GitHub/LineageOS/G12/android_kernel_amlogic_linux-4.9.git
4 years agovpp: display frame_width/height info for video_composer path [1/1]
Pengcheng Chen [Mon, 2 Dec 2019 09:30:48 +0000 (17:30 +0800)]
vpp: display frame_width/height info for video_composer path [1/1]

PD#SWPL-17616

Problem:
in video_composer path, frame_width/height is NA

Solution:
display frame_width/height info for video_composer path

Verify:
franklin

Change-Id: If334ca8186ee4b47ad0643f11c7ba1b0ea051f7f
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
4 years agohdmitx: correct the threshold value of rxsense [1/1]
Zongdong Jiao [Wed, 4 Dec 2019 08:23:54 +0000 (16:23 +0800)]
hdmitx: correct the threshold value of rxsense [1/1]

PD#SWPL-17841

Problem:
need correct the threshold value of rxsense

Solution:
correct the threshold value of rxsense, regress OTT-1968 to trunk

Verify:
G12/U212

Change-Id: I53f370cf51c0ad39e92fd64a0cc98cc4680eea4d
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
4 years agohdmitx: revert original hpd_state mark [1/1]
Zongdong Jiao [Wed, 4 Dec 2019 12:11:15 +0000 (20:11 +0800)]
hdmitx: revert original hpd_state mark [1/1]

PD#SWPL-17710

Problem:
Suppose hdmitx already output in uboot, and next kernel boots up,
the output will disable and then enable

Solution:
When hdmitx hpd_states initializes, read edid when hpd is high

Verify:
G12/U212

Change-Id: I538b8af2d59a7b2b34739f14d0b0cb6fd89aed2e
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
4 years agohdmitx: Fix parsing of HDR10plus and Dovi VSVDBs [1/1]
Zongdong Jiao [Tue, 3 Dec 2019 07:53:48 +0000 (15:53 +0800)]
hdmitx: Fix parsing of HDR10plus and Dovi VSVDBs [1/1]

PD#SWPL-17786

Problem:
No dolby vision on latest Vizio TVs providing an EDID containing
both dolby vision and HDR10plus VSVDBs in the EDID.

Solution:
If an EDID contains both HDR10plus and Dovi VSVDBs, we are
incorrectly resetting both dovi and hdr10plus parsed info structs.
As a result, we are only left with the parsed info of the later
VSVDB in the EDID. So, if we have a hdr10plus VSVDB after dovi
VSVDB, dovi info will be reset while parsing for hdr10plus, and
we will not report any dovi capabilities to the frameworks. This
is what is happening on these TVs in question.

Parse HDR10plus and Dovi VSVDBs independently of one another. Do
not reset parsed info of one while parsing for the other.

Since Dovi->HDR10+->Dovi transitions are not defined, send zero
drm/vsif packets while exiting HDR10+ playback if going to dovi
mode next.

Verify:
HDR10plus and Dovi capabilities can be simultaneously and correctly
reported to the frameworks.

Change-Id: I7a344638e5a923c88ef23f0b3a03480c5a564548
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
4 years agohdmitx: add a fake hpd for system test [1/1]
Zongdong Jiao [Fri, 15 Nov 2019 08:03:43 +0000 (16:03 +0800)]
hdmitx: add a fake hpd for system test [1/1]

PD#SWPL-16891

Problem:
For automatical test, the HDMI cable are always connected between Tx and Rx
without manully action.

Solution:
add a fake hpd for system test
echo 0[1] > /sys/class/amhdmitx/amhdmitx0/fake_plug

Verify:
G12/U212

Change-Id: I826fc0094845a13066a1535a4a87727b4ecad452
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
(cherry picked from commit e764ebfccc9502eb027f20527b58f7d4f509d1fd)

4 years agohdmitx: update edid attach after reading once [1/1]
Zongdong Jiao [Mon, 25 Nov 2019 14:10:13 +0000 (22:10 +0800)]
hdmitx: update edid attach after reading once [1/1]

PD#SWPL-17476

Problem:
After kernel boots up, there will no edid info for vinfo

Solution:
Need update edid attach after reading once
In kernel init, hdmitx / vout / edid with attach once time
Normal replug, plugin / edid / uevent / mode setting / attach
Also, init DRM_DB[0] data.

Verify:
G12/U212

Change-Id: I188f0972c882a17efcbc568ef7d0db861b9b6c0a
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
(cherry picked from commit 1c185cb0c069638c6391f424637f57abdaf99570)

4 years agohdmitx: optimize for power up flow [1/1]
Hang Cheng [Wed, 20 Nov 2019 10:04:26 +0000 (18:04 +0800)]
hdmitx: optimize for power up flow [1/1]

PD#SWPL-17155

Problem:
box set env to 2160p30hz 422,12bit, connect to
Hisense LED50EC660US 4K TV which only support
maximum 3G mode, then power on box, the hpd is
low when run hdmitx hpd command, so hdmi is not
set. after enter kernel, vout set 2160p30hz
before edid is read, and the color space is
forcely changed to rgb, but the color depth is
still 12bit, so TV can't detect normally.

Solution:
keep hpd state low until edid read finishes,
so that the mode vout set is filtered by edid

Verify:
S905X3

Change-Id: I11d126c4a2165ffaf3892534b1602bad3d78618d
Signed-off-by: Hang Cheng <hang.cheng@amlogic.com>
(cherry picked from commit f0a8ac5dcd74b92919e9ae55dd88de781c567c91)

4 years agohdmitx: Coverity(upgraded) defect cleanup: hdmitx [1/1]
Zhengrong.Zhu [Fri, 22 Nov 2019 11:16:30 +0000 (19:16 +0800)]
hdmitx: Coverity(upgraded) defect cleanup: hdmitx [1/1]

PD#SWPL-16574

Problem:
Coverity(upgraded) defect cleanup: hdmitx

Solution:
Modify code according to coverity

Verify:
verify on the board of U212

Change-Id: Iac69fb896dd22b38ae56b58701d6fd26aefe958c
Signed-off-by: Zhengrong Zhu <zhengrong.zhu@amlogic.com>
(cherry picked from commit cbbac9335c19d09c6bf50e6069d79f973129210a)

4 years agohdmitx: Playback HLG video color is abnormal during playing [1/1]
Zhengrong.Zhu [Fri, 15 Nov 2019 10:37:54 +0000 (18:37 +0800)]
hdmitx: Playback HLG video color is abnormal during playing [1/1]

PD#SWPL-16374

Problem:
Playback HLG video color display is abnormal during playing

Solution:
Modify code related to color space.

