]:DVBC QAM is set to auto to search channels Solution: add new searching method note: use "demod_dvbc_speedup_en" to mark the new method it's disabled as default, can be enabled if needed we can make it always enabled after all testing are passed enable: echo fast_search on > /sys/kernel/debug/demod/dvbc_channel_fast Verify: verified by t962x2_x301 Change-Id: Icaaab9f27eb058a062d7048c6ca9fa2e3bff008e Signed-off-by: Zhiwei Yuan <zhiwei.yuan@amlogic.com> dtv_demod: TL1 dtmb suspend hangup issue [1/1] PD#SWPL-5202 Problem: cma memory is not released when suspend Solution: provide suspend interface to release cma memory Verify: verified by t962x2_x301 Change-Id: I1d808a8a6119a2f385961c4c5bded5ab71c9c9d1 Signed-off-by: zhiwei.yuan <zhiwei.yuan@amlogic.com> dtv_demod: Prevent NULL pointer crash caused by tuner attach failure [1/1] PD#TV-1539 Problem: Prevent NULL pointer crash caused by tuner attach failure. Solution: Prevent NULL pointer crash caused by tuner attach failure. Verify: verified by x301 Change-Id: I57cf32947775626467eb952dd2298ae9ec84601d Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com> dtv_demod: add AGC control for board t309 [1/1] PD#SWPL-5175 Problem: r842 agc control need be controlled by demod Solution: add agc control function Verify: verified by t962x2_x301 t962x2_t309 Change-Id: If712e22276b97c457e2e2ed1c79bdf3673813dff Signed-off-by: zhiwei.yuan <zhiwei.yuan@amlogic.com> |
||
|
Jianxin Pan
|
31e3a8ac1f |
schedtune: fix crash when there is cpu boot fail in hmp [1/1]
PD#SWPL-7656 Problem: crash when there is cpu boot fail in hmp Solution: fix crash when there is cpu boot fail in hmp Verify: W400 Change-Id: I0153975593adb0bfcbc3c3bd6543f0fb2e6bf2e0 Signed-off-by: Jianxin Pan <jianxin.pan@amlogic.com> |
||
|
tao zeng
|
8021e13c8e |
mm: optimize thread stack usage on ARMv7 [1/1]
PD#SWPL-2681 Problem: Kernel stack usage is large when running many tasks. Solution: Map kernel stack to module space and handle page-fault for stack fault. This can save about 50% memory of stack usage Verify: p212 Change-Id: Ie894bc8f00cb525ddf8ac63c6d99d9c6e937fdc0 Signed-off-by: tao zeng <tao.zeng@amlogic.com> |
||
|
Jianxiong Pan
|
e8da821ba8 |
printk: fix print_time bug when run dmesg [1/1]
PD#SWPL-4900 Problem: printk_prefix() passes NULL buf to print_time() to get the length of the time prefix; when printk times are enable, the current code just return "snprint(NULL, 0, "[%5lu.000000]", (unsigned long)ts)" but in the SMP case, we add cpu core id at the end of time prefix, the format is "[%5lu, %06lu@%d]", this gives a wrong size(shorter). this will make buf get out of boundary, when run dmesg. Solution: replace [%5lu.000000] with [%5lu.000000@0] Verify: g12a-u200 Change-Id: Ic74b9f466b5af5aac7e6569a1afc7a46d9594050 Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com> |
||
|
tao zeng
|
4d6ae43593 |
mm: optimize thread stack usage on arm64 [1/1]
PD#SWPL-1219 Problem: On arm64, thread stack is 16KB for each task. If running task number is large, this type of memory may over 40MB. It's a large amount on small memory platform. But most case thread only use less 4KB stack. It's waste of memory and we need optimize it. Solution: 1. Pre-allocate a vmalloc address space for task stack; 2. Only map 1st page for stack and handle page fault in EL1 when stack growth triggered exception; 3. handle stack switch for exception. Verify: p212 Change-Id: I47f511ccfa2868d982bc10a820ed6435b6d52ba9 Signed-off-by: tao zeng <tao.zeng@amlogic.com> |
||
|
jiamin ma
|
96e7519870 |
debug: show more debug info when hardlockup happens
PD#173193: need more debug info when hardlockup happens Change-Id: Ia265a58ba776168ac4838cc7f4a6d7cfbeaf6557 Signed-off-by: jiamin ma <jiamin.ma@amlogic.com> |
||
|
Victor Wan
|
cc7b1eac54 |
Merge branch 'android-4.9' into amlogic-4.9-dev
Signed-off-by: Victor Wan <victor.wan@amlogic.com> Conflicts: drivers/md/dm-bufio.c drivers/media/dvb-core/dvb_frontend.c drivers/usb/dwc3/core.c drivers/usb/gadget/function/f_fs.c |
||
|
Sultan Alsawaf
|
47bbcd6bf8 |
ANDROID: Fix massive cpufreq_times memory leaks
Every time _cpu_up() is called for a CPU, idle_thread_get() is called which then re-initializes a CPU's idle thread that was already previously created and cached in a global variable in smpboot.c. idle_thread_get() calls init_idle() which then calls __sched_fork(). __sched_fork() is where cpufreq_task_times_init() is, and cpufreq_task_times_init() allocates memory for the task struct's time_in_state array. Since idle_thread_get() reuses a task struct instance that was already previously created, this means that every time it calls init_idle(), cpufreq_task_times_init() allocates this array again and overwrites the existing allocation that the idle thread already had. This causes memory to be leaked every time a CPU is onlined. In order to fix this, move allocation of time_in_state into _do_fork to avoid allocating it at all for idle threads. The cpufreq times interface is intended to be used for tracking userspace tasks, so we can safely remove it from the kernel's idle threads without killing any functionality. But that's not all! Task structs can be freed outside of release_task(), which creates another memory leak because a task struct can be freed without having its cpufreq times allocation freed. To fix this, free the cpufreq times allocation at the same time that task struct allocations are freed, in free_task(). Since free_task() can also be called in error paths of copy_process() after dup_task_struct(), set time_in_state to NULL immediately after calling dup_task_struct() to avoid possible double free. Bug description and fix adapted from patch submitted by Sultan Alsawaf <sultanxda@gmail.com> at https://android-review.googlesource.com/c/kernel/msm/+/700134 Bug: 110044919 Test: Hikey960 builds, boots & reports /proc/<pid>/time_in_state correctly Change-Id: I12fe7611fc88eb7f6c39f8f7629ad27b6ec4722c Signed-off-by: Connor O'Brien <connoro@google.com> |
||
|
Connor O'Brien
|
23a1412b82 |
ANDROID: Reduce use of #ifdef CONFIG_CPU_FREQ_TIMES
Add empty versions of functions to cpufreq_times.h to cut down on use of #ifdef in .c files. Test: kernel builds with and without CONFIG_CPU_FREQ_TIMES=y Change-Id: I49ac364fac3d42bba0ca1801e23b15081094fb12 Signed-off-by: Connor O'Brien <connoro@google.com> |
||
|
Greg Kroah-Hartman
|
