PD#SWPL-1767
Problem:
When driver/kernel call kmalloc with large size, memory may waste
if size is not equal to 2^n. For example, driver call kmalloc with
size 129KB, kmalloc will allocate a 256KB memory block to caller.
Then 127kb memory will be wasted if this caller don't free it.
Solution:
Free tail of slab memory if size is not match to 2^n. This change
can save about 900KB memory after boot, and more than 100KB during
run time.
Verify:
P212
Change-Id: Iba378792ec30003358b64384361c0f0c4c2800d8
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
ODROID-COMMON:Merge AMLogic BSP 181208
audioinfo: fix kasan compile problem [1/1]
PD#173553
Problem:
check stack out of bounds
Solution:
change variable's definition
Verify:
verifyed by r311
Change-Id: Id6efc3ac56b297b6812e2af786732e65867a07da
Signed-off-by: jiejing.wang <jiejing.wang@amlogic.com>
buildroot: Enable fb memory allocation in g12b_skt dts [1/1]
PD#SWPL-802
Problem:
Kernel crash when run GPU slt program
Solution:
Enable meson fb memory allocation in g12b_skt dts
Verify:
w400
Change-Id: I3074c2053cda06cf4ae237d24bad2c89d789f90d
Signed-off-by: Blance Tang <guofeng.tang@amlogic.com>
amvecm: hdr: add hdr10 plus support [1/2]
PD#SWPL-869
Problem:
new feature, add hdr10 plus support
Solution:
1. add hdr10 plus metadat parser
2. add hdr10 plus code support
3. add hdr10 plus bypass mode
Verify:
verify on U200
Change-Id: I957954ce148021882e17f2913dd2552e64f1298c
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
irqchip: add a new gpio IRQ driver to support double-edge detection [1/1]
PD#SWPL-922
Problem:
the existing gpio IRQ driver(porting from the upstream) does not
double-edge detection
Solution:
add a new gpio IRQ driver to support the double-edge detection,
the new driver is with different software structure but compatible
with functions of the existing driver
Verify:
test pass on s400
Change-Id: Id69dd0b4459eef20a6755947c7d95a24d7b17fa9
Signed-off-by: Xingyu Chen <xingyu.chen@amlogic.com>
pinctrl: fix some mistakes [1/1]
PD#172438
Problem:
1. return a negative when parsing the dts property
"output-high/output-low"
2. the prefix "bt565" don't meet corepinmux document
Solution:
1. add break statement in case PIN_CONFIG_OUTPUT to avoid
returning a negative value.
2. rename the "bt565" to "bt656" according to corepinmux document
Verify:
test pass on U200
Change-Id: Ieeb98367e2269007a6774592abac86ef05ee8f64
Signed-off-by: Xingyu Chen <xingyu.chen@amlogic.com>
amvecm: fix flicker when change between Launcher and Signal Source [1/1]
PD#SWPL-848
Problem:
ioctrl set the same gamma value again
Solution:
if the gamma value is same as the ioctrl userspace value, do not set again
Verify:
txlx
Change-Id: I6f17f5ff234513c5a886004aedea81b9945c5b98
Signed-off-by: Bencheng Jing <bencheng.jing@amlogic.com>
vpu: add support for tl1 [1/1]
PD#172587
Problem:
Bringup vpu for TL1.
Solution:
Add vpu support for TL1.
Verify:
Verified on PTM/PXP.
Change-Id: Ic2fb7682ddc1b21840e1be1d8462d4c0902e739d
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
lcd: add tl1 support [1/1]
PD#172587
Problem:
not support tl1
Solution:
add tl1 support
Verify:
test pass on PTM
Change-Id: I5a9073fc08a98ae95d4961297a7427864d92279b
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
backlight: add support for tl1 [1/1]
PD#172587
Problem:
do not suppoert for tl1
Solution:
add backlight support for fl1
Verify:
verified on PTM
Change-Id: I052a7ef93ec9cb21a544e4823a4c9a339d38b5f9
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
arm: dts: tl1: add decoder device tree [3/3]
PD#174543
PD#172587
Problem:
new added function for tl1
Solution:
add vdin afbce support for tl1
Verify:
verified on PTM
Change-Id: I42054b8228161713ac10446491ebdb50a71975ad
Signed-off-by: shihong.zheng <shihong.zheng@amlogic.com>
media: add codec support for tl1 [1/1]
PD#172587
Problem:
Bringup codec for TL1.
Solution:
Add codec iomap device tree node for TL1.
Add canvas device tree node for TL1.
Verify:
Verified on PXP/PTM.
Change-Id: I8b9a6645f1f9fd7d4aa9cae6166895e61ddc3e36
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
rdma: add rdma support for tl1 [1/1]
PD#172587
Problem:
Bringup rdma for TL1.
Solution:
Add rdma support for TL1.
Verify:
Verified on PXP/PTM.
Change-Id: I425edc1e47db2ea834fcc5acb0e3d0ee1f30a9f7
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
osd: tl1: add osd support for tl1 [1/1]
PD#172587
Problem:
Bringup osd for TL1.
Solution:
Add osd support for TL1.
Add ge2d device tree node for TL1.
Verify:
Verified on PXP/PTM.
Change-Id: I3d20934f015108f545c3c72b979d1bd8187b282e
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
amvideo: add amvideo support for tl1 [1/1]
PD#172587
Problem:
TL1 needs support amvideo driver.
Solution:
1. Add amvideo node into dts file, default enable.
2. Change the vd mif and sr reg offset.
3. Force bypass dolby vision function.
4. Change the register following hardware design.
5. Make amvideo afbc work for tl1.
Verify:
On PTM.
Change-Id: I0d857f8a707328383328daa23e6ba8e156c2e6e0
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
audio: auge: add sound card support for tl1 [1/1]
PD#172587
Problem:
Bringup tl1 sound card.
Solution:
Add tl1 sound card.
Add external interface for audio input/output.
Verify:
Tested by PTM
Sound card is setup.
TDM and SPDIF internel loopback is ok
Change-Id: I60830ca44a62ee2a8e16343e91e7311152cab161
Signed-off-by: Xing Wang <xing.wang@amlogic.com>
vdac: vdac: add vdac support for tl1 [1/1]
PD#172587
Problem:
do not suppoert for tl1
Solution:
add vdac support for fl1
Verify:
test pass on local
Change-Id: I7b3da34289dfe0638a24b812658db7e462ebdd4c
Signed-off-by: Nian Jing <nian.jing@amlogic.com>
avin: add avin detect support for tl1 [1/1]
PD#172587
Problem:
do not suppoert for tl1
Solution:
add avin detect support for fl1
Verify:
just coding for tl1 will test later
Change-Id: I809f7068c9d0a45c89bd7dc4e6615db99b11f015
Signed-off-by: Nian Jing <nian.jing@amlogic.com>
tvin: add vdin and viu support for tl1 [1/1]
PD#172587
Problem:
do not suppoert for tl1
Solution:
add vdin and viu support for fl1
Verify:
verified on PTM
Change-Id: If4e267356fa666541853cfc7b49008795f9ec301
Signed-off-by: xuhua zhang <xuhua.zhang@amlogic.com>
vdin: add afbce function for tl1 [1/1]
PD#172587
Problem:
new added function for tl1
Solution:
add vdin afbce support for tl1
Verify:
verified on PTM
Change-Id: I5e3e2487b6cdbed2d2ca794a6a21ed885e4bf469
Signed-off-by: xuhua zhang <xuhua.zhang@amlogic.com>
hdmirx: hdmirx driver for tl1 [1/1]
PD#172587
Problem:
hdmirx bringup for tl1
Solution:
1.modify write top api, add addr offset -qy
2.modify write edid api -qy
3.replace top hw reset api
4.add aocec access register idle -qy
5.for tl1 bypass sw eq -qy
6.replace pddq api -qy
7.modify arc control for tl1 -qy
8.add n/cts auto mode
9.add hdcp balance path
10.add sscp mode
11.optimize reg map(rm unused reg addr)
12.add recent commits from 4.9 trunk
13.modify top clk cntl bit definition
14.modify top offset addr handle(x4)
15.add audio pll setting -qy
16.update phy setting - qy
17.modify audio pll ctrl api for tl1 -qy
-add clk monitor function
-add clk stable api for tl1
18.add emp data to pfifo -qy
19.modify edid write and other function -qy
20.add emp data to ddr -qy
21.add tmds data to ddr -qy
22.add cec for tl1 -qy
-fix ctc 7-1
-add status register for cec a/b
23.add cec and hdmirx dts -qy
24.clean tl1 rx related clk measure
25.hdmirx and cec dts
Verify:
1.run PTM
2.need verify on chip
Change-Id: Ia7cc5a2d84925587bdfae825936ba763713926af
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
audio: codec: add tl1_acodec support for tl1 [1/1]
PD#172587
Problem:
Add aml_tl1_acodec driver support for tl1
Solution:
1.add files sound/soc/codecs/amlogic/aml_codec_tl1.acodec.c(.h)
2.update files sound/soc/codecs/amlogic/Kconfig(Makefile)
Verify:
kernel build passed, need verified after chipback
Change-Id: I19717a442f22f5b6844627475992ef7c8cc20a71
Signed-off-by: shuyu.li <shuyu.li@amlogic.com>
hdmirx: add hdmirx repeater function [1/6]
PD#SWPL-323:
Problem:
add hdmi repeater function
Solution:
1.add the edid receive and mix.
2.add the flow of repeater.
Verify:
R321
Change-Id: I9942c5f345e2fdfff110f01d4d0c2b4b23120c07
Signed-off-by: hongmin hua <hongmin.hua@amlogic.com>
ddr: fix bandwidth read result 0 on 32bit OS [1/1]
PD#SWPL-960
Problem:
Reading bandwidth of DDR from sysfs get all 0 result on 32
bit kernel
Solution:
Fix overflow when calculating bandwidth.
Verify:
P212
Change-Id: I35837db653bdc2d97ced98689546a9ffc0db21c7
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
mm: reduce watermark if free cma is too large [1/1]
PD#SWPL-807
Problem:
Sometimes driver can't allocation memory under atomic environment.
And free pages is enough but they are nearly ALL CMA pages.
Solution:
Reduce watermark with harf of free cma pages even allocation support
CMA.
Verify:
P212
Change-Id: I8e49768d4384ed064775537754a2b7f09a5bbb7c
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
avb: avb enable need to translate more types [1/1]
PD#SWPL-1095
Problem:
avb need more command line size.
Solution:
change max size to 2048
Verify:
verify by ampere
Change-Id: I9c1089a3ea0888e8ccd82f16d2b1111b6d88922a
Signed-off-by: Luan Yuan <luan.yuan@amlogic.com>
mm: avoid pages migrated to different zone [1/1]
PD#SWPL-881
Problem:
Amlogic modified code when compatcion from normal case.
And side effect is that pages in normal zone maybe migrated to
highmem zone. Which caused page_address get a NULL value and
kernel panic occurred.
Solution:
Avoid pages migarte to different zone by adding forbid_to_cma
flags.
Verify:
P212
Change-Id: I1d9c6653dc1069562db3c1be3f53a3510a51f0d1
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
usb: adb reboot and then adb disconnect [1/1]
PD#174155
Problem:
complete_ep 0xffffffc05bed2858, ep->queue empty!
1. After adb process be killed, data buffer is freed and
this memory is allocated for the other. But the address
is hold by the controller.
2. Adbd in PC is running. So, the controller receive the
data and write to this memory.
3.The value of this memory is modified by the controller.
This could cause the memory problem.
Solution:
whenever io_data->aio equals 1, the data buffer is from a fixed array.
Verify:
Test: adb devices, adb shell, exit, adb push
in android8(p212) and android9(w400)
verified by he he
Change-Id: Idac755d3646639e2944a82f42abf9adb9aeaea8c
Signed-off-by: he.he <he.he@amlogic.com>
dts: support 32bit for gxm_q201 [1/1]
PD#SWPL-1140
Problem:
add support 32bit for gxm_q201.
Solution:
add 32bit dts for gxm_q201.
Verify:
gxm_q201
Change-Id: I3c9a6ae72fad6575bd814818967d9ff80e62f5d4
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
amvecm: pq: add amvecm support for tl1 [1/1]
PD#172587
Problem:
new feature, add amvecm support for tl1
Solution:
1. add cm hist for tl1
2. add 3dlut for tl1
3. add hdr support for tl1
4. add wb for tl1
5. add local contrast
Verify:
verify on tl1 ptm
Change-Id: I1c7ebb83a1fb72a4529415fb9bf4acfd134e6b11
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
dts: add ir cut control GPIO [1/1]
PD#SWPL-900
Problem:
add ir cut control
Solution:
add ir cut control GPIO
Verify:
g12b-skt
Change-Id: I16c7d03ed4cc1dd3842cab80a987dde1a83e5cbc
Signed-off-by: yu.zhang <yu.zhang@amlogic.com>
hdmitx: add init code for repeater [2/6]
PD#SWPL-323
Problem:
For TXLX/T962E, it has both HDMI Rx and TX, and lacks of HDMI repeater
functions, including HDCP repeater function.
Solution:
Add the init code for hdmi repeater
Verify:
TXLX/T962E/R321
Change-Id: Iaf17ae62c590ff4f8478dd5556f3ed24b9ff3bb1
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
atv_demod: switch channel with the script, the ATV shows no signal [1/1]
PD#SWPL-791
Problem:
Switch channel with the script lasts 14 hours, the ATV shows no signal.