Verify:
verify on the board of U212

Change-Id: Ie01ba1f9931a5e8f12bd1cfd70fd9ea752194f9b
Signed-off-by: Zhengrong Zhu <zhengrong.zhu@amlogic.com>
(cherry picked from commit 82f2f9ff78c08c1b51738501dac9c6f213d4b33f)

4 years agohdmitx: fix sda hold low issue [1/1]
Zongdong Jiao [Fri, 25 Oct 2019 08:29:51 +0000 (16:29 +0800)]
hdmitx: fix sda hold low issue [1/1]

PD#SWPL-8231

Problem:
When HDMI cable plugin, the SDA may hold low forever or for short time.

Solution:
Add sw_reset_flt reset to resolve hold low forever issue

Verify:
G12/Dongle

Change-Id: I53bbf704bb48a2cd3cc2ebef1444c69ad1623e87
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
(cherry picked from commit b5c467ea8c3f6cd180b0cd00fa6a5413252119d4)

4 years agocec: cec b ARB timing cause cec 9.6.1 fail [1/1]
Yong Qin [Sun, 1 Dec 2019 07:41:26 +0000 (15:41 +0800)]
cec: cec b ARB timing cause cec 9.6.1 fail [1/1]

PD#SWPL-17658

Problem:
ARB timing setting is too large, it will cause
cec cts 9.6.1 fail

Solution:
reducec ARB timing

Verify:
sm1-AC214

Change-Id: If6bab18b729fc6f1dc010aa6289dc3e887a0f6f6
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
4 years agoamvideo: add dv support for video_receiver [1/2]
Brian Zhu [Tue, 26 Nov 2019 17:55:13 +0000 (01:55 +0800)]
amvideo: add dv support for video_receiver [1/2]

PD#SWPL-17393

Problem:
video receiver has no dv vframe operation.

Solution:
Add dv vframe operation.

Verify:
Verifed by u212

Change-Id: I0caf24d85a0d6d8382c13d72d9a439390ac21047
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
4 years agomedia_module: h264 playback crash [1/1]
Peng Yixin [Wed, 27 Nov 2019 09:57:59 +0000 (17:57 +0800)]
media_module: h264 playback crash [1/1]

PD#SWPL-17322

Problem:
    h264 playback crash.

Solution:
    Determine wether file_vf is null

Verify:
    U212

Change-Id: I1643c43ef0eb55aa88beeb47b2f3c8c180b9a5a2
Signed-off-by: Peng Yixin <yixin.peng@amlogic.com>
4 years agoosd: fix reboot quiescent caused screen blank issue [1/1]
Pengcheng Chen [Thu, 28 Nov 2019 10:53:26 +0000 (18:53 +0800)]
osd: fix reboot quiescent caused screen blank issue [1/1]

PD#SWPL-17605

Problem:
in reboot quiescent, suspend & resume caused afbc_cfg
reg set err

Solution:
in resume, reset afbc_start to 0

Verify:
franklin

Change-Id: I6e019ee88a3cfeb5d39db4777a7c2ac6c9648b02
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
4 years agoosd: modify register table update area [1/1]
Cao Jian [Thu, 14 Nov 2019 02:27:35 +0000 (10:27 +0800)]
osd: modify register table update area [1/1]

PD#SWPL-16663

Problem:
display freezes and becomes slow/lags
while selecting PIP mode options in VLC app

Solution:
remove wait-vsync nearly vsync coming
use polling to avoid  long time waiting

Verify:
W400

Change-Id: I7498a68de3f6c45e0cd48aad2805f73a23adb7e6
Signed-off-by: Cao Jian <jian.cao@amlogic.com>
(cherry picked from commit 6e8f000c86613d0c63712152c9fec3b219464fe6)

4 years agov4lvideo: fix cat dev panic [1/1]
jintao xu [Thu, 28 Nov 2019 06:24:40 +0000 (14:24 +0800)]
v4lvideo: fix cat dev panic [1/1]

PD#SWPL-17478

Problem:
cat /dev/video33 panic

Solution:
read return 0

Verify:
U212

Change-Id: Idcabf7eca7d99d020fb759261fd7f613da83b096
Signed-off-by: jintao xu <jintao.xu@amlogic.com>
4 years agovideo_composer: fix tunnel mode seek black screen [1/1]
jintao xu [Wed, 27 Nov 2019 06:23:53 +0000 (14:23 +0800)]
video_composer: fix tunnel mode seek black screen [1/1]

PD#SWPL-17424

Problem:
tunnel mode seek black screen

Solution:
set block_policy 0 when sideband

Verify:
U212

Change-Id: Iac9aff1db39c0366178025fc06ca91571948fa7c
Signed-off-by: jintao xu <jintao.xu@amlogic.com>
4 years agopartiton: change Q partition table [1/7]
Xindong Xu [Wed, 27 Nov 2019 02:00:03 +0000 (10:00 +0800)]
partiton: change Q partition table [1/7]

PD#SWPL-17483

Problem:
partition size is not enough

Solution:
1. add factory partition
2. change tee size to 32MB

Verify:
newton

Change-Id: I0a8cb2695fb67d1d6200f70864b31fcd7cee57bc
Signed-off-by: Xindong Xu <xindong.xu@amlogic.com>
4 years agov4l2: v4l2 vf video support [2/3]
rongrong zhou [Sat, 23 Nov 2019 10:49:24 +0000 (18:49 +0800)]
v4l2: v4l2 vf video support [2/3]