7f6f94c98c |
Merge 4.9.113 into android-4.9
Changes in 4.9.113 nvme: validate admin queue before unquiesce MIPS: Call dump_stack() from show_regs() MIPS: Fix ioremap() RAM check mmc: dw_mmc: fix card threshold control configuration ibmasm: don't write out of bounds in read handler ata: Fix ZBC_OUT command block check ata: Fix ZBC_OUT all bit handling vmw_balloon: fix inflation with batching ahci: Disable LPM on Lenovo 50 series laptops with a too old BIOS USB: serial: ch341: fix type promotion bug in ch341_control_in() USB: serial: cp210x: add another USB ID for Qivicon ZigBee stick USB: serial: keyspan_pda: fix modem-status error handling USB: yurex: fix out-of-bounds uaccess in read handler USB: serial: mos7840: fix status-register error handling usb: quirks: add delay quirks for Corsair Strafe xhci: xhci-mem: off by one in xhci_stream_id_to_ring() HID: usbhid: add quirk for innomedia INNEX GENESIS/ATARI adapter Fix up non-directory creation in SGID directories ALSA: hda - Handle pm failure during hotplug fs, elf: make sure to page align bss in load_elf_library tools build: fix # escaping in .cmd files for future Make i2c: tegra: Fix NACK error handling iw_cxgb4: correctly enforce the max reg_mr depth nvme-pci: Remap CMB SQ entries on every controller reset uprobes/x86: Remove incorrect WARN_ON() in uprobe_init_insn() netfilter: nf_queue: augment nfqa_cfg_policy netfilter: x_tables: initialise match/target check parameter struct loop: add recursion validation to LOOP_CHANGE_FD PM / hibernate: Fix oops at snapshot_write() RDMA/ucm: Mark UCM interface as BROKEN loop: remember whether sysfs_create_group() was done Linux 4.9.113 Signed-off-by: Greg Kroah-Hartman <gregkh@google.com> |
||
|
Tetsuo Handa
|
34f841a3c3 |
PM / hibernate: Fix oops at snapshot_write()
commit
|
||
|
libo
|
60e98ef37d |
wifi: softirq: reduce ksoftirqd scheduled rate to improve TCP throughtput
PD#162113: improve ap62x8 rx tput
The reason why wifi TCP rx throughtput lower than kernel3.14 are as flollows:
On K3.14, RX has more time to work in ISR, but on K4.9, RX has more time to
work in kernel ksoftirqd thread, however, ksoftirqd has a very low priority.
On K4.9, a commit from upstream caused this difference, then revert it.
Revert "softirq: Let ksoftirqd do its job"
This reverts commit
|
||
|
Greg Kroah-Hartman
|
9e79039544 |
Merge 4.9.112 into android-4.9
Changes in 4.9.112 usb: cdc_acm: Add quirk for Uniden UBC125 scanner USB: serial: cp210x: add CESINEL device ids USB: serial: cp210x: add Silicon Labs IDs for Windows Update usb: dwc2: fix the incorrect bitmaps for the ports of multi_tt hub n_tty: Fix stall at n_tty_receive_char_special(). n_tty: Access echo_* variables carefully. staging: android: ion: Return an ERR_PTR in ion_map_kernel vt: prevent leaking uninitialized data to userspace via /dev/vcs* i2c: rcar: fix resume by always initializing registers before transfer ipv4: Fix error return value in fib_convert_metrics() kprobes/x86: Do not modify singlestep buffer while resuming netfilter: nf_tables: use WARN_ON_ONCE instead of BUG_ON in nft_do_chain() Revert "sit: reload iphdr in ipip6_rcv" net: phy: micrel: fix crash when statistic requested for KSZ9031 phy ARM: dts: imx6q: Use correct SDMA script for SPI5 core IB/hfi1: Fix user context tail allocation for DMA_RTAIL x86/xen: Add call of speculative_store_bypass_ht_init() to PV paths x86/cpu: Re-apply forced caps every time CPU caps are re-read mm: hugetlb: yield when prepping struct pages tracing: Fix missing return symbol in function_graph output scsi: sg: mitigate read/write abuse s390: Correct register corruption in critical section cleanup drbd: fix access after free cifs: Fix infinite loop when using hard mount option drm/udl: fix display corruption of the last line jbd2: don't mark block as modified if the handle is out of credits ext4: make sure bitmaps and the inode table don't overlap with bg descriptors ext4: always check block group bounds in ext4_init_block_bitmap() ext4: only look at the bg_flags field if it is valid ext4: verify the depth of extent tree in ext4_find_extent() ext4: include the illegal physical block in the bad map ext4_error msg ext4: clear i_data in ext4_inode_info when removing inline data ext4: add more inode number paranoia checks ext4: add more mount time checks of the superblock ext4: check superblock mapped prior to committing mlxsw: spectrum: Forbid linking of VLAN devices to devices that have uppers HID: i2c-hid: Fix "incomplete report" noise HID: hiddev: fix potential Spectre v1 HID: debug: check length before copy_to_user() PM / OPP: Update voltage in case freq == old_freq Kbuild: fix # escaping in .cmd files for future Make media: cx25840: Use subdev host data for PLL override mm, page_alloc: do not break __GFP_THISNODE by zonelist reset dm bufio: avoid sleeping while holding the dm_bufio lock dm bufio: drop the lock when doing GFP_NOIO allocation mtd: rawnand: mxc: set spare area size register explicitly dm bufio: don't take the lock in dm_bufio_shrink_count mtd: cfi_cmdset_0002: Change definition naming to retry write operation mtd: cfi_cmdset_0002: Change erase functions to retry for error mtd: cfi_cmdset_0002: Change erase functions to check chip good only netfilter: nf_log: don't hold nf_log_mutex during user access staging: comedi: quatech_daqp_cs: fix no-op loop daqp_ao_insn_write() Linux 4.9.112 Signed-off-by: Greg Kroah-Hartman <gregkh@google.com> |
||
|
Changbin Du
|
07cd8167aa |
tracing: Fix missing return symbol in function_graph output
commit |
||
|
Greg Kroah-Hartman
|
03c70feafd |
Merge 4.9.111 into android-4.9
Changes in 4.9.111
x86/spectre_v1: Disable compiler optimizations over array_index_mask_nospec()
x86/mce: Improve error message when kernel cannot recover
x86/mce: Check for alternate indication of machine check recovery on Skylake
x86/mce: Fix incorrect "Machine check from unknown source" message
x86/mce: Do not overwrite MCi_STATUS in mce_no_way_out()
x86: Call fixup_exception() before notify_die() in math_error()
m68k/mm: Adjust VM area to be unmapped by gap size for __iounmap()
serial: sh-sci: Use spin_{try}lock_irqsave instead of open coding version
signal/xtensa: Consistenly use SIGBUS in do_unaligned_user
usb: do not reset if a low-speed or full-speed device timed out
1wire: family module autoload fails because of upper/lower case mismatch.