Solution:
To avoid conflict, don't need to check the CLK
and PLL states in atv demod write and read function.
Verify:
verified by p321
Change-Id: I033c54f14541ed540dfce1b9c4e21d2b5a8b9487
Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com>
media: keep last normal frame before reset [2/2]
PD#OTT-71
Problem:
google cast display green lump when seek.
reset after seek opration.
Solution:
keep last normal frame before reset and buf free.
Verify:
U211, Atom
Change-Id: I02304998c9434f1055de1138700510f4c19f466f
Signed-off-by: shihong.zheng <shihong.zheng@amlogic.com>
unifykey: add the hdmirx repeater key block [3/6]
PD#SWPL-323:
Problem:
add the hdmirx repeater key block
Solution:
add the hdmirx repeater key block.
Verify:
R321
Change-Id: I61920b706b5dd08c6be9647add287f85b35fb0ee
Signed-off-by: hongmin hua <hongmin.hua@amlogic.com>
mm: make sure the gtp flag is low default [1/1]
PD#SWPL-807
Problem:
The app's window cound be black screen when launched if the ion system
does not retry to allocate memory with the gtp high flag.
Solution:
make sure the gtp flag is low default
Verify:
P212
Change-Id: I562125573f8175cd180196ca73eb04d7dd36add0
Signed-off-by: an.xi <an.xi@amlogic.com>
dts: emmc: uboot enable emmc config [2/3]
PD#SWPL-736
Problem:
The s400 emmc platform uboot cannot be loaded
into kernel normally
Solution:
enable enable emmc node in dtb by fdt command.
and modify malloc max size to 18M.
Verify:
s400
Change-Id: Iaef58f1b05a952793d414eb82ac476d88d910fe1
Signed-off-by: yuegui.he <yuegui.he@amlogic.com>
dts: GVSDK: modify loopback and led device name [1/1]
PD#173996
Problem:
1. to support hw loopback channel 2,3 for GVSDK.
2. led is31fl32xx device name register fail.
Solution:
1. change boolback to channel 2,3
2. register led is31fl32xx name success.
Verify:
S400_GVSDK/S420_GVSDK
Change-Id: Ie1bbc01c4624fd1e2e2662c4d9c876e56c301a99
Signed-off-by: Renjun Xu <renjun.xu@amlogic.com>
mm: exclude free cma pages when calculate watermark [1/1]
PD#SWPL-1210
Problem:
Some drivers(eth/wifi) occasionally can't allocate memory under
atomic context. From mem status print, there are enough free
pages but most of them are CMA.
Solution:
Exclude free cma pages when calculate water mark. This can push
kswapd/compaction work more aggressive.
Verify:
P212
Change-Id: Ia723f21c0316eff1a38e759ff9f044bb59aa8efa
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
dv: set stb core run_mode_delay as 0 [1/1]
PD#SWPL-781
Problem:
when starting to play video, the manu bar shakes
Solution:
stb core don't need run mode delay.
Verify:
t962e (txlx)
Change-Id: I22ba5d778eedd72f87b4c687b38acf1b8c6d6fc9
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
dv: fix the green screen when changing dv mode [1/1]
PD#SWPL-1029
Problem:
when changing dv mode from standard to ll mode.
wn tp vks should be enabled
Solution:
enable wn tp vks when setting ll mode
Verify:
t962e (txlx)
Change-Id: I3b399132cf1a3c7b1799f36f9bc3a0696883029f
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
lcd: lcd_extern: fix get invalid extern_driver null pointer mistake [1/1]
PD#SWPL-1152
Problem:
There is a null pointer mistake in lcd_extern_get_driver for invalid driver
Solution:
add null pointer protection
Verify:
txlx r311
Change-Id: I171d10ab26e6c298b8eb4065dfa054a237bfd53b
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
crypto: Enable ARMCE and set aes dma to slow mode [1/1]
PD#SWPL-574
Problem:
xfrm test failed randomly in VtsKernelNetTest
due to random failed in gcm(aes)
Solution:
Use ARMCE for small sized crypto processing and use
copy mode when HW dma is used
Verify:
verified on ampere
Change-Id: I31cd75cfcd85da9fc9e9640135c7ce76623ef715
Signed-off-by: Matthew Shyu <matthew.shyu@amlogic.com>
USB: host: tuning disconnect threshold on g12a/g12b [2/2]
PD#174736
Problem:
Some usb device may disconnect when handshake
Solution:
tuning disconnect threshold when usb host initial
now default is 0x24, only three value can tuning, 0x2c,0x34,0x3c
we set 0x34 now
Verify:
Google estelle boards which return from factory
no side effect on refernce u200 board
Change-Id: I71d974a6bac43452b20bd7f2a6827d3a46fa39fb
Signed-off-by: qi duan <qi.duan@amlogic.com>
audio: solve av out output level low issue [2/2]
PD#SWPL-766
Problem:
av output level only 100mv
Solution:
change the channel map for av out,
use orignal pcm for output w/o volume
control and postprocess.
Verify:
verified by r311
Change-Id: Ibeaf0ecfda03f81ff6061dc9c4975049452ba659
Signed-off-by: Lianlian Zhu <lianlian.zhu@amlogic.com>
Signed-off-by: Jian Xu <jian.xu@amlogic.com>
chipid: meson: add ops id for S805Y [1/2]
PD#SWPL-940
Problem:
Need ops id api for S805Y
Solution:
Add ops id api
Verify:
s805y_p215
Change-Id: I179f87acbc3f57456e0359bd0d502c58cc3a9b9c
Signed-off-by: Qiufang Dai <qiufang.dai@amlogic.com>
hdmitx: add feature limited [3/3]
PD#174490
Problem:
Lack feature limited for certain types
Solution:
Add feature limited for certain types
Verify:
TBD
Change-Id: Ifbe104fe98fe1e3875f2f1af7793348856275b92
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
dts: support 32bit for g12b_w400 [1/1]
PD#SWPL-1010
Problem:
add support 32bit for g12b_w400
Solution:
add 32bit dts of g12b_w400.
Verify:
g12b_a311d_w400
Change-Id: I30ac9cb80721b04a13532ceb5c14f76bdf1cc575
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
display: dts modify [1/1]
PD#TV-405
Problem:
Should create a 720p ui display mode on p321.
Solution:
Created a folder named t962_p321_720p under device/amlogic
and modified some variate.
The diffs that compare with txl_t962_p321.dts
amlogic-dt-id = "txl_p321_v1_720p";
size = <0x400000>;
alloc-ranges = <0x3fc00000 0x400000>;
mem_size = <0x00400000 0x00b00000 0x00100000>;
/* uboot logo,fb0/fb1 memory size,if afbcd fb0=0x0b51000*/
display_mode_default = "720p60hz";
/*1280*720*4*3 = 0xA8C000*/
display_size_default = <1280 720 1280 2160 32>;
logo_addr = "0x3fc00000";
Verify:
p321
Change-Id: Ia397cd6cbcf4fdd9fb139259099805bdad90b92f
Signed-off-by: Guotai Shen <guotai.shen@amlogic.com>
dts: support 32bit for gxl_p230 [1/1]
PD#SWPL-1116
Problem:
add support 32bit for gxl_p230.
Solution:
add 32bit dts for gxl_p230.
Verify:
gxl_p230
Change-Id: Icc62c45e7bd36e8248ad0bb11a67a0151cf7ea34
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
dts: sync difference between arch32 and arch64 [1/1]
PD#SWPL-959
Problem:
sync arch32 and arch64 dts.
Solution:
sync arch32 and arch64 dts.
Verify:
r311
Change-Id: I493c83f38899628ac65c637e81f053a7c22bd24f
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
cec: for support multi-logical address [1/2]
PD#SWPL-418
Problem:
cec: for support mult-logical address
Solution:
1.add interface: remove logical address
2.for support multi-logical address
Verify:
r311 r321
Change-Id: I9ea8b1004f43fb84855d41dd684c117fa5cbd7ae
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
cec: for support multi-logical address [2/2]
PD#SWPL-418
Problem:
cec: support multi-logical address
Solution:
1.enable cec_a, cec_b
2.enable two interrupt
3.enable two pinmux oa_7, ao_8
4.cec_a only send all msg
5.cec_b only receive all msg
6.discard ceca broadcast msg
Verify:
r311 r321
Change-Id: I8f983ed6ad329ca5ec0144587a7ad1f03ad68031
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
emmc: ffu: add emmc ffu update support [1/1]
PD#174683
Problem:
kernel emmc ffu is not supported.
Solution:
add emmc ffu update support with mmc untils.
Verify:
p212
Change-Id: Ice9ffb174c061e3f114b0b68af290492808b0a5e
Signed-off-by: Nan Li <nan.li@amlogic.com>
codec_mm: add reserved & cma support for 4.9. [1/1]
PD#172483
Problem:
Add reserved & cma support for 4.9.
Solution:
1. add tvp padding, for test.
2. add no-cma-tvp,to force not used cma for tvp.
3. add nomap for reserved memroy detective.
4. add res & cma two mem region on same board.
5. add clear-map property for codec_mm_cma.
Verify:
Test this function with Android_O-MR1(GTVS) & ExoPlayer.
Change-Id: Ib20e0e9bc4725afb271de97543ce5ebf3bfcfb29
Signed-off-by: Zhi Zhou <zhi.zhou@amlogic.com>
mm: clear MMU mapping for secure os memory [1/1]
PD#SWPL-1631
Problem:
If Cortex A73 accessed memory near secure memory range, an
SError may occur due to prefetch of A73 core.
Solution:
Remove mmu maping of secure memory range.
Note: must add clear-map and set clear range property in DTS
Verify:
W400
Change-Id: I718919c1b4873584eb87e00332622a1626672808
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
cec: revert changesy [1/1]
PD#SWPL-418
Problem:
TXL can't bootup
Solution:
revert it
Verify:
p321
Revert "cec: for support multi-logical address [2/2]"
This reverts commit cc185dc81d.
Revert "cec: for support multi-logical address [1/2]"
This reverts commit b7922078ea.
Change-Id: I1cef0ac194464d75ffff3fe765f15b5f944406b2
Signed-off-by: Lei Qian <lei.qian@amlogic.com>
aarch64: enable swp software emulation [1/1]
PD#IPTV-14
Problem:
3D Benchmark.apk crash
Solution:
Add swp instruction support on armv8
Verify:
local
Change-Id: Ice2bef5c0ef7110b3adb4af811f99d4bfca1e8dc
Signed-off-by: Jianxin Pan <jianxin.pan@amlogic.com>
emmc: Modify delay of command line to 1/4 command window [1/1]
PD#SWPL-1184
Problem:
emmc data crc error
Solution:
Modify delay of command line
Verify:
A113d_s400
Change-Id: I7211da9da75569aebaf0c2790e3134e88a6f3077
Signed-off-by: Ruixuan Li <ruixuan.li@amlogic.com>
adc_key: delete early_suspend code [1/1]
PD#OTT-144
Problem:
after the dark screen, the adc button does not work
Solution:
The adc_keypad introduces earlysuspend for MID project:
after the LCD backlight disable, prevent the user from
pressing the key to open some applications and not complete standby.
Now check the adc button using the "system_freezable_wq" system queue,
which replaces the function of the code.So can remove it.
Verify:
g12b socket
Change-Id: I2333ba3dc406d2624962d116293a0b9d36f97534
Signed-off-by: Yingyuan Zhu <yingyuan.zhu@amlogic.com>
audio: fix errors of coverity [1/1]
PD#166793
Problem:
coverity check errors of format
Solution:
fix err
Verify:
verify locally
Change-Id: Ie3a404fe50c5bdaa6a9cf4a4194659502defdbd2
Signed-off-by: jiejing.wang <jiejing.wang@amlogic.com>
hdmirx: fix sound noise when switch hdmi port [1/1]
PD#SWPL-1224
Problem:
There will sound noise when switch hdmi port.
Solution:
Remove bandgap off when close port.
Verify:
Locally
Change-Id: Ibd9be3aa6f32072918f7b1ec151463452b2b879a
Signed-off-by: yicheng shen <yicheng.shen@amlogic.com>
mm: clear MMU mapping for secure os memory [1/1]
PD#SWPL-1631
Problem:
If Cortex A73 accessed memory near secure memory range, an
SError may occur due to prefetch of A73 core.
Solution:
Add clear mmu property for arm platform
Verify:
W400
Change-Id: I2e396f77772aeea231d0bf5f38a81440117ecdd6
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
video: merged from A wait AFBC idle when report state [1/1]
PD#169128:
[Problem]
When quick switch video between 2 vsync isr, the AFBC may read wrong
data, and green screen or distortion image occurs.
[Solution]
1. Add delay when get video layer state, to make sure afbc is stopped
2. Remove ambiguous log print
[Platform]
Blanche
BranchTo: NONE
[Test]
Change-Id: I94ccf25373f29ce188829ab1b7db6f9df1fb49ad
Signed-off-by: Zhi Zhou <zhi.zhou@amlogic.com>
dts: add isp and galcore support to w400-buildroot [1/2]
PD#SWPL-1819
Problem:
lack of isp and galcore support in buildroot dts
Solution:
merge configurations from skt board
Verify:
local
Change-Id: I5936341fc24719f345c7707e16e692ae292ca6fe
Signed-off-by: Jun Zhang <jun.zhang@amlogic.com>
dts: modified atom.dts support 32bit. [1/1]
PD#SWPL-1942
Problem:
add atom.dts support 32bit.