PD#SWPL-16226

Problem:
v4l2 can not display on vf video

Solution:
add v4l2 vf video support

Verify:
U212

Change-Id: If04fc991edaa57c202f1adcf6b1be6d453bb2af9
Signed-off-by: rongrong zhou <rongrong.zhou@amlogic.com>
4 years agovideo_composer: add multiple video layer display. [1/1]
jintao xu [Mon, 12 Aug 2019 10:50:27 +0000 (18:50 +0800)]
video_composer: add multiple video layer display. [1/1]

PD#SWPL-5991

Problem:
video_composer bringup

Solution:
add v4lvideo、dma to display on the video layer via video_composer

Verify:
on U212

Change-Id: Iec1b75cc8c9aa0475912148a0687d95250aeb8a0
Signed-off-by: jintao xu <jintao.xu@amlogic.com>
4 years agodv: graphic blend test failed [1/1]
yao liu [Tue, 19 Nov 2019 08:27:21 +0000 (16:27 +0800)]
dv: graphic blend test failed [1/1]

PD#SWPL-16994

Problem:
DV in/DV out case, gmax is high

Solution:
Set gmax to 300 for DV in/DV out case
By the way, remove err code in osd

Verify:
SM1

Change-Id: I9b73d57d424dc7609d284811c01a0d809d49c974
Signed-off-by: yao liu <yao.liu@amlogic.com>
4 years ago[Sabrina] HDMI compliance 7-10 at 222MHz fail [1/1]
Edward Ho [Tue, 13 Aug 2019 03:36:01 +0000 (11:36 +0800)]
[Sabrina] HDMI compliance 7-10 at 222MHz fail [1/1]

PD#GH-57
BUG:138766274

Problem:
HDMI compliance 7-10 at 222MHz fail

Solution
Modify register [FF63C3A0]=33EB4252

Verify:
sabrina-P1

Change-Id: I285b54a9faa740fc4552910a946b187cf26b2237
Signed-off-by: Edward Ho <edward.ho@amlogic.com>
4 years agovpp: new vpp driver to support vd layer switch [1/1]
Brian Zhu [Mon, 4 Nov 2019 16:20:36 +0000 (00:20 +0800)]
vpp: new vpp driver to support vd layer switch [1/1]

PD#SWPL-10060

Problem:
VD1 and VD2 used the fixed vfm path, it can not
switch the vfm path with frame base.

Solution:
1) Add new interface for set/get display path.
2) Using new vpp function which will remove the
dependency between vfm and vd mif.

Verify:
Verifed by U212, AC211

Change-Id: I84df441315005c8fa885ae83504e234f3f4a843f
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
4 years agodts: fixed dmx2 wrong irq in Q [1/1]
pengfei.liu [Mon, 18 Nov 2019 02:20:36 +0000 (10:20 +0800)]
dts: fixed dmx2 wrong irq in Q [1/1]

PD#RSP-287

Problem:
dmx2 IRQ is wrong in Q

Solution:
correct dmx2 IRQ num in Q

Verify:
local verify

Change-Id: I00b48755675cbf198087ac7f0825eea0597c6fe5
Signed-off-by: pengfei.liu <pengfei.liu@amlogic.com>
5 years agodts: add partition dts for ab_update in Q [1/10]
Xindong Xu [Wed, 13 Nov 2019 08:29:40 +0000 (16:29 +0800)]
dts: add partition dts for ab_update in Q [1/10]

PD#SWPL-15306

Problem:
need to support ab_update in Q

Solution:
add partition dts for ab_update in Q

Verify:
franklin

Change-Id: I0b8e16f2cf95b8b817c7659f0ef1b9d5c13e8777
Signed-off-by: Xindong Xu <xindong.xu@amlogic.com>
5 years agohdmitx: rewrite allm / contenttype [1/1]
Zongdong Jiao [Tue, 12 Nov 2019 14:39:57 +0000 (22:39 +0800)]
hdmitx: rewrite allm / contenttype [1/1]

PD#SWPL-14891

Problem:
need rewrite allm / contenttype

Solution:
rewrite allm / contenttype

Verify:
G12/U212

Change-Id: I3a41c963b1ba359546ed1286cc5c587f579aa441
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
5 years agohdr: hdr policy default follow sink(always hdr). [1/1]
Luan Yuan [Fri, 8 Nov 2019 09:57:33 +0000 (17:57 +0800)]
hdr: hdr policy default follow sink(always hdr). [1/1]

PD#SWPL-16581

Problem:
the policy is follow source(adaptive hdr).

Solution:
change the policy to follow sink(always hdr).

Verify:
verify by franklin.

Change-Id: Ic3608f296dca2975252586169c9d85b46d5c976e
Signed-off-by: Luan Yuan <luan.yuan@amlogic.com>
5 years agohdmitx: rewrite edidinfo to vinfo [1/1]
Zongdong Jiao [Thu, 7 Nov 2019 12:27:38 +0000 (20:27 +0800)]
hdmitx: rewrite edidinfo to vinfo [1/1]

PD#SWPL-14910

Problem:
When HDMI plugin, it will read edid, parse edid, (and here dv mode maybe changed),
notify, etc...

Solution:
rewrite edidinfo to vinfo and update when tx mode is ready

Verify:
G12/U212

Change-Id: I81eb1943f6444fd2c87e0353f6e2782bd9c6c9c1
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
(cherry picked from commit 967b21510a9c046bb89d559524c3753f958e16bf)

5 years agodts: fix the error of gxl_p212_1g.dts. [1/1]
Jianxiong Pan [Wed, 13 Nov 2019 07:06:14 +0000 (15:06 +0800)]
dts: fix the error of gxl_p212_1g.dts. [1/1]

PD#SWPL-12224

Problem:
have the errors in gxl_p212_1g.dts.

Solution:
fix the errors.

Verify:
build pass.