ASoC: dapm: delete dapm_kcontrol_data paths list before freeing it
ASoC: cirrus: i2s: Fix LRCLK configuration
ASoC: cirrus: i2s: Fix {TX|RX}LinCtrlData setup
clk: renesas: cpg-mssr: Stop using printk format %pCr
lib/vsprintf: Remove atomic-unsafe support for %pCr
mips: ftrace: fix static function graph tracing
branch-check: fix long->int truncation when profiling branches
ipmi:bt: Set the timeout before doing a capabilities check
Bluetooth: hci_qca: Avoid missing rampatch failure with userspace fw loader
fuse: atomic_o_trunc should truncate pagecache
fuse: don't keep dead fuse_conn at fuse_fill_super().
fuse: fix control dir setup and teardown
powerpc/mm/hash: Add missing isync prior to kernel stack SLB switch
powerpc/ptrace: Fix setting 512B aligned breakpoints with PTRACE_SET_DEBUGREG
powerpc/ptrace: Fix enforcement of DAWR constraints
powerpc/powernv/ioda2: Remove redundant free of TCE pages
cpuidle: powernv: Fix promotion from snooze if next state disabled
powerpc/fadump: Unregister fadump on kexec down path.
ARM: 8764/1: kgdb: fix NUMREGBYTES so that gdb_regs[] is the correct size
arm64: kpti: Use early_param for kpti= command-line option
arm64: mm: Ensure writes to swapper are ordered wrt subsequent cache maintenance
of: unittest: for strings, account for trailing \0 in property length field
IB/qib: Fix DMA api warning with debug kernel
IB/{hfi1, qib}: Add handling of kernel restart
IB/mlx5: Fetch soft WQE's on fatal error state
IB/isert: Fix for lib/dma_debug check_sync warning
IB/isert: fix T10-pi check mask setting
RDMA/mlx4: Discard unknown SQP work requests
mtd: cfi_cmdset_0002: Change write buffer to check correct value
mtd: cfi_cmdset_0002: Use right chip in do_ppb_xxlock()
mtd: cfi_cmdset_0002: fix SEGV unlocking multiple chips
mtd: cfi_cmdset_0002: Fix unlocking requests crossing a chip boudary
mtd: cfi_cmdset_0002: Avoid walking all chips when unlocking.
MIPS: BCM47XX: Enable 74K Core ExternalSync for PCIe erratum
PCI: Add ACS quirk for Intel 7th & 8th Gen mobile
PCI: Add ACS quirk for Intel 300 series
PCI: pciehp: Clear Presence Detect and Data Link Layer Status Changed on resume
printk: fix possible reuse of va_list variable
MIPS: io: Add barrier after register read in inX()
time: Make sure jiffies_to_msecs() preserves non-zero time periods
X.509: unpack RSA signatureValue field from BIT STRING
Btrfs: fix return value on rename exchange failure
Btrfs: fix unexpected cow in run_delalloc_nocow
iio:buffer: make length types match kfifo types
scsi: qla2xxx: Fix setting lower transfer speed if GPSC fails
scsi: zfcp: fix missing SCSI trace for result of eh_host_reset_handler
scsi: zfcp: fix missing SCSI trace for retry of abort / scsi_eh TMF
scsi: zfcp: fix misleading REC trigger trace where erp_action setup failed
scsi: zfcp: fix missing REC trigger trace on terminate_rport_io early return
scsi: zfcp: fix missing REC trigger trace on terminate_rport_io for ERP_FAILED
scsi: zfcp: fix missing REC trigger trace for all objects in ERP_FAILED
scsi: zfcp: fix missing REC trigger trace on enqueue without ERP thread
linvdimm, pmem: Preserve read-only setting for pmem devices
clk: at91: PLL recalc_rate() now using cached MUL and DIV values
md: fix two problems with setting the "re-add" device state.
rpmsg: smd: do not use mananged resources for endpoints and channels
ubi: fastmap: Cancel work upon detach
ubi: fastmap: Correctly handle interrupted erasures in EBA
UBIFS: Fix potential integer overflow in allocation
backlight: as3711_bl: Fix Device Tree node lookup
backlight: max8925_bl: Fix Device Tree node lookup
backlight: tps65217_bl: Fix Device Tree node lookup
mfd: intel-lpss: Program REMAP register in PIO mode
perf tools: Fix symbol and object code resolution for vdso32 and vdsox32
perf intel-pt: Fix sync_switch INTEL_PT_SS_NOT_TRACING
perf intel-pt: Fix decoding to accept CBR between FUP and corresponding TIP
perf intel-pt: Fix MTC timing after overflow
perf intel-pt: Fix "Unexpected indirect branch" error
perf intel-pt: Fix packet decoding of CYC packets
media: v4l2-compat-ioctl32: prevent go past max size
media: cx231xx: Add support for AverMedia DVD EZMaker 7
media: dvb_frontend: fix locking issues at dvb_frontend_get_event()
nfsd: restrict rd_maxcount to svc_max_payload in nfsd_encode_readdir
NFSv4: Fix possible 1-byte stack overflow in nfs_idmap_read_and_verify_message
NFSv4: Revert commit
|
||
|
Geert Uytterhoeven
|
8fd86587ea |
time: Make sure jiffies_to_msecs() preserves non-zero time periods
commit
|
||
|
Tetsuo Handa
|
db2baeef79 |
printk: fix possible reuse of va_list variable
commit |
||
|
jianxin.pan
|
1efcf84db4 |
schedutil: limit up/down init rate
PD#165090: schedutil: limit up/down min rate Change-Id: Ib3aa6653d56056298df05bdede2e2bf6aea46882 Signed-off-by: jianxin.pan <jianxin.pan@amlogic.com> |
||
|
jianxin.pan
|
874ce1dbf4 |
sched: disable SD_WAKE_AFFINE
PD#165090: remove WAKE_AFFINE to get better balance when heavy loading Change-Id: Id5650e9c3fd12b23be04f8f52a0f5c2e11c49199 Signed-off-by: jianxin.pan <jianxin.pan@amlogic.com> |
||
|
Juri Lelli
|
e7fd5b1876 |
UPSTREAM: cpufreq: schedutil: use now as reference when aggregating shared policy requests
Currently, sugov_next_freq_shared() uses last_freq_update_time as a
reference to decide when to start considering CPU contributions as
stale.