Solution:
modified length of reg, mem etc.
Verify:
r321
Change-Id: I2a518af2a76b8cb17ded68e994c3065072b92b9c
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
vccktest: add vccktest dts for buildroot [1/1]
PD#174851
Problem:
buildroot no vccktest dts
Solution:
add vccktest dts
Verify:
g12a/g12b test pass
Change-Id: I6907fa3cfc087ac84395956ddbda678f7a5e3f4b
Signed-off-by: Huan Biao <huan.biao@amlogic.com>
script: segments can appear in descriptive information [1/1]
PD#174488
Problem:
segmentation in descriptive information is reasonable and
should be allowed.
Solution:
modify the check codes.
Verify:
test locally.
Change-Id: I5e34ef2dd26077bb54adee7597a1aa3c789566c9
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
vlock: optimize vlock process for interlace input [1/1]
PD#SWPL-743
Problem:
The picthre is shaking when press DVD remote
control in AV source
Solution:
add delay work for vlock process interlace input
Verify:
T962X-R311
Change-Id: Ic2ed077e0684ecc6b0990336e7a4ab7022931886
Signed-off-by: Dezhi Kong <dezhi.kong@amlogic.com>
atv_demod: pc bad value in panic after change source [1/1]
PD#SWPL-1636
Problem:
1.pc bad value in kernel panic after change source.
2.System crash down when change source.
Solution:
optimize atv demod code.
Verify:
verified by p321
Change-Id: I655d1253950475dfb49ae1b174597797000adb73
Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com>
vout: add bist test on vout [1/1]
PD#SWPL-1630
Problem:
add cvbs/hdmi/lcd bist test on vout
Solution:
add cvbs bist test on vout
Verify:
test pass on u200/p212
Change-Id: Ie61d9b0d63649ec3022fd720068b689f4d721e3b
Signed-off-by: Nian Jing <nian.jing@amlogic.com>
osd: meson-hwc screen always blank [1/1]
PD#163001
Problem:
TL1 code merged caused meson-hwc screen always blank
Solution:
fix meson-hwc screen always blank
Verify:
verified in u200 board
Change-Id: I0a9371b42546e2713e6dcb926e7c94598e29da1d
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
osd: osd shift-workaround one line fixed [1/1]
PD#163001
Problem:
current osd code have one line missing issue in meson-hwc
Solution:
osd blend used +1 line to fix shift issue,
and used freescale to cut it.
Verify:
verified in u200 board
Change-Id: I2354742d456327d9f1aa5713ee782648895034db
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
smartcard: fix defects for coverity [1/1]
PD#166793
Problem:
conditional statement not is valid.
Solution:
change type u32 to type int.
Verify:
module owner confirmed.
Change-Id: I9fdfa444b29104c7afc9b7a380748ebe82ed390b
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
osd: uboot logo scaler down in 4k issue [1/1]
PD#163001
Problem:
uboot logo scaler down in 4k
Solution:
uboot logo size adjust when vout changed before
first page flip command
Verify:
verified in u200 board
Change-Id: Ic23e695fe716811eaf3faa62069c808a03c26f9d
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
osd: add hdr mode set via meson-hwc support [1/1]
PD#163001
Problem:
hdr mode set not supported
Solution:
add hdr mode set via meson-hwc support
Verify:
verified in u200 board
Change-Id: Iaad75982003870f9469e4155aedb9c13349b60b9
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
dts: fix some errors on g12b_w400 for 32bit [1/1]
PD#SWPL-1010
Problem:
some errors of reg address.
Solution:
modify to support 32bit.
Verify:
w400.
Change-Id: Ia35e95ac59084a9c57a9a4ab32fecb64f78c772e
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
ddr: detect bandwidth except mali and reduce bandwidth [2/3]
PD#SWPL-122
Problem:
ddr bandwidth is tightly in many cases, such as HDR/4K video playing,
and then refresh osd at the same time
Solution:
in most cases the problem is refreshing osd when bandwidth tightly,
so detect ddr bandwidth usage except mali and reduce it when it is tightly
Verify:
verify by p321/r311
Change-Id: Ia594f08dbe9a8a7a8d78b5104c93bddaa8229c0b
Signed-off-by: wenbiao zhang <wenbiao.zhang@amlogic.com>
arm: dts: gxl: replace watchdog with new version [1/1]
PD#SWPL-1600
Problem:
The current watchdog is not the upstream standard driver.
Solution:
Replace watchdog device tree to use the standard watchdog driver.
Verify:
S905X P215/P212 Android-32bit Kernel-32bit.
Change-Id: I972c8ebb0158caa2df393d9a2d687d9298181102
Signed-off-by: Bo Yang <bo.yang@amlogic.com>
arm64: dts: g12a: fine-tune voltage OPP [1/1]
PD#165143
Proble：
fine-tune voltage OPP
Solution:
fine-tune voltage OPP
Verify:
G12A/G12B,PASS
Change-Id: I0b3fb0e4624e2ddfa635c8225a740ed7a8ff3e51
Signed-off-by: Hong Guo <hong.guo@amlogic.com>
audio: fix audio driver del_timer BUG_ON crash issue [1/1]
PD#SWPL-972
Problem:
audio crash when ATV switch channel long time burning test
Solution:
there is a risk to fetch the timer lock. when input stop, stop_timer
will call del_timer, at the same time the function
"aml_i2s_hrtimer_callback" is waiting for the timer lock,after
stop_timer release the lock,"aml_i2s_hrtimer_callback" get the
lock and call mod_timer again, which will set the timer to pending
status. It will cause the next "start input" stage,add_timer will
trigger BUG_ON.Now we put the lock before the active status
checking then we will not touch the timer.
Verify:
Need burning test
Change-Id: I1fb66903a4d31e9491ac0533e477e1597575d4cf
Signed-off-by: Jian Xu <jian.xu@amlogic.com>
hdmitx: enable repeater_tx in r321/atom dts [2/3]
PD#SWPL-1303
Problem:
For TXLX/T962E/R321 cases, there are both HDMI TX and RX.
Should consider it as repeater.
Solution:
Enable repeater_tx in r321/atom dts
Also, add a missing ksv_lists[] clear.
Verify:
r321/atom
Change-Id: I7b512c34123481380220f9262466fde56b0ae6e3
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
hdmitx: remove a wrong Y422 format judgement [1/1]
PD#IPTV-55
Problem:
When connect to a Samsung TV(support 4k60 422 12bit), get a wrong
0 of 'echo 2160p60hz422,12bit > valid_mode'
Solution:
Remove a wrong Y422 format judgement
Verify:
P212
Change-Id: Ic0f2cefc1e0893390ac38298b55272877d81e04b
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
tsync: tsync change the apts set code [1/2]
PD#SWPL-1774
Problem:
DD program has avysnc problem over one night play
Solution:
fix the apts set bug
Verify:
p321
Change-Id: I0f6a5b47d3e3171a3dd8fa9a0b721cd11ec20b9e
Signed-off-by: live.li <live.li@amlogic.com>
amvecm: overscan: fix video local play, screen mode set error [1/1]
PD#SWPL-206
Problem:
overscan set error when local play
Solution:
overscan don't set at vpp
Verify:
verify on txlx
Change-Id: Ifd4ad4f5fcdb9fbc7ad05a540d0ed11b24d4ca66
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
thermal: sensor: fix thermal disabled mode issue [1/1]
PD#SWPL-2130
Problem:
tsensor driver update thermal event when mode disabled
Solution:
fix tsensor disable mode issue
Verify:
A311D W400 pass
Change-Id: If9e3d3ea15732a0efbd8314003b27bc0bdf3e476
Signed-off-by: Huan Biao <huan.biao@amlogic.com>
deinterface: add afbc support on g12a and txlx [1/1]
PD#SWPL-1618
Problem:
Add afbc support on g12a and txlx
Solution:
add afbc support on g12a and txlx
Verify:
T962X-R311, U200(G12A)
Change-Id: Ia0cce8516a98653b01c3bc628e812c3a81b1df59
Signed-off-by: huaihao.guo <huaihao.guo@amlogic.com>
dts: change CMA layout [1/1]
PD#SWPL-1909
Problem:
On 32bit kernel, codec cma is located in normal zone by
"alloc-ranges" parameter with large size. This takse up
memory space for kernel and drivers. Sometimes we can see
memory allocation fail log but there are still enought
free pages. But most of these pages are CMA pages.
Solution:
Move codec cma location to high memzone. On 32bit, low
768MB memory is normal zone.
This change also correct first 1MB memory lost problem
for chips later than TXLX
Verify:
P212
Change-Id: I4592d34d08ee4dbb6700bbbfc4e0fadaceab0310
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
media: remap the addr to prevent crash on the 32bit & 2G board [1/1]
PD#SWPL-1909
Problem:
mem rw exception caused crashed.
Solution:
1. add mapping the highmem address by the func vmap().
2. remove the flag CODEC_MM_FLAGS_CPU if not necessary.
Verify:
p212, w400
Change-Id: I982c775d7c009335cae5802f0eb6287d22037db6
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
media: fix some coverity error: [1/1]
PD#SWPL-2053
Problem:
Coverity detected some code defects.
Solution:
Fixed these code defects.
Verify:
Verified u212
Change-Id: Ie3c0907c7ee3db94f65f58aa5d5c96812a7fe38c
Signed-off-by: Peng Yixin <yixin.peng@amlogic.com>
dts: modify the remaining 64bit dts to support 32bit [1/1]
PD#SWPL-2046
Problem:
some of the remaining 64bit boards do not support 32bit.
Solution:
modify the dts to support 32bit.
Verify:
axg, g12a, g12b, gxl, gxm, txl, txlx.
Change-Id: I951f8d32b18243a64e2abc5008bbb015c2e21e7a
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
dv: set avi info limited range when dv output hdr mode [1/1]
PD#SWPL-1127
Problem:
when dv output hdr mode, the quantization range is default value
instead of limited range
Solution:
set quantization range to limited range in avi info
Verify:
t962e (txlx)
Change-Id: Ib4fbeb1c7efcedd7194143772ee5c663284a1c77
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
drm: pq: G12A DRM support gamma and ctm setting [1/1]
PD#165492
Problem:
G12A DRM add gamma and ctm setting support
Solution:
1. amvecm: add gamma apis for drm
2. drm: add gamma and ctm setting
Verify:
Verified on customer platfor
Change-Id: Iff5204ae7f719542463ac15140e471cd762c9c27
Signed-off-by: Bencheng Jing <bencheng.jing@amlogic.com>
di: fix dump_status log error [1/1]
PD#174473
Problem:
dump_status log error
Solution:
fix dump_status log error
Verify:
verified in t962X_r311 board
Change-Id: I53cec7b93f6e2649bd527b8591995d54bec6450d
Signed-off-by: wenfeng.guo <wenfeng.guo@amlogic.com>
sdio: optimize sd & wifi TDMA [1/1]
PD#SWPL-1908
Problem:
sdio wifi error TDMA.
Solution:
optimize host reg value set error.
Verify:
u211 & u212
Change-Id: I0e0d52ac2bce42cf163f0065d46badc938832a65
Signed-off-by: Nan Li <nan.li@amlogic.com>
atv_demod: Output big "PO" sound when change source from ATV [1/3]
PD#SWPL-318
Problem:
Output big "PO" sound when change source from ATV to others,
or switch between small window and full screen in ATV source.
Solution:
disable atv status early when exit ATV source.
Verify:
verified by p321
Change-Id: I82af7f8eaef181f1f72ef1f9d194e33c1efdcdb5
Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com>
deinterlace: fix di coverity & di_trace [1/1]
PD#SWPL-2138
Problem:
Fix di coverity & di_trace
Solution:
fix di coverity & di
Verify:
T962X-R311
Change-Id: I9be91832eeed89ebd715276c75414e4e6da55240
Signed-off-by: wenfeng.guo <wenfeng.guo@amlogic.com>
vdin: enable vdin0 urgent for T950 default [1/1]
PD#TV-288
Problem:
There is flicker and jitter in HDMI
Solution:
enable vdin0 urgent for T950 default
Verify:
p321-T950
Change-Id: I17e472d7a47411093e9ade70a342807bed060a3a
Signed-off-by: Dezhi Kong <dezhi.kong@amlogic.com>
clk: g12a/g12b: fix 32bit set mpll clk overflow [1/1]
PD#SWPL-1933
Problem:
32bit system clk overflow
Solution:
let mpll clock not overflow
Verify:
g12b
Change-Id: Ie1c7c611e637776348bb35a3e0c1624cee57716f
Signed-off-by: shunzhou.jiang <shunzhou.jiang@amlogic.com>
reboot: reason: add reboot_reason quiescent [1/4]
PD#SWPL-1887
Problem:
Android TV need reboot_reason quiescent
Solution:
add reboot_reason quiescent
Verify:
ampere
Change-Id: Ie5a748faae05be93c56722b7806df6179116a047
Signed-off-by: Xindong Xu <xindong.xu@amlogic.com>
arm64: dts: p230: add dvb module
PD#172865
Problem:
Customer requests a new release for linux dvb
Need enable dvb for QA testing
Solution:
Need enable dvb for QA testing
Verify:
Change-Id: Ib082c457d6f43ef4d6a5d16059a7e8ce068a332e
Signed-off-by: yeping.miao <yeping.miao@amlogic.com>
emmc: clean emmc reg val for probe [1/1]
PD#SWPL-2286
Problem:
emmc response crc error in emmc init.