Change-Id: I8b87213392f8ca79dbf83c2ddfd15dae63627577
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
5 years agovdin: tm2,add h/v shrink functionality [1/1]
zhiwei.yuan [Thu, 31 Oct 2019 06:28:46 +0000 (14:28 +0800)]
vdin: tm2,add h/v shrink functionality [1/1]

PD#SWPL-15510

Problem:
small path of double write needs h/v shrink

Solution:
add to driver

Verify:
verified by t962x3_ab301

Change-Id: I1b2d2a5062cb092be681f4e15623c5ff58846714
Signed-off-by: zhiwei.yuan <zhiwei.yuan@amlogic.com>
5 years agopagetrace: fix /proc/pagetrace lost in 64bit os [1/1]
Tao Zeng [Thu, 7 Nov 2019 07:37:53 +0000 (15:37 +0800)]
pagetrace: fix /proc/pagetrace lost in 64bit os [1/1]

PD#SWPL-14800

Problem:
/proc/pagetrace is not exist in 64bit kernel.

Solution:
Using right macro for create fs node.

Verify:
x301

Change-Id: I27b8db9a349f37db4bc0d6752b06f3d6a6a6c46c
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
5 years agoethernet: improve ethernet connect quality [1/1]
Zhuo Wang [Thu, 7 Nov 2019 08:10:13 +0000 (16:10 +0800)]
ethernet: improve ethernet connect quality [1/1]

PD#SWPL-16498

Problem:
ethernet connect quality not good enough

Solution:
disable adjust logic

Verify:
AB311

Change-Id: I63db4e756006baf8818d7f35aeb1fd1e64fd797b
Signed-off-by: Zhuo Wang <zhuo.wang@amlogic.com>
5 years agoge2d: disable matrix for fillrect function [1/1]
Cao Jian [Fri, 1 Nov 2019 10:09:25 +0000 (18:09 +0800)]
ge2d: disable matrix for fillrect function [1/1]

PD#SWPL-14823

Problem:
ge2d FILLRECTANGLE operation not working properly
if src and dst are in different color spaces (yuv&rgb)

Solution:
disable matrix for this function

Verify:
ae400

Change-Id: Id11d96191e9e214506bd13621777499e35e216b2
Signed-off-by: Cao Jian <jian.cao@amlogic.com>
5 years agohdr: default disable hdr hist module [1/1]
MingLiang Dong [Fri, 1 Nov 2019 07:06:56 +0000 (15:06 +0800)]
hdr: default disable hdr hist module [1/1]

PD#SWPL-16107

Problem:
tm2 default hdr hist module enable,
it cause power consumption is high.

Solution:
1. default disable hdr hist module
2. if enable hdr2sdr, enable hist module

Verify:
verify on TM2

Change-Id: I648e9d5ac6739c1b6ef04db5ac3aec6e8f695cf7
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
5 years agomtd: nand: support new slc nand [2/2]
xianjun.liu [Fri, 11 Oct 2019 03:32:22 +0000 (11:32 +0800)]
mtd: nand: support new slc nand [2/2]

PD#SWPL-14719

Problem:
4Gb SLC nand MT29F4G08ABAFAWP,MX30LF4G28AD requsted by Market

Solution:
add the nand id into amlogic nand flash id table

Verify:
AXG-S400

Change-Id: If2bd3cc90585b2a3cbcd500a9354cde4c65bc355
Signed-off-by: xianjun.liu <xianjun.liu@amlogic.com>
5 years agoDTS: TL1/TM2 internal phy led is H0/H1. [1/1]
qi duan [Mon, 4 Nov 2019 10:27:06 +0000 (18:27 +0800)]
DTS: TL1/TM2 internal phy led is H0/H1. [1/1]

PD#SWPL-16192

Problem:
ab301 ethernet can not link

Solution:
LED GPIO is GPIOH_0 and GPIOH_1
TL1 is the same gpio

Verify:
ab301

Change-Id: Id4e0a94a021d72d3b952d2dd15a86caf45a16ca0
Signed-off-by: qi duan <qi.duan@amlogic.com>
5 years agohdmirx: Coverity defect cleanup:hdmi rx [1/1]
Zhengrong.Zhu [Sat, 2 Nov 2019 03:38:11 +0000 (11:38 +0800)]
hdmirx: Coverity defect cleanup:hdmi rx [1/1]

PD#SWPL-15167

Problem:
Coverity defect cleanup:hdmi rx

Solution:
Modify code according to coverity

Verify:
verify on the board of x3011

Change-Id: I1f8bac51101a5ae1a8a27ffd14e899b75ab99c04
Signed-off-by: Zhengrong Zhu <zhengrong.zhu@amlogic.com>
5 years agodts: km3: add attestationdevidbox to more platforms [1/1]
Mingyen Hung [Tue, 5 Nov 2019 08:34:59 +0000 (16:34 +0800)]
dts: km3: add attestationdevidbox to more platforms [1/1]

PD#SWPL-12224

Problem:
Support Device ID attestation on multiple platforms

Solution:
1. Add new key slot, attestationdevidbox, to unify keys

Verify:
1. Android P + all android platforms
2. run cts -m CtsKeystoreTestCases -t \
android.keystore.cts.KeyAttestationTest#testDeviceIdAttestation

Change-Id: I95ba42b2df808a4aec5ad47d00ec9684c36e4ad3
Signed-off-by: Mingyen Hung <mingyen.hung@amlogic.com>
5 years agohdmirx: fix hdcp2.2 error after resume [1/1]
yicheng shen [Wed, 16 Oct 2019 08:46:59 +0000 (16:46 +0800)]
hdmirx: fix hdcp2.2 error after resume [1/1]

PD#SWPL-15872

Problem:
ESM init failed after boot-up twice

Solution:
Use the correct HPI address when request_mem_region called.

Verify:
TL1 4g ddr

Change-Id: I5fd30166b3a36e5f4a12b8b6b893df9c9f20ddaf
Signed-off-by: yicheng shen <yicheng.shen@amlogic.com>
5 years agodrm: solve debug node write string issue [1/1]
Ao Xu [Wed, 23 Oct 2019 11:24:27 +0000 (19:24 +0800)]
drm: solve debug node write string issue [1/1]

PD#SWPL-15143

Problem:
userspace program use the write() syscall to write the node.
when write the strlen() length content, it will fail.