However, since last_freq_update_time is set by the last CPU that issued
a frequency transition, this might cause problems in certain cases. In
practice, the detection of stale utilization values fails whenever the
CPU with such values was the last to update the policy. For example (and
please note again that the SCHED_CPUFREQ_RT flag is not the problem
here, but only the detection of after how much time that flag has to be
considered stale), suppose a policy with 2 CPUs:
CPU0 | CPU1
|
| RT task scheduled
| SCHED_CPUFREQ_RT is set
| CPU1->last_update = now
| freq transition to max
| last_freq_update_time = now
|
more than TICK_NSEC nsecs
|
a small CFS wakes up |
CPU0->last_update = now1 |
delta_ns(CPU0) < TICK_NSEC* |
CPU0's util is considered |
delta_ns(CPU1) = |
last_freq_update_time - |
CPU1->last_update = 0 |
< TICK_NSEC |
CPU1 is still considered |
CPU1->SCHED_CPUFREQ_RT is set |
we stay at max (until CPU1 |
exits from idle) |
* delta_ns is actually negative as now1 > last_freq_update_time
While last_freq_update_time is a sensible reference for rate limiting,
it doesn't seem to be useful for working around stale CPU states.
Fix the problem by always considering now (time) as the reference for
deciding when CPUs have stale contributions.
Bug: 109836581
Signed-off-by: Juri Lelli <juri.lelli@arm.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
(cherry picked from commit
|
||
|
Patrick Bellasi
|
db2c520bb5 |
ANDROID: sched/tune: fix boost_group spin_lock re-initialization
In: |
||
|
Patrick Bellasi
|
eafebcabec |
ANDROID: sched/tune: cleanup schedtune_boostgroup_{init,release}
This update the definition of these two methods to make more clear their role. In this refactored version, the slow path entry functions: schedtune_css_alloc() schedtune_css_free() are in charge just to allocate and release the memory used to track schedtune boost groups. These two functions rely respectively on: schedtune_boostgroup_init() schedtune_boostgroup_release() for everything related to setting up data structures and properly initializing them. Change-Id: I9336102b5c6a6b5726fd466e99b7d6b28d38f455 Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com> |
||
|
Patrick Bellasi
|
ca17b83d9c |
ANDROID: sched/tune: remove unused variable
This value was added in:
|
||
|
Patrick Bellasi
|
c964a2ba92 |
ANDROID: sched/fair: cosmetics
Change-Id: Ibbefdf1cdb4d72c7075c7b3edf6789630475a8e2 Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com> |
||
|
Greg Kroah-Hartman
|
42a730adb6 |
Merge 4.9.107 into android-4.9
Changes in 4.9.107
arm64: lse: Add early clobbers to some input/output asm operands
powerpc/64s: Clear PCR on boot
USB: serial: cp210x: use tcflag_t to fix incompatible pointer type
Revert "pinctrl: msm: Use dynamic GPIO numbering"
xfs: detect agfl count corruption and reset agfl
Revert "ima: limit file hash setting by user to fix and log modes"
Input: elan_i2c_smbus - fix corrupted stack
tracing: Fix crash when freeing instances with event triggers
selinux: KASAN: slab-out-of-bounds in xattr_getsecurity
cfg80211: further limit wiphy names to 64 bytes
dma-buf: remove redundant initialization of sg_table
rtlwifi: rtl8192cu: Remove variable self-assignment in rf.c
ASoC: Intel: sst: remove redundant variable dma_dev_name
platform/chrome: cros_ec_lpc: remove redundant pointer request
x86/amd: revert commit
|
||
|
Steven Rostedt (VMware)
|
dfc80dcea2 |
tracing: Fix crash when freeing instances with event triggers
commit |
||
|
Greg Kroah-Hartman
|
8b94b87fe7 |
Merge 4.9.106 into android-4.9
Changes in 4.9.106
objtool: Improve detection of BUG() and other dead ends
objtool: Move checking code to check.c
tools lib: Add for_each_clear_bit macro
tools: add more bitmap functions
tools: enable endian checks for all sparse builds
tools include: Introduce linux/compiler-gcc.h
radix tree test suite: Remove types.h
tools include: Adopt __compiletime_error
tools include: Introduce atomic_cmpxchg_{relaxed,release}()
tools include: Add UINT_MAX def to kernel.h
tools include: Adopt kernel's refcount.h
perf tools: Force fixdep compilation at the start of the build
perf tools: Move headers check into bash script
tools include uapi: Grab copies of stat.h and fcntl.h
tools include: Introduce linux/bug.h, from the kernel sources
tools include: Adopt __same_type() and __must_be_array() from the kernel
tools include: Move ARRAY_SIZE() to linux/kernel.h
tools include: Drop ARRAY_SIZE() definition from linux/hashtable.h
tools include: Include missing headers for fls() and types in linux/log2.h
objtool: sync up with the 4.14.47 version of objtool
objtool: Support GCC 8's cold subfunctions
objtool: Support GCC 8 switch tables
objtool: Detect RIP-relative switch table references
objtool: Detect RIP-relative switch table references, part 2
objtool: Fix "noreturn" detection for recursive sibling calls
objtool, x86: Add several functions and files to the objtool whitelist
perf/tools: header file sync up
objtool: header file sync-up
x86/xen: Add unwind hint annotations to xen_setup_gdt
objtool: Enclose contents of unreachable() macro in a block
Linux 4.9.106
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
|
||
|
Josh Poimboeuf
|
935893a17a |
objtool, x86: Add several functions and files to the objtool whitelist
commit
|
||
|
Rik van Riel
|
dbe0f61220 |
ANDROID: add extra free kbytes tunable
Add a userspace visible knob to tell the VM to keep an extra amount of memory free, by increasing the gap between each zone's min and low watermarks. This is useful for realtime applications that call system calls and have a bound on the number of allocations that happen in any short time period. In this application, extra_free_kbytes would be left at an amount equal to or larger than than the maximum number of allocations that happen in any burst. It may also be useful to reduce the memory use of virtual machines (temporarily?), in a way that does not cause memory fragmentation like ballooning does. [ccross] Revived for use on old kernels where no other solution exists. The tunable will be removed on kernels that do better at avoiding direct reclaim. [surenb] Will be reverted as soon as Android framework is reworked to use upstream-supported watermark_scale_factor instead of extra_free_kbytes. Bug: 86445363 Change-Id: I765a42be8e964bfd3e2886d1ca85a29d60c3bb3e Signed-off-by: Rik van Riel<riel@redhat.com> Signed-off-by: Colin Cross <ccross@android.com> Signed-off-by: Suren Baghdasaryan <surenb@google.com> |
||
|
tao zeng
|
9cbd99daf1 |
ramdump: add ramdump support for kernel [4/6]
PD#165764: add ramdump support 1. Add Makefile and driver of ramdump; 2. Flush all cache before panic reboot; 3. Change panic reboot reason when ramdump is disabled; 4. Add dts support/open watch dog for chips. Change-Id: Ieeb418f038bfda119c3156a7f8f8b05c8bc58ad2 Signed-off-by: tao zeng <tao.zeng@amlogic.com> |
||
|
Satya Durga Srinivasu Prabhala
|
a56900fabe |
ANDROID: sched/walt: Fix compilation issue for x86_64
Below compilation errors observed when SCHED_WALT enabled and
FAIR_GROUP_SCHED is disabled, fix it.