Solution:
clean emmc reg val from uboot to kernel.
Verify:
w400
Change-Id: I6a4763187f18e9e284c887c5c788f3a88c8548ff
Signed-off-by: Nan Li <nan.li@amlogic.com>
dts: dtsi: add dtsi for ab update [1/6]
PD#SWPL-1513
Problem:
ab update can not work on P
Solution:
add dtsi for ab update for P
Verify:
test pass in ampere
Change-Id: I6ff219170a16c0081fba7297110e8dfaadcff401
Signed-off-by: Xindong Xu <xindong.xu@amlogic.com>
backlight: ldim: update usage for new spi driver [1/1]
PD#SWPL-1603
Problem:
don't support old spi driver now
Solution:
update new spi driver api for ldim driver
Verify:
txlx r311
Change-Id: I4dff13e4cab5666a30488730e27c6275b933a922
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
hdmitx: fix wrong power uevent in suspend/resume [1/1]
PD#SWPL-1931
Problem:
Wrong hdmi_power uevent in suspend/resume flow
Solution:
Remove current mode judgement
Verify:
R321
Change-Id: Ib36e8d6cb4988f7b2be91146b12e974c194526ca
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
ddr: fix dmc_monitor register bit missmatch on G12A/G12B [1/1]
PD#172256
Problem:
On G12A/G12B, bit for ddr0 protection 0 vilation is not same
Solution:
Fix bit missmatch problem
Verify:
U212
Change-Id: I5dca14c8bc96bdcb85dbe8bbd5a9a18157834bcf
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
deinterlace: fix kasan bug in di_task_handle [1/1]
PD#173820
Problem:
kasan bug in di_task_handle
Solution:
delete cma_mutex because it has no effect
Verify:
P321
Change-Id: Ie6ca7973576e7350870c5466b54b38b490dff65e
Signed-off-by: Jihong Sui <jihong.sui@amlogic.com>
mmc: fixup toshiba emmc secure discard [1/1]
PD#SWPL-1937
Problem:
secure discard and trim may cost 5mins on toshiba
"004GA0" emmc.
Solution:
using normal discard operation instead.
Verify:
W400
Change-Id: I7fd23867f8ae0af9192377de8f8c15609ef809ed
Signed-off-by: Yonghui Yu <yonghui.yu@amlogic.com>
ionvideo: Add thread stop protection when the process exits [1/1]
PD#SWPL-1689
Problem:
The thread still exists when the process exits
Solution:
Check if the thread exits when the process exits
Verify:
local
Change-Id: I6d805da6bfa0df85b6d39d065c95eed49b3cb7da
Signed-off-by: Renjiang Han <renjiang.han@amlogic.com>
dts: the memory of the reserved 1MB was recovered. [1/1]
PD#SWPL-2250
Problem:
h265 & vp9 4K playback abnormal because HEVC FRONT &
VPU READ2 pollute the phy address of near 0x0.
Solution:
it's only a temporary modification, the real reason
needs to take more time fixed.
Verify:
U212
Change-Id: I3dcc47f8979960e95c89cf53fe10eadc1a7ae5ad
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
dts: r314: increase board dts file [2/3]
PD#SWPL-1978
Problem:
bring up board r314.
Solution:
increase board dts file
feature:
The Dts file with Einstein is identical except for the memory difference
R314 1.5G, Einstein 1G.
Verify:
Verified on txlx_t962x_r314 board.
Change-Id: I271cb92a9cbcab174fb0bb76cfbc8beff9c730d3
Signed-off-by: guotai.shen <guotai.shen@amlogic.com>
vrtc: vrtc time update for stick mem [2/2]
PD#174867
Problem:
vrtc time update for stick mem.
Solution:
vrtc time update for stick mem.
Verify:
g12a txl.
Change-Id: I65074422569370b253882a4e66fb771da10ccdbf
Signed-off-by: Hong Guo <hong.guo@amlogic.com>
irblaster: add synchronous transmit [1/1]
PD#SWPL-2170
Problem:
cts fail
android.hardware.consumerir.cts.ConsumerIrTest#test_timing
Solution:
add synchronous transmit
when it returns the pattern has been transmitted
Verify:
verify on franklin
Change-Id: I7e490236e5effc13f81e4fab3a1c7fc10c5dcfb3
Signed-off-by: Wencai You <wencai.you@amlogic.com>
amvecm: pq: add dnlp/bri/contrast for tl1 [1/1]
PD#172587
Problem:
dnlp/bri/contrast can not finetune
Solution:
1. add dnlp regisiter 10bit write
2. add brightness/contrast/sr for tl1
Verify:
verify on tl1
Change-Id: I90b363f57fbf2cd485e2acaddb5ce60a8715a47b
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
ramdump: add ramdump support for ARM32 [1/1]
PD#165764
Problem:
ramdump is not enabled on ARM32
Solution:
Enable it and fix compatible with ARM64
Verify:
R311
Change-Id: I337ef019ecab05d25b7d9dc0cacab7a03149aa7b
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
avsync: tsync: fix the avsync problem [1/1]
PD#SWPL-1158
Problem:
The video will become un-smooth after change
between several audio languages
Solution:
when apts and vpts distance is large then 60*90000
then set the tsync mode to vmaster and set the tysnc
enable to 0
Verify:
P321
Change-Id: I7c7e32f49ca12645cf6a1aa58caeb968a9234c8c
Signed-off-by: live.li <live.li@amlogic.com>
dtv_demod: fix dtv demod crash switch once every 2s [1/1]
PD#SWPL-2293
Problem:
dtv demod switch once every 2s cause crash
Solution:
fix dtv demod crash switch once every 2s
Verify:
P321
Change-Id: Ibfd919a503e3dae41f51637cdff6e01ed27bac71
Signed-off-by: wenfeng.guo <wenfeng.guo@amlogic.com>
ramdump: fix a warning of cast [1/1]
PD#165764
Problem:
have a warning of the "ram" variable cast.
Solution:
add a "void *" to the ram variable.
Verify:
test locally.
Change-Id: I037a89a28e98ae8c8ab965e908877bf4a72d3ead
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
vpp: fix p212 osd sdr2hdr luma too dark [1/1]
PD#SWPL-1707
Problem:
{NTS}Manual,HDR-001-TC2 Failed step 8/12/16/20 VID
and GFX doesn't match
Solution:
add osd hdr2sdr convert table for 400nit on p212
Verify:
verify on p212
Change-Id: I9a9dcf13ab0deaa8b676c5c409919367420b6d1b
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
vpp: fix p212 osd sdr2hdr luma too dark compile [1/1]
PD#SWPL-1707
Problem:
{NTS}Manual,HDR-001-TC2 Failed step 8/12/16/20 VID
and GFX doesn't match
Solution:
add osd hdr2sdr convert table for 400nit on p212
Verify:
verify on p212
Change-Id: I16637652b6d07ba81b25f46c2684ff7ae3e1c68c
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
vdac: fix atv/dtv resume no signal issue [1/1]
PD#SWPL-2351
Problem:
atv/dtv no signal when suspend after switch from hdmi
Solution:
disable vdac bandgap when suspend
Verify:
test pass on P321
Change-Id: I0449f0d9673f46928b8951249f5759a4a1f16562
Signed-off-by: Nian Jing <nian.jing@amlogic.com>
spicc: fix miss bytes/time consume/loopback issue [1/1]
PD#SWPL-215
Problem:
1. additional or missing bytes be sent on mosi sometimes.
2. Customer DSP load firmware through SPICC. the loading time is 12s with
buildroot release 20180907 while 6s with 20180131.
3. rx error when work in loopback mode at high speed.
Solution:
1. change to disable irq at the irq-handle begining;
change to enable irq after all data pulled.
2. pre-setup of every spi transfer spends most of time on clk_set_rate().
this time is not obvious while cpu work at a high frequence such as
1000MHz.
In fact, a slave speed is almost fixed and we needn't set it for every
transfer but set only when speed changed.
3. disable auto io delay when in loopback mode.
Verify:
verified on axg-s400-v03 and tl1-skt
Change-Id: I61bcceccc243b218879b2b0711d0aff7538151f6
Signed-off-by: Sunny Luo <sunny.luo@amlogic.com>
framebuffer: remove lock in fbmem ioctl. [1/1]
PD#SWPL-2497
Problem:
on 32bit kernel, when user space call WAITFORVSYNC,
we cannot send other command to framebuffer.
Solution:
remove lock in fbioctl as fb compat_ioctl do.
Our driver have locks to protect internal resources.
Verify:
verified on franklin.
Change-Id: I1789c09a7ea459aed4b782748847687c7f974526
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
dv: fix osd color matrix when dv enabled [1/1]
PD#SWPL-1804
Problem:
fix osd color matrix when dv enabled
Solution:
enable osd matrix when dv enabled
Verify:
verified on u212 dev board
Change-Id: I4b4206f4d8c447873f23a3a0066af0d0fa85e18c
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
osd: fix osd color error when osd hdr enabled [1/1]
PD#SWPL-1804
Problem:
osd color error when osd hdr enabled
Solution:
div alpha when osd hdr enabled
Verify:
verified on u212 dev board
Change-Id: I1f5ea0db404d1956f454bd5da2532bded89626ee
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
osd: add meson-hwc repodution rate support [1/2]
PD#SWPL-1810
Problem:
osd blend repodution rate not supported
Solution:
add meson-hwc repodution rate support
Verify:
verified on u212 dev board
Change-Id: Ied10393e240e8d3f2a1eefe17a2d4d3c965e3383
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
dts: GXL: defendkey: support secure upgrade check [3/3]
PD#SWPL-2100
Problem:
GXL need to support secure upgrade check
Solution:
add defendkey support in all GXL dts
Verify:
GXL skt board verify pass
Change-Id: I73b84b86f23f512ddc556b61f3a07d015c717e8c
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
GXL: defendkey: support secure upgrade check [2/3]
PD#SWPL-2100
Problem:
GXL need to support secure upgrade check
Solution:
1.add dtb decrypt support
2.add 32bit defendkey support
Verify:
GXL skt board verify pass
Change-Id: I501967530b2a61d9b90c20241b82f92b00829453
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
PM_SUSPEND: suspend: set wakeup reason to stick mem [2/2]
PD#SWPL-2287
Problem:
set wakeup reason to stick mem
Solution:
set wakeup reason to stick mem
Verify:
gxl,test pass
Change-Id: Iff6f1824ea4ac9ec12b8295d4283992fdac03b14
Signed-off-by: Hong Guo <hong.guo@amlogic.com>
deinterlace: fix CMA layout adaptation di_cma_reserved [1/1]
PD#SWPL-2308
Problem:
CMA layout adaptation di_cma_reserved cause crash
Solution:
fix CMA layout adaptation di_cma_reserved
Verify:
P321
Change-Id: Ia24e811a02d47d70afa1c06361fd0852c357394c
Signed-off-by: wenfeng.guo <wenfeng.guo@amlogic.com>
media: media_sync: remove serial port debug logs [1/1]
PD#SWPL-2494
Problem:
serial port always print timestamp log
Solution:
remove this log output
Verify:
P321
Change-Id: I875dffab483269374b9939f556a3ed01e9d2b215
Signed-off-by: xiaoming.sui <xiaoming.sui@amlogic.com>
dts: GXL/TXL: adjust defendkey_reserved location [1/1]
PD#SWPL-2539
Problem:
defendkey_reserved have influenced on display
during the system boot up
Solution:
adjust defendkey_reserved node location
Verify:
GXL/TXL skt board verify pass
Change-Id: I5cf0245f9653e20224c4e3bab1f6f05527011210
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
DTS: p230 board config spdif out pin [3/3]
PD# SWPL-1223
Problem:
p230 board spdif out no audio sound
Solution:
config p230 board spdif out pin
Verify:
verify on p230
Change-Id: I0c35e6d526aab98955206f78322fcc873ff01b8c
Signed-off-by: Chaomin Zheng <chaomin.zheng@amlogic.com>
hdmitx: disable encryption hdcp1.4 fail [1/1]
PD#SWPL-2302
Problem:
Play online video,half an hour later, the TV shows
a pink screen and the sound is normal.
Solution:
Check hdcp1.4 status and disable encryption when fail.
Verify:
S922X
Change-Id: I6ce51824726b267d140750f9e9d80aa8bb921cac
Signed-off-by: Kaifu Hu <kaifu.hu@amlogic.com>
dv: fix the flickered problem [1/1]
PD#SWPL-1207
Problem:
fix the filckered problem when playing transition
video in sdr tv
Solution:
when dv core2 don't run, the reset can't be executed
Verify:
r321
Change-Id: I719325f1722589e02a40d46442258b0d1e3feb17
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
display: Modify the value of register afbc_gclk_ctrl [1/1]
PD#SWPL-2565
Problem:
The splash screen problem for first frame after starting up.
Solution:
Modify the value of register afbc_gclk_ctrl.
Verify:
u211
Change-Id: Ie1ddfade0566ff8cd571c43ba8490a8c277a8758
Signed-off-by: Peng Yixin <yixin.peng@amlogic.com>
dts: fix SD card read issue for g12a board [2/3]
PD#SWPL-2588
Problem:
g12a u220/u221 boad cann't read SD card
Solution:
auto detect revA and revB chip
Verify:
test pass on u220/u221 board
Change-Id: Ia153bb3f966fe2a58d5f3ed1bf1d3f610d00ff5a
Signed-off-by: Sandy Luo <sandy.luo@amlogic.com>
frame_sync: fix last few seconds of the DD+ stream are not smooth [4/4]
PD#SWPL-714
Problem:
The last few seconds of the DD+ stream are not smooth by DTV playback.