Solution:
use the buf[size] to instead of buf[size-1]

Verify:
U200

Change-Id: I886d9a1cbf3da459476bca76c9a5708ecbc20afe
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
5 years agoRevert "io: make register access in amvecm simple [1/1]"
Jianxin Pan [Tue, 5 Nov 2019 11:55:40 +0000 (04:55 -0700)]
Revert "io: make register access in amvecm simple [1/1]"

This reverts commit 7436b7dc4acf29ea9549249cbdd2ef6df1b91e92.

revert this now, print_vpp_lut() stack is 1840, when this applied,
it will bigger than 1880. werror will be triggerd.

Change-Id: Ia7b084ee800081a75c70f4e6ea6ce739f113f2b8
Signed-off-by: zihuan.ling <zihuan.ling@amlogic.com>
5 years agovdin: vdin add dolby vision de-scramble & scramble
Yong Qin [Sat, 12 Oct 2019 09:40:14 +0000 (17:40 +0800)]
vdin: vdin add dolby vision de-scramble & scramble

PD#SWPL-14888

Problem:
New feature, dolby source vdin have de-scramble,
scrable function

Solution:
add feature for dv de scramble function

Verify:
tm2 ab301

Change-Id: Ib7c10fcf53232c84411ae366618b2ddcd9948527
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
5 years agoatv_demod: fix atv audio no output when line frequency offset over default [1/1]
nengwen.chen [Fri, 1 Nov 2019 13:09:23 +0000 (21:09 +0800)]
atv_demod: fix atv audio no output when line frequency offset over default [1/1]

PD#SWPL-16188, PD#TV-11641, PD#TV-10181, PD#TV-9714

Problem:
fix atv audio no output when line frequency offset over 200Hz.

Solution:
1.fix atv audio no output when line frequency offset over 200Hz.
2.atv demod version: [V2.15].

Verify:
verified by tl1.

Change-Id: Ic43c707d3401ea93909856ff83a6842e14f0fc61
Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com>
5 years agoaudio: fix might_sleep warning at bootup [1/1]
Shuai Li [Tue, 5 Nov 2019 06:47:37 +0000 (14:47 +0800)]
audio: fix might_sleep warning at bootup [1/1]

PD#SWPL-16123

Problem:
might_sleep() is triggered by regmap_read().

[    3.106894@0] Workqueue: events tl1_acodec_release_fast_mode_work_func
[    3.106903@0] [bc059af4+  16][<c020e3a8>] show_stack+0x20/0x24
[    3.106910@0] [bc059b1c+  40][<c05dff94>] dump_stack+0xb8/0xf4
[    3.106917@0] [bc059b44+  40][<c0258138>] ___might_sleep+0x1c8/0x208
[    3.106922@0] [bc059b6c+  40][<c02581ec>] __might_sleep+0x74/0xac
[    3.106928@0] [bc059bb4+  72][<c0393a80>] kmem_cache_alloc_trace+0x12c/0x2c0
[    3.106935@0] [bc059bfc+  72][<c06bd03c>] regcache_rbtree_write+0x2dc/0x4f0
[    3.106941@0] [bc059c1c+  32][<c06bbbd8>] regcache_write+0x64/0x6c
[    3.106945@0] [bc059c44+  40][<c06b9260>] _regmap_write+0x144/0x16c
[    3.106948@0] [bc059c64+  32][<c06bab4c>] regmap_write+0x4c/0x6c
[    3.106953@0] [bc059c74+  16][<c0c79088>] snd_soc_component_write+0x28/0x48
[    3.106958@0] [bc059c84+  16][<c0c790c4>] snd_soc_write+0x1c/0x20
[    3.106962@0] [bc059c9c+  24][<c0c7c0b0>] tl1_acodec_start_up+0x24/0x44
[    3.106967@0] [bc059cbc+  32][<c0c7c1b8>]
tl1_acodec_release_fast_mode_work_func+0x54/0xe4
[    3.106973@0] [bc059cfc+  64][<c0245dc8>] process_one_work+0x200/0x624
[    3.106977@0] [bc059d3c+  64][<c0246dc4>] worker_thread+0x58/0x5c4
[    3.106982@0] [bc059d94+  88][<c024c5c0>] kthread+0x128/0x144
[    3.106987@0] [00000000+   0][<c020885c>] ret_from_fork+0x14/0x38

Solution:
This is a mmio regmap constrain using spinlock_irq to protect the
critical read and write.
But if the default reg cache is not allocated, the read routine will
malloc() buffer which is non-atomic. That is the root cause of this warning.
So we add num_reg_defaults_raw to tell regmap to malloc the buffer at
initializing stage.

Verify:
TL1.

Change-Id: I42f87fa89bbe99762ada36337b5ed6565b06b891
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
5 years agodeintlace: register setting default disable hdr2 [1/1]
qianqian.cai [Fri, 1 Nov 2019 10:16:16 +0000 (18:16 +0800)]
deintlace: register setting default disable hdr2  [1/1]

PD#SWPL-16107

Problem:
cpu_after_eq(MESON_CPU_MAJOR_ID_TM2)default disable hdr2

Solution:
default disable hdr2,from VLSI feijun
DI_HDR2_HIST_CTRL, 0x5510
DI_HDR2_HIST_H_START_END, 0x10000
DI_HDR2_HIST_V_START_END, 0x0

Verify:
verify on x301_t96x2

Change-Id: Ib646111fb866570a75775534e446807f1e3b4a91
Signed-off-by: qianqian.cai <qianqian.cai@amlogic.com>
5 years agodts: add support s805y config. [2/3]
Luan Yuan [Mon, 4 Nov 2019 09:14:04 +0000 (17:14 +0800)]
dts: add support s805y config. [2/3]

PD#SWPL-16193

Problem:
add support s805y config for anning.

Solution:
add support s805y config for anning.
Anning(P244) copy from P215, and change soc with s805y

Verify:
verify by anning.