CC kernel/sched/walt.o
kernel/sched/walt.c: In function .walt_inc_cfs_cumulative_runnable_avg.:
kernel/sched/walt.c:157:8: error: .struct cfs_rq. has no member named .cumulative_runnable_avg.
cfs_rq->cumulative_runnable_avg += p->ravg.demand;
^
kernel/sched/walt.c: In function .walt_dec_cfs_cumulative_runnable_avg.:
kernel/sched/walt.c:163:8: error: .struct cfs_rq. has no member named .cumulative_runnable_avg.
cfs_rq->cumulative_runnable_avg -= p->ravg.demand;
^
make[2]: *** [kernel/sched/walt.o] Error 1
make[1]: *** [kernel/sched] Error 2
make: *** [kernel] Error 2
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
|
||
|
Greg Kroah-Hartman
|
9797dcb8c7 |
Merge 4.9.104 into android-4.9
Changes in 4.9.104
MIPS: c-r4k: Fix data corruption related to cache coherence
MIPS: ptrace: Expose FIR register through FP regset
MIPS: Fix ptrace(2) PTRACE_PEEKUSR and PTRACE_POKEUSR accesses to o32 FGRs
KVM: Fix spelling mistake: "cop_unsuable" -> "cop_unusable"
affs_lookup(): close a race with affs_remove_link()
aio: fix io_destroy(2) vs. lookup_ioctx() race
ALSA: timer: Fix pause event notification
do d_instantiate/unlock_new_inode combinations safely
mmc: sdhci-iproc: remove hard coded mmc cap 1.8v
mmc: sdhci-iproc: fix 32bit writes for TRANSFER_MODE register
libata: Blacklist some Sandisk SSDs for NCQ
libata: blacklist Micron 500IT SSD with MU01 firmware
xen-swiotlb: fix the check condition for xen_swiotlb_free_coherent
drm/vmwgfx: Fix 32-bit VMW_PORT_HB_[IN|OUT] macros
IB/hfi1: Use after free race condition in send context error path
Revert "ipc/shm: Fix shmat mmap nil-page protection"
ipc/shm: fix shmat() nil address after round-down when remapping
kasan: fix memory hotplug during boot
kernel/sys.c: fix potential Spectre v1 issue
kernel/signal.c: avoid undefined behaviour in kill_something_info
KVM/VMX: Expose SSBD properly to guests
KVM: s390: vsie: fix < 8k check for the itdba
KVM: x86: Update cpuid properly when CR4.OSXAVE or CR4.PKE is changed
kvm: x86: IA32_ARCH_CAPABILITIES is always supported
firewire-ohci: work around oversized DMA reads on JMicron controllers
x86/tsc: Allow TSC calibration without PIT
NFSv4: always set NFS_LOCK_LOST when a lock is lost.
ALSA: hda - Use IS_REACHABLE() for dependency on input
kvm: x86: fix KVM_XEN_HVM_CONFIG ioctl
netfilter: ipv6: nf_defrag: Pass on packets to stack per RFC2460
tracing/hrtimer: Fix tracing bugs by taking all clock bases and modes into account
PCI: Add function 1 DMA alias quirk for Marvell 9128
Input: psmouse - fix Synaptics detection when protocol is disabled
i40iw: Zero-out consumer key on allocate stag for FMR
tools lib traceevent: Simplify pointer print logic and fix %pF
perf callchain: Fix attr.sample_max_stack setting
tools lib traceevent: Fix get_field_str() for dynamic strings
perf record: Fix failed memory allocation for get_cpuid_str
iommu/vt-d: Use domain instead of cache fetching
dm thin: fix documentation relative to low water mark threshold
net: stmmac: dwmac-meson8b: fix setting the RGMII TX clock on Meson8b
net: stmmac: dwmac-meson8b: propagate rate changes to the parent clock
nfs: Do not convert nfs_idmap_cache_timeout to jiffies
watchdog: sp5100_tco: Fix watchdog disable bit
kconfig: Don't leak main menus during parsing
kconfig: Fix automatic menu creation mem leak
kconfig: Fix expr_free() E_NOT leak
mac80211_hwsim: fix possible memory leak in hwsim_new_radio_nl()
ipmi/powernv: Fix error return code in ipmi_powernv_probe()
Btrfs: set plug for fsync
btrfs: Fix out of bounds access in btrfs_search_slot
Btrfs: fix scrub to repair raid6 corruption
btrfs: fail mount when sb flag is not in BTRFS_SUPER_FLAG_SUPP
HID: roccat: prevent an out of bounds read in kovaplus_profile_activated()
fm10k: fix "failed to kill vid" message for VF
device property: Define type of PROPERTY_ENRTY_*() macros
jffs2: Fix use-after-free bug in jffs2_iget()'s error handling path
powerpc/numa: Use ibm,max-associativity-domains to discover possible nodes
powerpc/numa: Ensure nodes initialized for hotplug
RDMA/mlx5: Avoid memory leak in case of XRCD dealloc failure
ntb_transport: Fix bug with max_mw_size parameter
gianfar: prevent integer wrapping in the rx handler
tcp_nv: fix potential integer overflow in tcpnv_acked
kvm: Map PFN-type memory regions as writable (if possible)
ocfs2: return -EROFS to mount.ocfs2 if inode block is invalid
ocfs2/acl: use 'ip_xattr_sem' to protect getting extended attribute
ocfs2: return error when we attempt to access a dirty bh in jbd2
mm/mempolicy: fix the check of nodemask from user
mm/mempolicy: add nodes_empty check in SYSC_migrate_pages
asm-generic: provide generic_pmdp_establish()
sparc64: update pmdp_invalidate() to return old pmd value
mm: thp: use down_read_trylock() in khugepaged to avoid long block
mm: pin address_space before dereferencing it while isolating an LRU page
mm/fadvise: discard partial page if endbyte is also EOF
openvswitch: Remove padding from packet before L3+ conntrack processing
IB/ipoib: Fix for potential no-carrier state
drm/nouveau/pmu/fuc: don't use movw directly anymore
netfilter: ipv6: nf_defrag: Kill frag queue on RFC2460 failure