Solution:
check the apts and vpts both discontinue,then replay the stream.
Verify:
P321
Change-Id: I6cbbdecc052dfe9fef76b44f36545b231332ee12
Signed-off-by: Yinming Ding <yinming.ding@amlogic.com>
dvb: tuner: USB DVB tuner support from Google
PD#SWPL-932
Problem:
USB DVB Dongle search program Function is not implemented
Solution:
merge usb dvb tuner from kernel3.14
http://10.8.9.5/kernel/common/commit/?h=amlogic-3.14-dev&id=7af9c5c38efdc96f5d3235f17788232a630b3d32
fixed frontend.h structures
Verify:
Ampere P
Change-Id: I16d68c2507711eda43313619d210d05699abcc25
Signed-off-by: qi duan <qi.duan@amlogic.com>
scripts/dtc: Update to upstream version v1.4.5-3-gb1a60033c110
PD#SWPL-2193
This adds the following commits from upstream:
b1a60033c110 tests: Add a test for overlays syntactic sugar
737b2df39cc8 overlay: Add syntactic sugar version of overlays
497432fd2131 checks: Use proper format modifier for size_t
22a65c5331c2 dtc: Bump version to v1.4.5
c575d8059fff Add fdtoverlay to .gitignore
b6a6f9490d19 fdtoverlay: Sanity check blob size
8c1eb1526d2d pylibfdt: Use Python2 explicitly
ee3d26f6960b checks: add interrupts property check
c1e7738988f5 checks: add gpio binding properties check
b3bbac02d5e3 checks: add phandle with arg property checks
fe50bd1ecc1d fdtget: Split out cell list display into a new function
62d812308d11 README: Add a note about test_tree1.dts
5bed86aee9e8 pylibfdt: Add support for fdt_subnode_offset()
46f31b65b3b3 pylibfdt: Add support for fdt_node_offset_by_phandle()
a3ae43723687 pylibfdt: Add support for fdt_parent_offset()
a198af80344c pylibfdt: Add support for fdt_get_phandle()
b9eba92ea50f tests: Return a failure code when any tests fail
155faf6cc209 pylibfdt: Use local pylibfdt module
50e5cd07f325 pylibfdt: Add a test for use of uint32_t
ab78860f09f5 pylibfdt: Add stdint include to fix uint32_t
36f511fb1113 tests: Add stacked overlay tests on fdtoverlay
1bb00655d3e5 fdt: Allow stacked overlays phandle references
a33c2247ac8d Introduce fdt_setprop_placeholder() method
0016f8c2aa32 dtc: change default phandles to ePAPR style instead of both
e3b9a9588a35 tests: fdtoverlay unit test
42409146f2db fdtoverlay: A tool that applies overlays
aae22722fc8d manual: Document missing options
13ce6e1c2fc4 dtc: fix sprintf() format string error, again
d990b8013889 Makefile: Fix build on MSYS2 and Cygwin
51f56dedf8ea Clean up shared library compile/link options
21a2bc896e3d Suppress expected error message in fdtdump test
2a42b14d0d03 dtc: check.c fix compile error
a10cb3c818d3 Fix get_node_by_path string equality check
548aea2c436a fdtdump: Discourage use of fdtdump
c2258841a785 fdtdump: Fix over-zealous version check
9067ee4be0e6 Fix a few whitespace and style nits
e56f2b07be38 pylibfdt: Use setup.py to build the swig file
896f1c133265 pylibfdt: Use Makefile constructs to implement NO_PYTHON
90db6d9989ca pylibfdt: Allow setup.py to operate stand-alone
e20d9658cd8f Add Coverity Scan support
b04a2cf08862 pylibfdt: Fix code style in setup.py
1c5170d3a466 pylibfdt: Rename libfdt.swig to libfdt.i
580a9f6c2880 Add a libfdt function to write a property placeholder
ab15256d8d02 pylibfdt: Use the call function to simplify the Makefile
9f2e3a3a1f19 pylibfdt: Use the correct libfdt version in the module
e91c652af215 pylibfdt: Enable installation of Python module
8a892fd85d94 pylibfdt: Allow building to be disabled
741cdff85d3e .travis.yml: Add builds with and without Python library prerequisites
14c4171f4f9a pylibfdt: Use package_dir to set the package directory
89a5062ab231 pylibfdt: Use environment to pass C flags and files
4e0e0d049757 pylibfdt: Allow pkg-config to be supplied in the environment
6afd7d9688f5 Correct typo: s/pylibgfdt/pylibfdt/
Change-Id: Ib23ca90f1e34ed85db3aae9d223fc921e6f2114e
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: jiamin ma <jiamin.ma@amlogic.com>
deinterface: temporary remove afbc support on chip after g12a [1/1]
PD#172587
Problem:
afbc now support on chip txlx and g12a,
but we use cpu_after_eq(G12A) to judge
if afbc is support or not.
Solution:
use is_meson_g12a_cpu() replace cpu_after_eq(G12A).
Verify:
u200(g12a),t962x_r311
Change-Id: I3fb7ed54f7137fada97481177e40a2543cf8df9f
Signed-off-by: huaihao guo <huaihao.guo@amlogic.com>
osd: osd: a black line below the screen when play video [1/1]
PD#SWPL-2558
Problem:
a black horizontal line below the screen when play video
Solution:
set dummy data alpha
Verify:
verified on p321
Change-Id: I36dc924b9bc89f4f8ea61eb86c6f5bc9199e40ae
Signed-off-by: Jian Cao <jian.cao@amlogic.com>
osd: call osd_update_enable if needed [2/2]
PD#SWPL-2580
Problem:
The vd1 postblend bit in vpp misc is overwriten.
Solution:
osd: call osd_update_enable if needed
Verify:
verify by p212
Change-Id: I438848dc843ea370d65dc3c6b4befb32d1f2b5f8
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
video: vpp: add more check for vd1 postblend in vpp_misc [1/2]
PD#SWPL-2580
Problem:
The vd1 postblend bit in vpp misc is overwriten.
Solution:
add more check for vd1 postblend bit in vsync
Verify:
verify by p212
Change-Id: I367e920c13764af41446d7d759aaf8388fb08722
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
osd: adjust free_scale_dst if interlaced [1/1]
PD#SWPL-2600
Problem:
osd display half menu when in 1080i
Solution:
osd: adjust free_scale_dst if interlaced
Verify:
verify by p212
Change-Id: Ie9faffbb5942faa317b380ca0cdbee072a30c01f
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
Signed-off-by: Jian Cao <jian.cao@amlogic.com>
osd: call osd_update_enable directly if osd cursor [1/1]
PD#SWPL-2744
Problem:
osd mouse can't see before platform g12a
Solution:
call osd_update_enable directly if osd cursor
Verify:
verify by p212
Change-Id: Ia8488a5bd0f707de2a02aa2b74e4a07582ef7f60
Signed-off-by: pengcheng chen <pengcheng.chen@amlogic.com>
deinterlace: there's a change to hang in shutdown [1/1]
PD#OTT-778
Problem:
1.kernel received reboot command, but hang in shutdown
Solution:
move tasklet_disable behind tasklet_kill
Verify:
P321
Change-Id: I994c21c5aeb42dbbb4e3b50ccac8376db4d4039d
Signed-off-by: Jihong Sui <jihong.sui@amlogic.com>
Conflicts:
drivers/amlogic/media/deinterlace/deinterlace.c
fdto: peripheral driver support dtbo [1/4]
PD#SWPL-2193
Problem:
We cannot compile dtbo file in a way like
make ARCH=arm foo.dtbo
Solution:
Add dtbo target and reuse the dtb rules
Verify:
Locally verifed
Change-Id: I34b88e61b409defe553cf5ed1077c160b9a6f350
Signed-off-by: jiamin ma <jiamin.ma@amlogic.com>
fdto: peripheral driver support dtbo [2/4]
PD#SWPL-2193
Problem:
Node dummy-battery and dummy-charger are android version
dependent, and should be "fdt overlayed"
Solution:
Add fdt overlay file android_p_overlay_dt.dts, and configure
dummy-battery and dummy-charger in it
Verify:
Locally on P212 and R311
Change-Id: I75a27a0ad1253556cae5273f36e0eb828291fd73
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
Signed-off-by: jiamin ma <jiamin.ma@amlogic.com>
dts: fdto: the fdt is damaged in booting process [1/1]
PD#SWPL-2960
Problem:
If the overlay dts refers to a node which does not exist in the base
device tree, the final fdt will be in a damaged state after doing fdt
overlay
Solution:
Use the standard overlay dts syntax to note node to be applied instead
of the google recommended one
Verify:
Locally on P230
Change-Id: Ibf45abff1348437bc3fc2745e6d955cc8292db46
Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com>
dv: fix the error hsize and vsize of dv core 2 [1/1]
PD#OTT-776
Problem:
dv core 2 get the error hsize and vsize from OSD module
Solution:
when opening 3 osd paths. we get the accurate the parameters again
Verify:
g12b s922
Change-Id: Ib6af084910c8cf1896060295c7cea75971cd86b6
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
codec_mm: show current tvp flag [1/2]
PD#OTT-937
Problem:
When kill mediacodec during playing drm, tvp memory not be freed
Solution:
Check tvp flag in kernel, free tvp memory if flag is not zero
Verify:
P212
Change-Id: I35cbb2808bf179710ec59bd5227995949946ee77
Signed-off-by: Tao Guo <tao.guo@amlogic.com>
Revert "arm: dts: gxl: replace watchdog with new version [1/1]"
This reverts commit d932cb0022.
Change-Id: Ie9af906663de4f49a5f21fe026d68e55238b3642
defendkey: G12A: support secure upgrade check [3/4]
PD#SWPL-2347
Problem:
G12A need to support secure upgrade check
Solution:
1.modify reserve memory mode
2.only use cpu0 in defendkey write process
Verify:
G12A skt board verify pass
Change-Id: Iad771381e959a79dcba7a0adb862295fa5ba5dee
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
dts: G12A: defendkey: support secure upgrade check [4/4]
PD#SWPL-2347
Problem:
G12A need to support secure upgrade check
Solution:
add defendkey support in all G12A dts
Verify:
G12A skt board verify pass
Change-Id: I7343c2c553dfe776eefcfff9ab3a270c87014118
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
emmc: modify device node name [1/1]
PD#SWPL-2719
Problem:
Can't idetify correctlly when move disk have multi
partition
Solution:
Remove the function of using the partition name as
the device node name
Verify:
test pass on tl1 ref board
Change-Id: I113e63f209c529149fb94b0bb10b0b254717c2bf
Signed-off-by: Ruixuan Li <ruixuan.li@amlogic.com>
arm: dts: gxl: replace watchdog with new version [1/1]
PD#SWPL-1600
Problem:
The current watchdog is not the upstream standard driver.
Solution:
Replace watchdog device tree to use the standard watchdog driver.
Verify:
S905X P215/P212 Android-32bit Kernel-32bit.
Change-Id: I972c8ebb0158caa2df393d9a2d687d9298181102
Signed-off-by: Bo Yang <bo.yang@amlogic.com>
watchdog: Change the feed watchdog mode [1/1]
PD#SWPL-2378
Problem:
Using Android to feed watchdog is not easy to debug the
crash problem, in addition, the Android does not open
the dog feeding service.
Solution:
Change the DTS configuration and change into driver feed watchdog.
Verify:
test pass on g12a-u200
Change-Id: Ie0a91fd8451fdccf3038ba6290a277c9048c52cd
Signed-off-by: Yingyuan Zhu <yingyuan.zhu@amlogic.com>
hdmitx: add parsing for non-standard dv edid [1/1]
PD#SWPL-1968
Problem:
some error edid need to be parsed
Solution:
add parsing for non-standard dv edid
Verify:
t962e (txlx)
Change-Id: Ia952ed59650fd0b1c665ca03175a38c1cf3ae5bb
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
media: video: fix cvbs black dot issue [1/1]
PD#SWPL-1910
Problem:
Black dot on cvbsout when local play 4k
Solution:
Set limit rang when vpp output yuv for cvbs
Verify:
test pass on u200/s905x
Change-Id: Ia3e93078eb1a00d8c974bae8cc373a3a679f8b9f
Signed-off-by: Nian Jing <nian.jing@amlogic.com>
hdmitx: fix a resmue panic isse [1/1]
PD#SWPL-3188
Problem:
When HDMI cable is not connected(that is CVBS out), and system
may be panic after resume.