Change-Id: I4e09521833f37d96066bf0643b0e99077cd0c7be
Signed-off-by: Luan Yuan <luan.yuan@amlogic.com>
5 years agohdmitx: Coverity defect cleanup: VOUT [1/1]
Zhengrong.Zhu [Sat, 2 Nov 2019 02:01:55 +0000 (10:01 +0800)]
hdmitx: Coverity defect cleanup: VOUT [1/1]

PD#SWPL-13989

Problem:
Coverity defect cleanup: VOUT

Solution:
Modify code according to coverity

Verify:
verify on the board of U212

Change-Id: I787429350be37da84ccd85332e97d472f45d3826
Signed-off-by: Zhengrong Zhu <zhengrong.zhu@amlogic.com>
5 years agoamlvideo2: Screen capture adds support for decoder data. [1/1]
renjiang.han [Thu, 17 Oct 2019 12:32:19 +0000 (20:32 +0800)]
amlvideo2: Screen capture adds support for decoder data. [1/1]

PD#SWPL-16197

Problem:
The data from the decoder is not equipped with a canvas.

Solution:
screenshot adds support for vframes without canvas.

Verify:
on X301

Change-Id: I706b87d6f59da27ffda7ace0d3486d6d978a292a
Signed-off-by: renjiang.han <renjiang.han@amlogic.com>
5 years agoio: make register access in amvecm simple [1/1]
Jianxin Pan [Thu, 24 Oct 2019 11:02:49 +0000 (19:02 +0800)]
io: make register access in amvecm simple [1/1]

PD#SWPL-15713

Problem:
There are too many register access in vsync isr from pq

Solution:
Make register access quicker

Verify:
TL1

Change-Id: I72972ccb8f2cf1bd53eda7fc40697cb5df4f5ede
Signed-off-by: Jianxin Pan <jianxin.pan@amlogic.com>
5 years agocec: finetune cecb ARB rising time [1/1]
Yong Qin [Tue, 22 Oct 2019 07:08:32 +0000 (15:08 +0800)]
cec: finetune cecb ARB rising time [1/1]

PD#SWPL-15616

Problem:
when cec bus conect other cec connector, may lose cec ARB

Solution:
finetune ARB rising time.

Verify:
tl1

Change-Id: Icf422505602f75307f0b91c7841f0aae5e09c22d
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
5 years agohdr: correct hdr10+ process path [1/1]
MingLiang Dong [Thu, 17 Oct 2019 11:08:23 +0000 (19:08 +0800)]
hdr: correct hdr10+ process path [1/1]

PD#SWPL-16139

Problem:
1. hdr10+ gen ebz curve error
2. hdr10+ process error

Solution:
1. fix hdr10+ ebz gen
2. correct hdr10+ process path

Verify:
verify on TL1

Change-Id: Ia33488ada2e0ebfe9b156f9f667188cf02164089
Signed-off-by: MingLiang Dong <mingliang.dong@amlogic.com>
5 years agobacklight: add bl policy support [1/1]
Evoke Zhang [Mon, 4 Nov 2019 05:04:31 +0000 (13:04 +0800)]
backlight: add bl policy support [1/1]

PD#TV-11347

Problem:
sometimes need power on from uboot state in kernel

Solution:
add bl policy to support different power on mode

Verify:
x301

Change-Id: I13c289523c6a77d713b3a770a5cdaf02afd01a11
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
5 years agodts: km4: add attestationdevidbox to more platforms [1/1]
Mingyen Hung [Tue, 5 Nov 2019 06:43:14 +0000 (14:43 +0800)]
dts: km4: add attestationdevidbox to more platforms [1/1]

PD#SWPL-12224

Problem:
Support Device ID attestation on Marconi and Dalton

Solution:
1. Add new key slot, attestationdevidbox, to unify keys

Verify:
1. Android Q + (Marconi, Dalton)
2. run cts -m CtsKeystoreTestCases -t \
android.keystore.cts.KeyAttestationTest#testDeviceIdAttestation

Change-Id: I2c61422a56fd29c201a05a29871af5daeb612070
Signed-off-by: Mingyen Hung <mingyen.hung@amlogic.com>
5 years agoaudio: bringup a new resample module on SM1/TM2 [1/1]
Zhe Wang [Sat, 6 Apr 2019 10:27:24 +0000 (18:27 +0800)]
audio: bringup a new resample module on SM1/TM2 [1/1]

PD#SWPL-5644

Problem:
bringup new hw resample modules on SM1/TM2

Solution:
bringup new hw resample modules on SM1/TM2
1) resample A is used for HDMIRX/SPDIFIN
2) resample B is used for LOOPBACK anti-aliasing
   if capture loopback data with 16KHz, enable it in dts.

Verify:
AB301

Change-Id: I0c2e403ff089a28e8d5695d854d13358ecffa313
Signed-off-by: Zhe Wang <Zhe.Wang@amlogic.com>
5 years agoclk: tm2: disable the pcie pll internal module when suspend [1/1]
Jian Hu [Fri, 18 Oct 2019 10:41:02 +0000 (18:41 +0800)]
clk: tm2: disable the pcie pll internal module when suspend [1/1]

PD#SWPL-15470

Problem:
tm2 pcie pll power dissipation is high when suspend

Solution:
disable the pcie pll internal module when suspend

Verify:
tm2

Change-Id: I298a7a15c58ebd68c379b0281284d71425207b0e
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
5 years agoethernet: switch to gpio when suspend [1/1]
Zhuo Wang [Thu, 31 Oct 2019 02:34:05 +0000 (10:34 +0800)]
ethernet: switch to gpio when suspend [1/1]

PD#SWPL-16033

Problem:
save pwr when suspend

Solution:
switch to gpio when suspend

Verify:
AB311

Change-Id: I1186263c1fccc82ac4299d95d606d35f5ee7c171
Signed-off-by: Zhuo Wang <zhuo.wang@amlogic.com>
5 years agodv: remove the wrong operation for txlx when dv off [1/1]
Brian Zhu [Fri, 1 Nov 2019 10:17:12 +0000 (18:17 +0800)]
dv: remove the wrong operation for txlx when dv off [1/1]

PD#SWPL-5164

Problem:
Vsync operate the wrong registers on txlx chip.