x86/power: Fix swsusp_arch_resume prototype
firmware: dmi_scan: Fix handling of empty DMI strings
ACPI: processor_perflib: Do not send _PPC change notification if not ready
ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs
bpf: fix selftests/bpf test_kmod.sh failure when CONFIG_BPF_JIT_ALWAYS_ON=y
MIPS: generic: Fix machine compatible matching
MIPS: TXx9: use IS_BUILTIN() for CONFIG_LEDS_CLASS
xen-netfront: Fix race between device setup and open
xen/grant-table: Use put_page instead of free_page
RDS: IB: Fix null pointer issue
arm64: spinlock: Fix theoretical trylock() A-B-A with LSE atomics
proc: fix /proc/*/map_files lookup
cifs: silence compiler warnings showing up with gcc-8.0.0
bcache: properly set task state in bch_writeback_thread()
bcache: fix for allocator and register thread race
bcache: fix for data collapse after re-attaching an attached device
bcache: return attach error when no cache set exist
tools/libbpf: handle issues with bpf ELF objects containing .eh_frames
bpf: fix rlimit in reuseport net selftest
vfs/proc/kcore, x86/mm/kcore: Fix SMAP fault when dumping vsyscall user page
locking/qspinlock: Ensure node->count is updated before initialising node
irqchip/gic-v3: Ignore disabled ITS nodes
cpumask: Make for_each_cpu_wrap() available on UP as well
irqchip/gic-v3: Change pr_debug message to pr_devel
ARC: Fix malformed ARC_EMUL_UNALIGNED default
ptr_ring: prevent integer overflow when calculating size
libata: Fix compile warning with ATA_DEBUG enabled
selftests: pstore: Adding config fragment CONFIG_PSTORE_RAM=m
selftests: memfd: add config fragment for fuse
ARM: OMAP2+: timer: fix a kmemleak caused in omap_get_timer_dt
ARM: OMAP3: Fix prm wake interrupt for resume
ARM: OMAP1: clock: Fix debugfs_create_*() usage
ibmvnic: Free RX socket buffer in case of adapter error
iwlwifi: mvm: fix security bug in PN checking
iwlwifi: mvm: always init rs with 20mhz bandwidth rates
NFC: llcp: Limit size of SDP URI
rxrpc: Work around usercopy check
mac80211: round IEEE80211_TX_STATUS_HEADROOM up to multiple of 4
mac80211: fix a possible leak of station stats
mac80211: fix calling sleeping function in atomic context
mac80211: Do not disconnect on invalid operating class
md raid10: fix NULL deference in handle_write_completed()
drm/exynos: g2d: use monotonic timestamps
drm/exynos: fix comparison to bitshift when dealing with a mask
locking/xchg/alpha: Add unconditional memory barrier to cmpxchg()
md: raid5: avoid string overflow warning
kernel/relay.c: limit kmalloc size to KMALLOC_MAX_SIZE
powerpc/bpf/jit: Fix 32-bit JIT for seccomp_data access
s390/cio: fix ccw_device_start_timeout API
s390/cio: fix return code after missing interrupt
s390/cio: clear timer when terminating driver I/O
PKCS#7: fix direct verification of SignerInfo signature
ARM: OMAP: Fix dmtimer init for omap1
smsc75xx: fix smsc75xx_set_features()
regulatory: add NUL to request alpha2
integrity/security: fix digsig.c build error with header file
locking/xchg/alpha: Fix xchg() and cmpxchg() memory ordering bugs
x86/topology: Update the 'cpu cores' field in /proc/cpuinfo correctly across CPU hotplug operations
mac80211: drop frames with unexpected DS bits from fast-rx to slow path
arm64: fix unwind_frame() for filtered out fn for function graph tracing
macvlan: fix use-after-free in macvlan_common_newlink()
kvm: fix warning for CONFIG_HAVE_KVM_EVENTFD builds
fs: dcache: Avoid livelock between d_alloc_parallel and __d_add
fs: dcache: Use READ_ONCE when accessing i_dir_seq
md: fix a potential deadlock of raid5/raid10 reshape
md/raid1: fix NULL pointer dereference
batman-adv: fix packet checksum in receive path
batman-adv: invalidate checksum on fragment reassembly
netfilter: ebtables: convert BUG_ONs to WARN_ONs
batman-adv: Ignore invalid batadv_iv_gw during netlink send
batman-adv: Ignore invalid batadv_v_gw during netlink send
batman-adv: Fix netlink dumping of BLA claims
batman-adv: Fix netlink dumping of BLA backbones
nvme-pci: Fix nvme queue cleanup if IRQ setup fails
clocksource/drivers/fsl_ftm_timer: Fix error return checking
ceph: fix dentry leak when failing to init debugfs
ARM: orion5x: Revert commit
|
||
|
Randy Dunlap
|
357cf023c0 |
kdb: make "mdr" command repeat
[ Upstream commit
|
||
|
Tejun Heo
|
c458c7c783 |
rcu: Call touch_nmi_watchdog() while printing stall warnings
[ Upstream commit
|
||
|
Richard Guy Briggs
|
85e924bb33 |
audit: return on memory error to avoid null pointer dereference
[ Upstream commit
|
||
|
Peter Zijlstra
|
bc09bf874d |
perf/core: Fix perf_output_read_group()
[ Upstream commit
|
||
|
Davidlohr Bueso
|
3d0632557e |
sched/rt: Fix rq->clock_update_flags < RQCF_ACT_SKIP warning
[ Upstream commit
|
||
|
Song Liu
|
d4271d86a4 |
perf/cgroup: Fix child event counting bug
[ Upstream commit
|
||
|
Arvind Yadav
|
30393949d1 |
workqueue: use put_device() instead of kfree()
[ Upstream commit
|
||
|
David Rientjes
|
79dc8f3865 |
kernel/relay.c: limit kmalloc size to KMALLOC_MAX_SIZE
[ Upstream commit |
||
|
Will Deacon
|
c8723ceed3 |
locking/qspinlock: Ensure node->count is updated before initialising node
[ Upstream commit
|
||
|
Arnd Bergmann
|