Solution:
Add NULL pointer protection check
Verify:
P212
Change-Id: I60818faaf7049667501e31990fa557b754ea7f1b
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
dts: G12B: defendkey: support secure upgrade check [3/3]
PD#SWPL-2348
Problem:
G12B need to support secure upgrade check
Solution:
add defendkey support in all G12B dts
Verify:
G12B skt board verify pass
Change-Id: I2c3f51e761531cf2169dda483991b3d876039686
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
dts: GXL: defendkey: modify reserve memory mode [2/2]
PD#SWPL-2100
Problem:
GXL need to modify reserve memory mode
Solution:
remove defendkey reserved memory in GXL dts
Verify:
GXL skt board verify pass
Change-Id: Ie5ed01c6dd8f88c5594d6bc73f13282e535071a2
Signed-off-by: Zhongfu Luo <zhongfu.luo@amlogic.com>
cec: for support multi-logical address [1/1]
PD#SWPL-418
Problem:
cec: support multi-logical address
Solution:
if working on multi-logical address, enable two ip
1.enable cec_a, cec_b
2.enable two interrupt
3.enable two pinmux oa_7, ao_8
4.cec_a only send all msg
5.cec_b only receive all msg
6.discard ceca broadcast msg
Verify:
r311
r321
p321
Change-Id: I1dc93429876ede951657102bcd7d41a500946719
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
cec: add device match for tl1 [1/1]
PD#172587
Problem:
add device match for tl1
Solution:
1.add cecb ver
2.add ceca status reg flag
Verify:
1.run TPM
2.run chip
Change-Id: I304abfd1d6f1f216f9f0c56ca19a8bdb2ccfeabb
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
cec: update cec platform data for tl1 [1/1]
PD#172587
Problem:
the line check always return low.
the print of check too much.
Solution:
modify the cec gpio from 7 to 10.
modify the check line.
Verify:
tl1 x301 board
Change-Id: I09ae5eee455d83e51e65957cd293e07da1c0046a
Signed-off-by: hongmin hua <hongmin.hua@amlogic.com>
cec: auto shutdown case crash [1/1]
PD#SWPL-2258
Problem:
1.running autoshutdown cause device crashed
Solution:
1.optimize cec power on proble flow
Verify:
1.p212
Change-Id: Ia78c76cdb4b72a1a275628b3ad77ea6d71696d89
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
commit 400e22499d upstream.
Commit 63f53dea0c ("mm: warn about allocations which stall for too
long") was a great step for reducing possibility of silent hang up
problem caused by memory allocation stalls. But this commit reverts it,
for it is possible to trigger OOM lockup and/or soft lockups when many
threads concurrently called warn_alloc() (in order to warn about memory
allocation stalls) due to current implementation of printk(), and it is
difficult to obtain useful information due to limitation of synchronous
warning approach.
Current printk() implementation flushes all pending logs using the
context of a thread which called console_unlock(). printk() should be
able to flush all pending logs eventually unless somebody continues
appending to printk() buffer.
Since warn_alloc() started appending to printk() buffer while waiting
for oom_kill_process() to make forward progress when oom_kill_process()
is processing pending logs, it became possible for warn_alloc() to force
oom_kill_process() loop inside printk(). As a result, warn_alloc()
significantly increased possibility of preventing oom_kill_process()
from making forward progress.
---------- Pseudo code start ----------
Before warn_alloc() was introduced:
retry:
if (mutex_trylock(&oom_lock)) {
while (atomic_read(&printk_pending_logs) > 0) {
atomic_dec(&printk_pending_logs);
print_one_log();
}
// Send SIGKILL here.
mutex_unlock(&oom_lock)
}
goto retry;
After warn_alloc() was introduced:
retry:
if (mutex_trylock(&oom_lock)) {
while (atomic_read(&printk_pending_logs) > 0) {
atomic_dec(&printk_pending_logs);
print_one_log();
}
// Send SIGKILL here.
mutex_unlock(&oom_lock)
} else if (waited_for_10seconds()) {
atomic_inc(&printk_pending_logs);
}
goto retry;
---------- Pseudo code end ----------
Although waited_for_10seconds() becomes true once per 10 seconds,
unbounded number of threads can call waited_for_10seconds() at the same
time. Also, since threads doing waited_for_10seconds() keep doing
almost busy loop, the thread doing print_one_log() can use little CPU
resource. Therefore, this situation can be simplified like
---------- Pseudo code start ----------
retry:
if (mutex_trylock(&oom_lock)) {
while (atomic_read(&printk_pending_logs) > 0) {
atomic_dec(&printk_pending_logs);
print_one_log();
}
// Send SIGKILL here.
mutex_unlock(&oom_lock)
} else {
atomic_inc(&printk_pending_logs);
}
goto retry;
---------- Pseudo code end ----------
when printk() is called faster than print_one_log() can process a log.
One of possible mitigation would be to introduce a new lock in order to
make sure that no other series of printk() (either oom_kill_process() or
warn_alloc()) can append to printk() buffer when one series of printk()
(either oom_kill_process() or warn_alloc()) is already in progress.
Such serialization will also help obtaining kernel messages in readable
form.
---------- Pseudo code start ----------
retry:
if (mutex_trylock(&oom_lock)) {
mutex_lock(&oom_printk_lock);
while (atomic_read(&printk_pending_logs) > 0) {
atomic_dec(&printk_pending_logs);
print_one_log();
}
// Send SIGKILL here.
mutex_unlock(&oom_printk_lock);
mutex_unlock(&oom_lock)
} else {
if (mutex_trylock(&oom_printk_lock)) {
atomic_inc(&printk_pending_logs);
mutex_unlock(&oom_printk_lock);
}
}
goto retry;
---------- Pseudo code end ----------
But this commit does not go that direction, for we don't want to
introduce a new lock dependency, and we unlikely be able to obtain
useful information even if we serialized oom_kill_process() and
warn_alloc().
Synchronous approach is prone to unexpected results (e.g. too late [1],
too frequent [2], overlooked [3]). As far as I know, warn_alloc() never
helped with providing information other than "something is going wrong".
I want to consider asynchronous approach which can obtain information
during stalls with possibly relevant threads (e.g. the owner of
oom_lock and kswapd-like threads) and serve as a trigger for actions
(e.g. turn on/off tracepoints, ask libvirt daemon to take a memory dump
of stalling KVM guest for diagnostic purpose).
This commit temporarily loses ability to report e.g. OOM lockup due to
unable to invoke the OOM killer due to !__GFP_FS allocation request.
But asynchronous approach will be able to detect such situation and emit
warning. Thus, let's remove warn_alloc().
[1] https://bugzilla.kernel.org/show_bug.cgi?id=192981
[2] http://lkml.kernel.org/r/CAM_iQpWuPVGc2ky8M-9yukECtS+zKjiDasNymX7rMcBjBFyM_A@mail.gmail.com
[3] commit db73ee0d46 ("mm, vmscan: do not loop on too_many_isolated for ever"))
Link: http://lkml.kernel.org/r/1509017339-4802-1-git-send-email-penguin-kernel@I-love.SAKURA.ne.jp
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reported-by: Cong Wang <xiyou.wangcong@gmail.com>
Reported-by: yuwang.yuwang <yuwang.yuwang@alibaba-inc.com>
Reported-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[Resolved backport conflict due to missing 8225196, a8e9925, 9e80c71 and
9a67f64 in 4.9 -- all of which modified this hunk being removed.]
Signed-off-by: Amit Shah <amit@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
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
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>
commit 7810e6781e upstream.
In __alloc_pages_slowpath() we reset zonelist and preferred_zoneref for
allocations that can ignore memory policies. The zonelist is obtained
from current CPU's node. This is a problem for __GFP_THISNODE
allocations that want to allocate on a different node, e.g. because the
allocating thread has been migrated to a different CPU.
This has been observed to break SLAB in our 4.4-based kernel, because
there it relies on __GFP_THISNODE working as intended. If a slab page
is put on wrong node's list, then further list manipulations may corrupt
the list because page_to_nid() is used to determine which node's
list_lock should be locked and thus we may take a wrong lock and race.
Current SLAB implementation seems to be immune by luck thanks to commit
511e3a0588 ("mm/slab: make cache_grow() handle the page allocated on
arbitrary node") but there may be others assuming that __GFP_THISNODE
works as promised.
We can fix it by simply removing the zonelist reset completely. There
is actually no reason to reset it, because memory policies and cpusets
don't affect the zonelist choice in the first place. This was different
when commit 183f6371aa ("mm: ignore mempolicies when using
ALLOC_NO_WATERMARK") introduced the code, as mempolicies provided their
own restricted zonelists.
We might consider this for 4.17 although I don't know if there's
anything currently broken.
SLAB is currently not affected, but in kernels older than 4.7 that don't
yet have 511e3a0588 ("mm/slab: make cache_grow() handle the page
allocated on arbitrary node") it is. That's at least 4.4 LTS. Older
ones I'll have to check.
So stable backports should be more important, but will have to be
reviewed carefully, as the code went through many changes. BTW I think
that also the ac->preferred_zoneref reset is currently useless if we
don't also reset ac->nodemask from a mempolicy to NULL first (which we
probably should for the OOM victims etc?), but I would leave that for a
separate patch.
Link: http://lkml.kernel.org/r/20180525130853.13915-1-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Fixes: 183f6371aa ("mm: ignore mempolicies when using ALLOC_NO_WATERMARK")
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
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>
Signed-off-by: Victor Wan <victor.wan@amlogic.com>
Conflicts:
arch/arm/configs/bcm2835_defconfig
arch/arm/configs/sunxi_defconfig
include/linux/cpufreq.h
init/main.c
PD#159608: mm: optimize for CMA allocate time
1. Make all amlogic-changed mm code configuarable, which are
wrapped by CONFIG_AMLOGIC_CMA/CONFIG_AMLOGIC_MEMORY_EXTEND
2. Implement some core code of CMA to a single file:
drivers/amlogic/memory_ext/aml_cma.c
3. detailed imporove steps:
a) use NOOP as default IO-scheduler for nand based storage.
which can avoid long time wait for page lock found in
CFQ scheduler;
b) use per-cpu thread to allocate CMA concurrent when driver
request large amount CMA memory; these threads have high
user nice value to reduce schedule delay;
c) increase task user nice of mmc queue and kswapd.
d) wake up kswapd if page are hold by kswap shrink list and
cma isolated test failed.
e) Fobidden low user nice task use CMA, which can avoid priority
inversion problem.
f) optimize for LRU usage, devide each type of LRU to 2 parts,
normal pages are linked after LRU head, CMA pages are linked
after cma_list.
g) avoid compaction case move cma forbidden pages to cma area.
h) Increase strength of lowmemory killer.
4. Improve read speed of /proc/pagetrace, a filter can be set to
reduce message which not print functions allocate memory less
than filter value:
echo filter=xxx > /proc/pagetrace
Change-Id: Ie79288b7947aa642e4f7eacc25565559a73660df
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
commit c73322d098 upstream.
Patch series "mm: kswapd spinning on unreclaimable nodes - fixes and
cleanups".
Jia reported a scenario in which the kswapd of a node indefinitely spins
at 100% CPU usage. We have seen similar cases at Facebook.
The kernel's current method of judging its ability to reclaim a node (or
whether to back off and sleep) is based on the amount of scanned pages
in proportion to the amount of reclaimable pages. In Jia's and our
scenarios, there are no reclaimable pages in the node, however, and the
condition for backing off is never met. Kswapd busyloops in an attempt
to restore the watermarks while having nothing to work with.
This series reworks the definition of an unreclaimable node based not on
scanning but on whether kswapd is able to actually reclaim pages in
MAX_RECLAIM_RETRIES (16) consecutive runs. This is the same criteria
the page allocator uses for giving up on direct reclaim and invoking the
OOM killer. If it cannot free any pages, kswapd will go to sleep and
leave further attempts to direct reclaim invocations, which will either
make progress and re-enable kswapd, or invoke the OOM killer.
Patch #1 fixes the immediate problem Jia reported, the remainder are
smaller fixlets, cleanups, and overall phasing out of the old method.
Patch #6 is the odd one out. It's a nice cleanup to get_scan_count(),
and directly related to #5, but in itself not relevant to the series.
If the whole series is too ambitious for 4.11, I would consider the
first three patches fixes, the rest cleanups.
This patch (of 9):
Jia He reports a problem with kswapd spinning at 100% CPU when
requesting more hugepages than memory available in the system:
$ echo 4000 >/proc/sys/vm/nr_hugepages
top - 13:42:59 up 3:37, 1 user, load average: 1.09, 1.03, 1.01
Tasks: 1 total, 1 running, 0 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.0 us, 12.5 sy, 0.0 ni, 85.5 id, 2.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem: 31371520 total, 30915136 used, 456384 free, 320 buffers
KiB Swap: 6284224 total, 115712 used, 6168512 free. 48192 cached Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
76 root 20 0 0 0 0 R 100.0 0.000 217:17.29 kswapd3
At that time, there are no reclaimable pages left in the node, but as
kswapd fails to restore the high watermarks it refuses to go to sleep.
Kswapd needs to back away from nodes that fail to balance. Up until
commit 1d82de618d ("mm, vmscan: make kswapd reclaim in terms of
nodes") kswapd had such a mechanism. It considered zones whose
theoretically reclaimable pages it had reclaimed six times over as
unreclaimable and backed away from them. This guard was erroneously
removed as the patch changed the definition of a balanced node.
However, simply restoring this code wouldn't help in the case reported
here: there *are* no reclaimable pages that could be scanned until the
threshold is met. Kswapd would stay awake anyway.
Introduce a new and much simpler way of backing off. If kswapd runs
through MAX_RECLAIM_RETRIES (16) cycles without reclaiming a single
page, make it back off from the node. This is the same number of shots
direct reclaim takes before declaring OOM. Kswapd will go to sleep on
that node until a direct reclaimer manages to reclaim some pages, thus
proving the node reclaimable again.
[hannes@cmpxchg.org: check kswapd failure against the cumulative nr_reclaimed count]
Link: http://lkml.kernel.org/r/20170306162410.GB2090@cmpxchg.org
[shakeelb@google.com: fix condition for throttle_direct_reclaim]
Link: http://lkml.kernel.org/r/20170314183228.20152-1-shakeelb@google.com
Link: http://lkml.kernel.org/r/20170228214007.5621-2-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Reported-by: Jia He <hejianet@gmail.com>
Tested-by: Jia He <hejianet@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Dmitry Shmidt <dimitrysh@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b050e3769c upstream.