Solution:
Add chip condition for txlx

Verify:
Verified on R311

Change-Id: Ic3c10b7ebdab9d72cc9f05fe68be2442a1be6200
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
5 years agoaudio: auge: add mixer for eARC RX/TX latency and capability [1/1]
Xing Wang [Fri, 1 Nov 2019 07:47:10 +0000 (15:47 +0800)]
audio: auge: add mixer for eARC RX/TX latency and capability [1/1]

PD#SWPL-6918

Problem:
lack function to read/write eARC RX/TX latency and capability

Solution:
add mixer for eARC RX set/get latency and capability
add mixer for eARC RX get latency and capability

Verify:
ab311

Change-Id: I0d8aa00af7d856493417dc881a1a8c40497b12a5
Signed-off-by: Xing Wang <xing.wang@amlogic.com>
5 years agomm: fix crash in vmap stack [1/1]
Tao Zeng [Mon, 28 Oct 2019 09:22:50 +0000 (17:22 +0800)]
mm: fix crash in vmap stack [1/1]

PD#SWPL-12145

Problem:
When switch IRQ stack, there is a 16 bytes miss match of irq
stack. This may cause context not match some times;

Solution:
Save sp to right place

Verify:
X301

Change-Id: I75dfa5208cb3cec050c1568ac78ae2e783ca106b
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
5 years agokasan: fix pc crash if open kasan with vmap stack [1/1]
Tao Zeng [Fri, 25 Oct 2019 07:01:24 +0000 (15:01 +0800)]
kasan: fix pc crash if open kasan with vmap stack [1/1]

PD#SWPL-10141

Problem:
If open KASAN, connect wifi, and auto-start apk dianshijia, it will
soon crash as:
[ 56.655524@2] Unable to handle kernel NULL pointer dereference
               at virtual address 00000050
[ 56.658155@2] pgd = d0104000
[ 56.661008@2] [00000050] *pgd=00000000
[ 56.664744@2] Internal error: Oops: 80000005 1 PREEMPT SMP ARM
               common(O) firmware(O) media_clock(O) tb_detect(O)
               dnlp_alg ldim_alg mxl661_fe_32(O)
[ 56.709028@2] CPU: 2 PID: 4930 Comm: kworker/u9:3 Tainted: O 4.9.113 #1
[ 56.716908@2] Hardware name: Generic DT based system
[ 56.721858@2] task: f2632bc0 task.stack: c2010000
[ 56.726530@2] PC is at 0x50
[ 56.729301@2] LR is at 0x0

Solution:
Fix wrong pmd check address

Verify:
TL1 X301

Change-Id: I85e01418789801e0ef7691a9d8963c233e115b28
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
5 years agohdmirx: Coverity defect cleanup:hdmi rx [1/1]
Zhengrong.Zhu [Fri, 1 Nov 2019 07:07:45 +0000 (15:07 +0800)]
hdmirx: Coverity defect cleanup:hdmi rx [1/1]

PD#SWPL-15167

Problem:
Coverity defect cleanup:hdmi rx

Solution:
Modify code according to coverity

Verify:
verify on the board of x3011

Change-Id: If325bcd7ca3a7f85cd17ac8b57e4a15284231ccc
Signed-off-by: Zhengrong Zhu <zhengrong.zhu@amlogic.com>
5 years agotvafe: vbi: fix coverity [1/1]
Evoke Zhang [Fri, 1 Nov 2019 03:32:34 +0000 (11:32 +0800)]
tvafe: vbi: fix coverity [1/1]

PD#SWPL-15873

Problem:
Coverity defect cleanup.

Solution:
Coverity defect cleanup.

Verify:
x301

Change-Id: I4c5c5e8d01dbd8eb3389937e45461775ab1ac4ca
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
5 years agodts: increase codecmm to enable multiple encoder [1/1]
Ruoran Xi [Fri, 1 Nov 2019 08:58:56 +0000 (16:58 +0800)]
dts: increase codecmm to enable multiple encoder [1/1]

PD#SWPL-16131

Problem:
  codec memory too small to change video encode

Solution:
  increase codec memory size

Verify:
local

Change-Id: If09d00b97d607ba319c030516344dd3e9f96f7ec
Signed-off-by: Ruoran Xi <ruoran.xi@amlogic.com>
5 years agoreserved_memory: change print of all reserved memory [1/1]
Tao Zeng [Fri, 1 Nov 2019 09:03:02 +0000 (17:03 +0800)]
reserved_memory: change print of all reserved memory [1/1]

PD#SWPL-16049

Problem:
When boot, print information of all reserved memory is not nice
for reading.

Solution:
Change print format of reserved memory like:
        07400000 - 07500000,     1024 KB, ramoops@0x07400000
        05000000 - 05400000,     4096 KB, linux,secmon
        3f800000 - 40000000,     8192 KB, linux,meson-fb
        31c00000 - 3f800000,   225280 KB, linux,codec_mm_cma
        31c00000 - 31c00000,        0 KB, linux,codec_mm_reserved
        2e000000 - 30000000,    32768 KB, linux,ion-dev
        30800000 - 31c00000,    20480 KB, linux,vdin1_cma
        30000000 - 30800000,     8192 KB, linux,demod_cma
        2b800000 - 2e000000,    40960 KB, linux,di_cma
        2b400000 - 2b800000,     4096 KB, linux,emp_cma

Verify:
TL1 X301

Change-Id: I5a010118665a886896c7db8be2bb911813647e9e
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
5 years agodv: blank screen during Transition tests [1/1]
yao liu [Thu, 31 Oct 2019 07:47:49 +0000 (15:47 +0800)]
dv: blank screen during Transition tests [1/1]

PD#SWPL-16051

Problem:
Soure changed from DV to sdr, but core1 lut
is not update due to stb_core_setting_update_flag
is reset when frame dropped by amvideo

Solution:
Don't clear stb_core_setting_update_flag until
it has been used.