b2e949bfba |
x86/power: Fix swsusp_arch_resume prototype
[ Upstream commit
|
||
|
zhongjiang
|
ec1975ac98 |
kernel/signal.c: avoid undefined behaviour in kill_something_info
commit
|
||
|
Gustavo A. R. Silva
|
960828aaa0 |
kernel/sys.c: fix potential Spectre v1 issue
commit
|
||
|
Joel Fernandes
|
7fd40752c3 |
UPSTREAM: sched/fair: Consider RT/IRQ pressure in capacity_spare_wake
capacity_spare_wake in the slow path influences choice of idlest groups, as we search for groups with maximum spare capacity. In scenarios where RT pressure is high, a sub optimal group can be chosen and hurt performance of the task being woken up. Several tests with results are included below to show improvements with this change. 1) Hackbench on Pixel 2 Android device (4x4 ARM64 Octa core) ------------------------------------------------------------ Here we have RT activity running on big CPU cluster induced with rt-app, and running hackbench in parallel. The RT tasks are bound to 4 CPUs on the big cluster (cpu 4,5,6,7) and have 100ms periodicity with runtime=20ms sleep=80ms. Hackbench shows big benefit (30%) improvement when number of tasks is 8 and 32: Note: data is completion time in seconds (lower is better). Number of loops for 8 and 16 tasks is 50000, and for 32 tasks its 20000. +--------+-----+-------+-------------------+---------------------------+ | groups | fds | tasks | Without Patch | With Patch | +--------+-----+-------+---------+---------+-----------------+---------+ | | | | Mean | Stdev | Mean | Stdev | | | | +-------------------+-----------------+---------+ | 1 | 8 | 8 | 1.0534 | 0.13722 | 0.7293 (+30.7%) | 0.02653 | | 2 | 8 | 16 | 1.6219 | 0.16631 | 1.6391 (-1%) | 0.24001 | | 4 | 8 | 32 | 1.2538 | 0.13086 | 1.1080 (+11.6%) | 0.16201 | +--------+-----+-------+---------+---------+-----------------+---------+ 2) Rohit ran barrier.c test (details below) with following improvements: ------------------------------------------------------------------------ This was Rohit's original use case for a patch he posted at [1] however from his recent tests he showed my patch can replace his slow path changes [1] and there's no need to selectively scan/skip CPUs in find_idlest_group_cpu in the slow path to get the improvement he sees. barrier.c (open_mp code) as a micro-benchmark. It does a number of iterations and barrier sync at the end of each for loop. Here barrier,c is running in along with ping on CPU 0 and 1 as: 'ping -l 10000 -q -s 10 -f hostX' barrier.c can be found at: http://www.spinics.net/lists/kernel/msg2506955.html Following are the results for the iterations per second with this micro-benchmark (higher is better), on a 44 core, 2 socket 88 Threads Intel x86 machine: +--------+------------------+---------------------------+ |Threads | Without patch | With patch | | | | | +--------+--------+---------+-----------------+---------+ | | Mean | Std Dev | Mean | Std Dev | +--------+--------+---------+-----------------+---------+ |1 | 539.36 | 60.16 | 572.54 (+6.15%) | 40.95 | |2 | 481.01 | 19.32 | 530.64 (+10.32%)| 56.16 | |4 | 474.78 | 22.28 | 479.46 (+0.99%) | 18.89 | |8 | 450.06 | 24.91 | 447.82 (-0.50%) | 12.36 | |16 | 436.99 | 22.57 | 441.88 (+1.12%) | 7.39 | |32 | 388.28 | 55.59 | 429.4 (+10.59%)| 31.14 | |64 | 314.62 | 6.33 | 311.81 (-0.89%) | 11.99 | +--------+--------+---------+-----------------+---------+ 3) ping+hackbench test on bare-metal sever (Rohit ran this test) ---------------------------------------------------------------- Here hackbench is running in threaded mode along with, running ping on CPU 0 and 1 as: 'ping -l 10000 -q -s 10 -f hostX' This test is running on 2 socket, 20 core and 40 threads Intel x86 machine: Number of loops is 10000 and runtime is in seconds (Lower is better). +--------------+-----------------+--------------------------+ |Task Groups | Without patch | With patch | | +-------+---------+----------------+---------+ |(Groups of 40)| Mean | Std Dev | Mean | Std Dev | +--------------+-------+---------+----------------+---------+ |1 | 0.851 | 0.007 | 0.828 (+2.77%)| 0.032 | |2 | 1.083 | 0.203 | 1.087 (-0.37%)| 0.246 | |4 | 1.601 | 0.051 | 1.611 (-0.62%)| 0.055 | |8 | 2.837 | 0.060 | 2.827 (+0.35%)| 0.031 | |16 | 5.139 | 0.133 | 5.107 (+0.63%)| 0.085 | |25 | 7.569 | 0.142 | 7.503 (+0.88%)| 0.143 | +--------------+-------+---------+----------------+---------+ [1] https://patchwork.kernel.org/patch/9991635/ Matt Fleming also ran cyclictest and several different hackbench tests on his test machines to santiy-check that the patch doesn't harm any of his usecases. Change-Id: I1de1e40a3a92a54a9adc1303ccc98104704c1e35 Cc: Dietmar Eggemann <dietmar.eggemann@arm.com> Cc: Vincent Guittot <vincent.guittot@linaro.org> Cc: Morten Ramussen <morten.rasmussen@arm.com> Cc: Brendan Jackman <brendan.jackman@arm.com> Tested-by: Rohit Jain <rohit.k.jain@oracle.com> Tested-by: Matt Fleming <matt@codeblueprint.co.uk> Signed-off-by: Joel Fernandes <joelaf@google.com> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com> |
||
|
Greg Kroah-Hartman
|
9c3041c524 |
Merge 4.9.102 into android-4.9
Changes in 4.9.102