Since commit 97a16fc82a ("mm, page_alloc: only enforce watermarks for
order-0 allocations"), __zone_watermark_ok() check for high-order
allocations will shortcut per-migratetype free list checks for
ALLOC_HARDER allocations, and return true as long as there's free page
of any migratetype. The intention is that ALLOC_HARDER can allocate
from MIGRATE_HIGHATOMIC free lists, while normal allocations can't.
However, as a side effect, the watermark check will then also return
true when there are pages only on the MIGRATE_ISOLATE list, or (prior to
CMA conversion to ZONE_MOVABLE) on the MIGRATE_CMA list. Since the
allocation cannot actually obtain isolated pages, and might not be able
to obtain CMA pages, this can result in a false positive.
The condition should be rare and perhaps the outcome is not a fatal one.
Still, it's better if the watermark check is correct. There also
shouldn't be a performance tradeoff here.
Link: http://lkml.kernel.org/r/20171102125001.23708-1-vbabka@suse.cz
Fixes: 97a16fc82a ("mm, page_alloc: only enforce watermarks for order-0 allocations")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PD#153120: mm: close debug print of cma alloc
These print message may print a lot when video playback
And cause it not smooth.
Change-Id: If1f18d5e8a0234f1daca2c6e803a23ba90354414
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
[ Upstream commit 2df26639e7 ]
Jia He has noticed that commit b9f00e147f ("mm, page_alloc: reduce
branches in zone_statistics") has an unintentional side effect that
remote node allocation requests are accounted as NUMA_MISS rathat than
NUMA_HIT and NUMA_OTHER if such a request doesn't use __GFP_OTHER_NODE.
There are many of these potentially because the flag is used very rarely
while we have many users of __alloc_pages_node.
Fix this by simply ignoring __GFP_OTHER_NODE (it can be removed in a
follow up patch) and treat all allocations that were satisfied from the
preferred zone's node as NUMA_HITS because this is the same node we
requested the allocation from in most cases. If this is not the local
node then we just account it as NUMA_OTHER rather than NUMA_LOCAL.
One downsize would be that an allocation request for a node which is
outside of the mempolicy nodemask would be reported as a hit which is a
bit weird but that was the case before b9f00e147f already.
Fixes: b9f00e147f ("mm, page_alloc: reduce branches in zone_statistics")
Link: http://lkml.kernel.org/r/20170102153057.9451-2-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Jia He <hejianet@gmail.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz> # with cbmc[1] superpowers
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Taku Izumi <izumi.taku@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 63cd448908 upstream.
If the call __alloc_contig_migrate_range() in alloc_contig_range returns
-EBUSY, processing continues so that test_pages_isolated() is called
where there is a tracepoint to identify the busy pages. However, it is
possible for busy pages to become available between the calls to these
two routines. In this case, the range of pages may be allocated.
Unfortunately, the original return code (ret == -EBUSY) is still set and
returned to the caller. Therefore, the caller believes the pages were
not allocated and they are leaked.
Update the comment to indicate that allocation is still possible even if
__alloc_contig_migrate_range returns -EBUSY. Also, clear return code in
this case so that it is not accidentally used or returned to caller.
Link: http://lkml.kernel.org/r/20171122185214.25285-1-mike.kravetz@oracle.com
Fixes: 8ef5849fa8 ("mm/cma: always check which page caused allocation failure")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Nazarewicz <mina86@mina86.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d135e57502 upstream.
In reset_deferred_meminit() we determine number of pages that must not
be deferred. We initialize pages for at least 2G of memory, but also
pages for reserved memory in this node.
The reserved memory is determined in this function:
memblock_reserved_memory_within(), which operates over physical
addresses, and returns size in bytes. However, reset_deferred_meminit()
assumes that that this function operates with pfns, and returns page
count.
The result is that in the best case machine boots slower than expected
due to initializing more pages than needed in single thread, and in the
worst case panics because fewer than needed pages are initialized early.
Link: http://lkml.kernel.org/r/20171021011707.15191-1-pasha.tatashin@oracle.com
Fixes: 864b9a393d ("mm: consider memblock reservations for deferred memory initialization sizing")
Signed-off-by: Pavel Tatashin <pasha.tatashin@oracle.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PD#153120: mm: close debug print of cma alloc
These print message may print a lot when video playback
And cause it not smooth.
Change-Id: I77920e902189ab4aa1ec594d351749ee4aa3a3ee
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
PD#151104: mm: add pagetrace function
1. implement pagetrace as a driver; you can get information
of how many pages allocated by each function by read:
cat /proc/pagetrace
2. fix wrong statistics of free memory of each migrate type.
Change-Id: Ib2dff4bb5b3dd288ee188007352fc7b353eda100
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
PD#152454: mm: fix cma allocate fail problem
boost work shoud return right value of __alloc_contig_migrate_range
Change-Id: I8275a8541cd263f5ea5574fba1c053a30e3cbc80
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
commit 4855e4a7f2 upstream.
There is race between page freeing and unreserved highatomic.
CPU 0 CPU 1
free_hot_cold_page
mt = get_pfnblock_migratetype
set_pcppage_migratetype(page, mt)
unreserve_highatomic_pageblock
spin_lock_irqsave(&zone->lock)
move_freepages_block
set_pageblock_migratetype(page)
spin_unlock_irqrestore(&zone->lock)
free_pcppages_bulk
__free_one_page(mt) <- mt is stale
By above race, a page on CPU 0 could go non-highorderatomic free list
since the pageblock's type is changed. By that, unreserve logic of
highorderatomic can decrease reserved count on a same pageblock severak
times and then it will make mismatch between nr_reserved_highatomic and
the number of reserved pageblock.
So, this patch verifies whether the pageblock is highatomic or not and
decrease the count only if the pageblock is highatomic.
Link: http://lkml.kernel.org/r/1476259429-18279-3-git-send-email-minchan@kernel.org
Signed-off-by: Minchan Kim <minchan@kernel.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Sangseok Lee <sangseok.lee@lge.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Miles Chen <miles.chen@mediatek.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PD#150914: mm: optimize for cma allocation/free time
1. NOT use __GFP_COLD during cma is under allcating to
avoid page lock long time;
2. Mark __GFP_BDEV for alloc_migrate_target to avoid
migrate allocate cma page again in cma allocate;
3. use SMP to allocate cma;
4. try to free large order of pages for cma.
Change-Id: I2213444a089ce27ee6486c84dea03e0584086cd3
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
PD#150206: mm: forward memory reclaim process
memory reclaim process is usually do when first allocate failed
from buddy. This will cause some process crash especially when
memory is low, because there is not enough time to do reclaim
process. This change try to wake up kswapd process to reclaim
memory if free memory is less than high water mark. After apply
this change, free memory is usually keeps high than high_water
mark.
Change-Id: I03b5b931b6ec1fe1345d6f5cf8c15150438a26d3
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
PD#150194: mm: update cma policy and debug sysfs
1. For movable memory, using CMA first, this can help to avoid
low memory issue and driver can't allocate pages issue;
2. change /proc/slabinfo sysfs with total size print of each slab;
3. change /proc/pagetypeinfo with toatal pages of each migrate type;
4. add statistics for free pages of each migrate type, and show in
echo m > /proc/sysrq-trigger.
Change-Id: I2fdab73c2d1278cd025185362a1159e4f683166b
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
commit 3010f87650 upstream.
There is existing use after free bug when deferred struct pages are
enabled:
The memblock_add() allocates memory for the memory array if more than
128 entries are needed. See comment in e820__memblock_setup():
* The bootstrap memblock region count maximum is 128 entries
* (INIT_MEMBLOCK_REGIONS), but EFI might pass us more E820 entries
* than that - so allow memblock resizing.
This memblock memory is freed here:
free_low_memory_core_early()
We access the freed memblock.memory later in boot when deferred pages
are initialized in this path:
deferred_init_memmap()
for_each_mem_pfn_range()
__next_mem_pfn_range()
type = &memblock.memory;
One possible explanation for why this use-after-free hasn't been hit
before is that the limit of INIT_MEMBLOCK_REGIONS has never been
exceeded at least on systems where deferred struct pages were enabled.
Tested by reducing INIT_MEMBLOCK_REGIONS down to 4 from the current 128,
and verifying in qemu that this code is getting excuted and that the
freed pages are sane.
Link: http://lkml.kernel.org/r/1502485554-318703-2-git-send-email-pasha.tatashin@oracle.com
Fixes: 7e18adb4f8 ("mm: meminit: initialise remaining struct pages in parallel with kswapd")
Signed-off-by: Pavel Tatashin <pasha.tatashin@oracle.com>
Reviewed-by: Steven Sistare <steven.sistare@oracle.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Bob Picco <bob.picco@oracle.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 75dddef325 upstream.
The RDMA subsystem can generate several thousand of these messages per
second eventually leading to a kernel crash. Ratelimit these messages
to prevent this crash.
Doug said:
"I've been carrying a version of this for several kernel versions. I
don't remember when they started, but we have one (and only one) class
of machines: Dell PE R730xd, that generate these errors. When it
happens, without a rate limit, we get rcu timeouts and kernel oopses.
With the rate limit, we just get a lot of annoying kernel messages but
the machine continues on, recovers, and eventually the memory
operations all succeed"
And:
"> Well... why are all these EBUSY's occurring? It sounds inefficient
> (at least) but if it is expected, normal and unavoidable then
> perhaps we should just remove that message altogether?
I don't have an answer to that question. To be honest, I haven't
looked real hard. We never had this at all, then it started out of the
blue, but only on our Dell 730xd machines (and it hits all of them),
but no other classes or brands of machines. And we have our 730xd
machines loaded up with different brands and models of cards (for
instance one dedicated to mlx4 hardware, one for qib, one for mlx5, an
ocrdma/cxgb4 combo, etc), so the fact that it hit all of the machines
meant it wasn't tied to any particular brand/model of RDMA hardware.
To me, it always smelled of a hardware oddity specific to maybe the
CPUs or mainboard chipsets in these machines, so given that I'm not an
mm expert anyway, I never chased it down.
A few other relevant details: it showed up somewhere around 4.8/4.9 or
thereabouts. It never happened before, but the prinkt has been there
since the 3.18 days, so possibly the test to trigger this message was
changed, or something else in the allocator changed such that the
situation started happening on these machines?
And, like I said, it is specific to our 730xd machines (but they are
all identical, so that could mean it's something like their specific
ram configuration is causing the allocator to hit this on these
machine but not on other machines in the cluster, I don't want to say
it's necessarily the model of chipset or CPU, there are other bits of
identicalness between these machines)"
Link: http://lkml.kernel.org/r/499c0f6cc10d6eb829a67f2a4d75b4228a9b356e.1501695897.git.jtoppins@redhat.com
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Reviewed-by: Doug Ledford <dledford@redhat.com>
Tested-by: Doug Ledford <dledford@redhat.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit adb1fe9ae2 upstream.
Linus suggested we try to remove some of the low-hanging fruit related
to kernel address exposure in dmesg. The only leaks I see on my local
system are:
Freeing SMP alternatives memory: 32K (ffffffff9e309000 - ffffffff9e311000)
Freeing initrd memory: 10588K (ffffa0b736b42000 - ffffa0b737599000)
Freeing unused kernel memory: 3592K (ffffffff9df87000 - ffffffff9e309000)
Freeing unused kernel memory: 1352K (ffffa0b7288ae000 - ffffa0b728a00000)
Freeing unused kernel memory: 632K (ffffa0b728d62000 - ffffa0b728e00000)
Linus says:
"I suspect we should just remove [the addresses in the 'Freeing'
messages]. I'm sure they are useful in theory, but I suspect they
were more useful back when the whole "free init memory" was
originally done.
These days, if we have a use-after-free, I suspect the init-mem
situation is the easiest situation by far. Compared to all the dynamic
allocations which are much more likely to show it anyway. So having
debug output for that case is likely not all that productive."
With this patch the freeing messages now look like this:
Freeing SMP alternatives memory: 32K
Freeing initrd memory: 10588K
Freeing unused kernel memory: 3592K
Freeing unused kernel memory: 1352K
Freeing unused kernel memory: 632K
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/6836ff90c45b71d38e5d4405aec56fa9e5d1d4b2.1477405374.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Kees Cook <keescook@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 864b9a393d upstream.
We have seen an early OOM killer invocation on ppc64 systems with
crashkernel=4096M:
kthreadd invoked oom-killer: gfp_mask=0x16040c0(GFP_KERNEL|__GFP_COMP|__GFP_NOTRACK), nodemask=7, order=0, oom_score_adj=0
kthreadd cpuset=/ mems_allowed=7
CPU: 0 PID: 2 Comm: kthreadd Not tainted 4.4.68-1.gd7fe927-default #1
Call Trace:
dump_stack+0xb0/0xf0 (unreliable)
dump_header+0xb0/0x258
out_of_memory+0x5f0/0x640
__alloc_pages_nodemask+0xa8c/0xc80
kmem_getpages+0x84/0x1a0
fallback_alloc+0x2a4/0x320
kmem_cache_alloc_node+0xc0/0x2e0
copy_process.isra.25+0x260/0x1b30
_do_fork+0x94/0x470
kernel_thread+0x48/0x60
kthreadd+0x264/0x330
ret_from_kernel_thread+0x5c/0xa4
Mem-Info:
active_anon:0 inactive_anon:0 isolated_anon:0
active_file:0 inactive_file:0 isolated_file:0
unevictable:0 dirty:0 writeback:0 unstable:0
slab_reclaimable:5 slab_unreclaimable:73
mapped:0 shmem:0 pagetables:0 bounce:0
free:0 free_pcp:0 free_cma:0
Node 7 DMA free:0kB min:0kB low:0kB high:0kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:52428800kB managed:110016kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:320kB slab_unreclaimable:4672kB kernel_stack:1152kB pagetables:0kB unstable:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes
lowmem_reserve[]: 0 0 0 0
Node 7 DMA: 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB 0*8192kB 0*16384kB = 0kB
0 total pagecache pages
0 pages in swap cache
Swap cache stats: add 0, delete 0, find 0/0
Free swap = 0kB
Total swap = 0kB
819200 pages RAM
0 pages HighMem/MovableOnly
817481 pages reserved
0 pages cma reserved
0 pages hwpoisoned
the reason is that the managed memory is too low (only 110MB) while the
rest of the the 50GB is still waiting for the deferred intialization to
be done. update_defer_init estimates the initial memoty to initialize
to 2GB at least but it doesn't consider any memory allocated in that
range. In this particular case we've had
Reserving 4096MB of memory at 128MB for crashkernel (System RAM: 51200MB)
so the low 2GB is mostly depleted.
Fix this by considering memblock allocations in the initial static
initialization estimation. Move the max_initialise to
reset_deferred_meminit and implement a simple memblock_reserved_memory
helper which iterates all reserved blocks and sums the size of all that
start below the given address. The cumulative size is than added on top
of the initial estimation. This is still not ideal because
reset_deferred_meminit doesn't consider holes and so reservation might
be above the initial estimation whihch we ignore but let's make the
logic simpler until we really need to handle more complicated cases.
Fixes: 3a80a7fa79 ("mm: meminit: initialise a subset of struct pages if CONFIG_DEFERRED_STRUCT_PAGE_INIT is set")
Link: http://lkml.kernel.org/r/20170531104010.GI27783@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Tested-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 62be1511b1 upstream.
Patch series "more robust PF_MEMALLOC handling"
This series aims to unify the setting and clearing of PF_MEMALLOC, which
prevents recursive reclaim. There are some places that clear the flag
unconditionally from current->flags, which may result in clearing a
pre-existing flag. This already resulted in a bug report that Patch 1
fixes (without the new helpers, to make backporting easier). Patch 2
introduces the new helpers, modelled after existing memalloc_noio_* and
memalloc_nofs_* helpers, and converts mm core to use them. Patches 3
and 4 convert non-mm code.
This patch (of 4):
__alloc_pages_direct_compact() sets PF_MEMALLOC to prevent deadlock
during page migration by lock_page() (see the comment in
__unmap_and_move()). Then it unconditionally clears the flag, which can
clear a pre-existing PF_MEMALLOC flag and result in recursive reclaim.
This was not a problem until commit a8161d1ed6 ("mm, page_alloc:
restructure direct compaction handling in slowpath"), because direct
compation was called only after direct reclaim, which was skipped when
PF_MEMALLOC flag was set.
Even now it's only a theoretical issue, as the new callsite of
__alloc_pages_direct_compact() is reached only for costly orders and
when gfp_pfmemalloc_allowed() is true, which means either
__GFP_NOMEMALLOC is in gfp_flags or in_interrupt() is true. There is no
such known context, but let's play it safe and make
__alloc_pages_direct_compact() robust for cases where PF_MEMALLOC is
already set.
Fixes: a8161d1ed6 ("mm, page_alloc: restructure direct compaction handling in slowpath")
Link: http://lkml.kernel.org/r/20170405074700.29871-2-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reported-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Boris Brezillon <boris.brezillon@free-electrons.com>
Cc: Chris Leech <cleech@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Josef Bacik <jbacik@fb.com>
Cc: Lee Duncan <lduncan@suse.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Richard Weinberger <richard@nod.at>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PD#144028: mm: avoid PFN busy when cma allocate
1. add cma_suitable function to check if GFP flags are
suitable for fallback of CMA.
2. cma page free should be relink to migrate type of CMA
freelist, not to MOVABLE freelist.
3. for block dev page cache, add GPF_BDEV flag to avoid
it using CMA.
4. use CONFIG_AMLOGIC_MODIFY to wrap AMLOGIC modified kernel
code.
Change-Id: I09daf9d7e3e62086e7b13654667e2db692f7ca08
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
commit e02dc017c3 upstream.
When @node_reclaim_node isn't 0, the page allocator tries to reclaim
pages if the amount of free memory in the zones are below the low
watermark. On Power platform, none of NUMA nodes are scanned for page
reclaim because no nodes match the condition in zone_allows_reclaim().
On Power platform, RECLAIM_DISTANCE is set to 10 which is the distance
of Node-A to Node-A. So the preferred node even won't be scanned for
page reclaim.
__alloc_pages_nodemask()
get_page_from_freelist()
zone_allows_reclaim()
Anton proposed the test code as below:
# cat alloc.c
:
int main(int argc, char *argv[])
{
void *p;
unsigned long size;
unsigned long start, end;
start = time(NULL);
size = strtoul(argv[1], NULL, 0);
printf("To allocate %ldGB memory\n", size);
size <<= 30;
p = malloc(size);
assert(p);
memset(p, 0, size);
end = time(NULL);
printf("Used time: %ld seconds\n", end - start);
sleep(3600);
return 0;
}
The system I use for testing has two NUMA nodes. Both have 128GB
memory. In below scnario, the page caches on node#0 should be reclaimed
when it encounters pressure to accommodate request of allocation.
# echo 2 > /proc/sys/vm/zone_reclaim_mode; \
sync; \
echo 3 > /proc/sys/vm/drop_caches; \
# taskset -c 0 cat file.32G > /dev/null; \
grep FilePages /sys/devices/system/node/node0/meminfo
Node 0 FilePages: 33619712 kB
# taskset -c 0 ./alloc 128
# grep FilePages /sys/devices/system/node/node0/meminfo
Node 0 FilePages: 33619840 kB
# grep MemFree /sys/devices/system/node/node0/meminfo
Node 0 MemFree: 186816 kB
With the patch applied, the pagecache on node-0 is reclaimed when its
free memory is running out. It's the expected behaviour.
# echo 2 > /proc/sys/vm/zone_reclaim_mode; \
sync; \
echo 3 > /proc/sys/vm/drop_caches
# taskset -c 0 cat file.32G > /dev/null; \
grep FilePages /sys/devices/system/node/node0/meminfo
Node 0 FilePages: 33605568 kB
# taskset -c 0 ./alloc 128
# grep FilePages /sys/devices/system/node/node0/meminfo
Node 0 FilePages: 1379520 kB
# grep MemFree /sys/devices/system/node/node0/meminfo
Node 0 MemFree: 317120 kB
Fixes: 5f7a75acdb ("mm: page_alloc: do not cache reclaim distances")
Link: http://lkml.kernel.org/r/1486532455-29613-1-git-send-email-gwshan@linux.vnet.ibm.com
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Anton Blanchard <anton@samba.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e47483bca2 upstream.
Ganapatrao Kulkarni reported that the LTP test cpuset01 in stress mode
triggers OOM killer in few seconds, despite lots of free memory. The
test attempts to repeatedly fault in memory in one process in a cpuset,
while changing allowed nodes of the cpuset between 0 and 1 in another
process.
The problem comes from insufficient protection against cpuset changes,
which can cause get_page_from_freelist() to consider all zones as
non-eligible due to nodemask and/or current->mems_allowed. This was
masked in the past by sufficient retries, but since commit 682a3385e7
("mm, page_alloc: inline the fast path of the zonelist iterator") we fix
the preferred_zoneref once, and don't iterate over the whole zonelist in
further attempts, thus the only eligible zones might be placed in the
zonelist before our starting point and we always miss them.
A previous patch fixed this problem for current->mems_allowed. However,
cpuset changes also update the task's mempolicy nodemask. The fix has
two parts. We have to repeat the preferred_zoneref search when we
detect cpuset update by way of seqcount, and we have to check the
seqcount before considering OOM.
[akpm@linux-foundation.org: fix typo in comment]
Link: http://lkml.kernel.org/r/20170120103843.24587-5-vbabka@suse.cz
Fixes: c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone in a zonelist twice")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reported-by: Ganapatrao Kulkarni <gpkulkarni@gmail.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 16096c25bf upstream.
Ganapatrao Kulkarni reported that the LTP test cpuset01 in stress mode
triggers OOM killer in few seconds, despite lots of free memory. The
test attempts to repeatedly fault in memory in one process in a cpuset,
while changing allowed nodes of the cpuset between 0 and 1 in another
process.
One possible cause is that in the fast path we find the preferred
zoneref according to current mems_allowed, so that it points to the
middle of the zonelist, skipping e.g. zones of node 1 completely. If
the mems_allowed is updated to contain only node 1, we never reach it in
the zonelist, and trigger OOM before checking the cpuset_mems_cookie.
This patch fixes the particular case by redoing the preferred zoneref
search if we switch back to the original nodemask. The condition is
also slightly changed so that when the last non-root cpuset is removed,
we don't miss it.
Note that this is not a full fix, and more patches will follow.
Link: http://lkml.kernel.org/r/20170120103843.24587-3-vbabka@suse.cz
Fixes: 682a3385e7 ("mm, page_alloc: inline the fast path of the zonelist iterator")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reported-by: Ganapatrao Kulkarni <gpkulkarni@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ea57485af8 upstream.
Patch series "fix premature OOM regression in 4.7+ due to cpuset races".
This is v2 of my attempt to fix the recent report based on LTP cpuset
stress test [1]. The intention is to go to stable 4.9 LTSS with this,
as triggering repeated OOMs is not nice. That's why the patches try to
be not too intrusive.
Unfortunately why investigating I found that modifying the testcase to
use per-VMA policies instead of per-task policies will bring the OOM's
back, but that seems to be much older and harder to fix problem. I have
posted a RFC [2] but I believe that fixing the recent regressions has a
higher priority.
Longer-term we might try to think how to fix the cpuset mess in a better
and less error prone way. I was for example very surprised to learn,
that cpuset updates change not only task->mems_allowed, but also
nodemask of mempolicies. Until now I expected the parameter to
alloc_pages_nodemask() to be stable. I wonder why do we then treat
cpusets specially in get_page_from_freelist() and distinguish HARDWALL
etc, when there's unconditional intersection between mempolicy and
cpuset. I would expect the nodemask adjustment for saving overhead in
g_p_f(), but that clearly doesn't happen in the current form. So we
have both crazy complexity and overhead, AFAICS.
[1] https://lkml.kernel.org/r/CAFpQJXUq-JuEP=QPidy4p_=FN0rkH5Z-kfB4qBvsf6jMS87Edg@mail.gmail.com
[2] https://lkml.kernel.org/r/7c459f26-13a6-a817-e508-b65b903a8378@suse.cz
This patch (of 4):
Since commit c33d6c06f6 ("mm, page_alloc: avoid looking up the first
zone in a zonelist twice") we have a wrong check for NULL preferred_zone,
which can theoretically happen due to concurrent cpuset modification. We
check the zoneref pointer which is never NULL and we should check the zone
pointer. Also document this in first_zones_zonelist() comment per Michal
Hocko.
Fixes: c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone in a zonelist twice")
Link: http://lkml.kernel.org/r/20170120103843.24587-2-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Ganapatrao Kulkarni <gpkulkarni@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a6de734bc0 upstream.
Vlastimil Babka pointed out that commit 479f854a20 ("mm, page_alloc:
defer debugging checks of pages allocated from the PCP") will allow the
per-cpu list counter to be out of sync with the per-cpu list contents if
a struct page is corrupted.
The consequence is an infinite loop if the per-cpu lists get fully
drained by free_pcppages_bulk because all the lists are empty but the
count is positive. The infinite loop occurs here
do {
batch_free++;
if (++migratetype == MIGRATE_PCPTYPES)
migratetype = 0;
list = &pcp->lists[migratetype];
} while (list_empty(list));
What the user sees is a bad page warning followed by a soft lockup with
interrupts disabled in free_pcppages_bulk().
This patch keeps the accounting in sync.
Fixes: 479f854a20 ("mm, page_alloc: defer debugging checks of pages allocated from the PCP")
Link: http://lkml.kernel.org/r/20161202112951.23346-2-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull gcc plugin fixes from Kees Cook:
- make sure required exports from gcc plugins are visible to gcc
- switch latent_entropy to unsigned long to avoid stack frame bloat
* tag 'gcc-plugins-v4.9-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
latent_entropy: Fix wrong gcc code generation with 64 bit variables
gcc-plugins: Export symbols needed by gcc
The stack frame size could grow too large when the plugin used long long
on 32-bit architectures when the given function had too many basic blocks.
The gcc warning was:
drivers/pci/hotplug/ibmphp_ebda.c: In function 'ibmphp_access_ebda':
drivers/pci/hotplug/ibmphp_ebda.c:409:1: warning: the frame size of 1108 bytes is larger than 1024 bytes [-Wframe-larger-than=]
This switches latent_entropy from u64 to unsigned long.
Thanks to PaX Team and Emese Revfy for the patch.
Signed-off-by: Kees Cook <keescook@chromium.org>
tao zeng