usbip: usbip_host: refine probe and disconnect debug msgs to be useful
usbip: usbip_host: delete device from busid_table after rebind
usbip: usbip_host: run rebind from exit when module is removed
usbip: usbip_host: fix NULL-ptr deref and use-after-free errors
usbip: usbip_host: fix bad unlock balance during stub_probe()
ALSA: usb: mixer: volume quirk for CM102-A+/102S+
ALSA: hda: Add Lenovo C50 All in one to the power_save blacklist
ALSA: control: fix a redundant-copy issue
spi: pxa2xx: Allow 64-bit DMA
spi: bcm-qspi: Avoid setting MSPI_CDRAM_PCS for spi-nor master
spi: bcm-qspi: Always read and set BSPI_MAST_N_BOOT_CTRL
KVM: arm/arm64: VGIC/ITS: protect kvm_read_guest() calls with SRCU lock
powerpc: Don't preempt_disable() in show_cpuinfo()
signals: avoid unnecessary taking of sighand->siglock
tracing/x86/xen: Remove zero data size trace events trace_xen_mmu_flush_tlb{_all}
netfilter: nf_tables: can't fail after linking rule into active rule list
i2c: designware: fix poll-after-enable regression
powerpc/powernv: Fix NVRAM sleep in invalid context when crashing
mm: don't allow deferred pages with NEED_PER_CPU_KM
s390/qdio: fix access to uninitialized qdio_q fields
s390/cpum_sf: ensure sample frequency of perf event attributes is non-zero
s390/qdio: don't release memory in qdio_setup_irq()
s390: remove indirect branch from do_softirq_own_stack
x86/pkeys: Override pkey when moving away from PROT_EXEC
x86/pkeys: Do not special case protection key 0
efi: Avoid potential crashes, fix the 'struct efi_pci_io_protocol_32' definition for mixed mode
ARM: 8771/1: kprobes: Prohibit kprobes on do_undefinstr
tick/broadcast: Use for_each_cpu() specially on UP kernels
ARM: 8769/1: kprobes: Fix to use get_kprobe_ctlblk after irq-disabed
ARM: 8770/1: kprobes: Prohibit probing on optimized_callback
ARM: 8772/1: kprobes: Prohibit kprobes on get_user functions
Btrfs: fix xattr loss after power failure
btrfs: fix crash when trying to resume balance without the resume flag
x86/amd: don't set X86_BUG_SYSRET_SS_ATTRS when running under Xen
btrfs: fix reading stale metadata blocks after degraded raid1 mounts
x86/nospec: Simplify alternative_msr_write()
x86/bugs: Concentrate bug detection into a separate function
x86/bugs: Concentrate bug reporting into a separate function
x86/bugs: Read SPEC_CTRL MSR during boot and re-use reserved bits
x86/bugs, KVM: Support the combination of guest and host IBRS
x86/bugs: Expose /sys/../spec_store_bypass
x86/cpufeatures: Add X86_FEATURE_RDS
x86/bugs: Provide boot parameters for the spec_store_bypass_disable mitigation
x86/bugs/intel: Set proper CPU features and setup RDS
x86/bugs: Whitelist allowed SPEC_CTRL MSR values
x86/bugs/AMD: Add support to disable RDS on Fam[15,16,17]h if requested
x86/KVM/VMX: Expose SPEC_CTRL Bit(2) to the guest
x86/speculation: Create spec-ctrl.h to avoid include hell
prctl: Add speculation control prctls
x86/process: Optimize TIF checks in __switch_to_xtra()
x86/process: Correct and optimize TIF_BLOCKSTEP switch
x86/process: Optimize TIF_NOTSC switch
x86/process: Allow runtime control of Speculative Store Bypass
x86/speculation: Add prctl for Speculative Store Bypass mitigation
nospec: Allow getting/setting on non-current task
proc: Provide details on speculation flaw mitigations
seccomp: Enable speculation flaw mitigations
x86/bugs: Make boot modes __ro_after_init
prctl: Add force disable speculation
seccomp: Use PR_SPEC_FORCE_DISABLE
seccomp: Add filter flag to opt-out of SSB mitigation
seccomp: Move speculation migitation control to arch code
x86/speculation: Make "seccomp" the default mode for Speculative Store Bypass
x86/bugs: Rename _RDS to _SSBD
proc: Use underscores for SSBD in 'status'
Documentation/spec_ctrl: Do some minor cleanups
x86/bugs: Fix __ssb_select_mitigation() return type
x86/bugs: Make cpu_show_common() static
x86/bugs: Fix the parameters alignment and missing void
x86/cpu: Make alternative_msr_write work for 32-bit code
KVM: SVM: Move spec control call after restore of GS
x86/speculation: Use synthetic bits for IBRS/IBPB/STIBP
x86/cpufeatures: Disentangle MSR_SPEC_CTRL enumeration from IBRS
x86/cpufeatures: Disentangle SSBD enumeration
x86/cpu/AMD: Fix erratum 1076 (CPB bit)
x86/cpufeatures: Add FEATURE_ZEN
x86/speculation: Handle HT correctly on AMD
x86/bugs, KVM: Extend speculation control for VIRT_SPEC_CTRL
x86/speculation: Add virtualized speculative store bypass disable support
x86/speculation: Rework speculative_store_bypass_update()
x86/bugs: Unify x86_spec_ctrl_{set_guest,restore_host}
x86/bugs: Expose x86_spec_ctrl_base directly
x86/bugs: Remove x86_spec_ctrl_set()
x86/bugs: Rework spec_ctrl base and mask logic
x86/speculation, KVM: Implement support for VIRT_SPEC_CTRL/LS_CFG
KVM: SVM: Implement VIRT_SPEC_CTRL support for SSBD
x86/bugs: Rename SSBD_NO to SSB_NO
Linux 4.9.102
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
|
||
|
Thomas Gleixner
|
094c2767c4 |
seccomp: Move speculation migitation control to arch code
commit
|
||
|
Kees Cook
|
ab677c2add |
seccomp: Add filter flag to opt-out of SSB mitigation
commit
|
||
|
Thomas Gleixner
|
c71def81cd |
seccomp: Use PR_SPEC_FORCE_DISABLE
commit
|