Verify:
U212

Change-Id: Ib1e445653978583ddda35c266000e812463f3f77
Signed-off-by: yao liu <yao.liu@amlogic.com>
5 years agoaudio: wrong mclk_pad setting for TL1 [1/1]
Shuai Li [Fri, 1 Nov 2019 19:41:47 +0000 (03:41 +0800)]
audio: wrong mclk_pad setting for TL1 [1/1]

PD#SWPL-16108

Problem:
Mclk_pad is set 0 by all three TDM module.
The result is that only the last one takes effect.

Solution:
Modify the mclk_pad setting and delete the unused one.

Verify:
TL1.

Change-Id: I2fa881e8ebcc2e7822dfbbea6a7ac0bfbc8ae0f9
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
5 years agoMerge "atv_demod: fix atv non-std signal handle [1/1]" into bringup/amlogic-4.9/sync2...
Jianxin Pan [Tue, 5 Nov 2019 09:19:20 +0000 (01:19 -0800)]
Merge "atv_demod: fix atv non-std signal handle [1/1]" into bringup/amlogic-4.9/sync20190611_forQ

5 years agoMerge "sdio: optimize sdio suspend/resume [1/1]" into bringup/amlogic-4.9/sync2019061...
Jianxin Pan [Tue, 5 Nov 2019 09:19:04 +0000 (01:19 -0800)]
Merge "sdio: optimize sdio suspend/resume [1/1]" into bringup/amlogic-4.9/sync20190611_forQ

5 years agoMerge "dv: do not reset the tv core when resolution changed [1/1]" into bringup/amlog...
Jianxin Pan [Tue, 5 Nov 2019 09:18:54 +0000 (01:18 -0800)]
Merge "dv: do not reset the tv core when resolution changed [1/1]" into bringup/amlogic-4.9/sync20190611_forQ

5 years agoatv_demod: fix atv non-std signal handle [1/1]
nengwen.chen [Wed, 30 Oct 2019 07:37:59 +0000 (15:37 +0800)]
atv_demod: fix atv non-std signal handle [1/1]

PD#TV-11542

Problem:
fix atv non std handle when video and audio are set at the same time.

Solution:
1.fix atv non std handle when video and audio are set at the same time.
2.atv demod version: [V2.14].
3.branch: amlogic-4.9-dev.

Verify:
verified by tl1.

Change-Id: I4bfc0e076ccde813329c315e4cbcead6e9bf6c94
Signed-off-by: nengwen.chen <nengwen.chen@amlogic.com>
5 years agosdio: optimize sdio suspend/resume [1/1]
Nan Li [Fri, 1 Nov 2019 08:39:14 +0000 (16:39 +0800)]
sdio: optimize sdio suspend/resume [1/1]

PD#SWPL-16036

Problem:
When the system starts up, wifi is turned on.
When we manually turn off wifi and start the
suspend & resume operation, we will find that
the sdio error is reported when we wake up

Solution:
Add the sdio_notify() interface to notify the
sdio controller when wifi is on/off, enabling
the keep_power standard to be configured when suspend.

Verify:
TL1

Change-Id: I48e0c31bf7ea80d3efd4b0d41b4af378e859dc15
Signed-off-by: Nan Li <nan.li@amlogic.com>
5 years agohdmitx: add hdr_priority [1/1]
Zongdong Jiao [Wed, 18 Sep 2019 07:47:33 +0000 (15:47 +0800)]
hdmitx: add hdr_priority [1/1]

PD#SWPL-14089

Problem:
Need add 'hdr_priority' to mask sysfs 'dv_cap'

Solution:
Parse the variable 'hdr_priority' from uboot environment

Verify:
G12/U212

Change-Id: I6fca18669c068ac793ca9230c6497b0ef2cbe07c
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
5 years agodv: do not reset the tv core when resolution changed [1/1]
Brian Zhu [Sat, 2 Nov 2019 10:46:55 +0000 (18:46 +0800)]
dv: do not reset the tv core when resolution changed [1/1]

PD#SWPL-5167

Problem:
Reset tv core will cause the display flashing.

Solution:
remove the reset operation

Verify:
Verified on R311

Change-Id: Id9aa97dd0462d04bb2d83e5f0247acf3bcaabcd7
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
5 years agodts: ac214 disable spdifin [1/1]
huijie.huang [Thu, 31 Oct 2019 08:59:50 +0000 (16:59 +0800)]
dts: ac214 disable spdifin [1/1]

PD#SWPL-16012

Problem:
spdifin gpio(GPIOH_5) conflict with HDMI

Solution:
don't use spdifin

Verify:
Android Q + Newton

Change-Id: Icb61bcab5421384b54ac03c697a16372e3e91577
Signed-off-by: huijie.huang <huijie.huang@amlogic.com>
5 years agopq: add yuv full to rgb full feature [2/2]
Yong Qin [Thu, 17 Oct 2019 02:29:51 +0000 (10:29 +0800)]
pq: add yuv full to rgb full feature [2/2]

PD#SWPL-15366

Problem:
according source type, add YUV full to RGB Full

Solution:
according source type, add YUV full to RGB Full

Verify:
TL1

Change-Id: I1afb7bc56161dee87d055a7686d99377db954be4
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
5 years agonet: stmmac: use correct DMA buffer size in the RX descriptor
Aaro Koskinen [Wed, 27 Mar 2019 20:35:35 +0000 (22:35 +0200)]
net: stmmac: use correct DMA buffer size in the RX descriptor

PD#TV-11316

[ Upstream commit 583e6361414903c5206258a30e5bd88cb03c0254 ]

We always program the maximum DMA buffer size into the receive descriptor,
although the allocated size may be less. E.g. with the default MTU size
we allocate only 1536 bytes. If somebody sends us a bigger frame, then
memory may get corrupted.

Fix by using exact buffer sizes.

Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Change-Id: I32f3f0758ee6df0217dd9d1275c08b04163d811b
Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com>
Signed-off-by: zihuan.ling <zihuan.ling@amlogic.com>
5 years agonet: stmmac: don't stop NAPI processing when dropping a packet
Aaro Koskinen [Wed, 27 Mar 2019 20:35:37 +0000 (22:35 +0200)]
net: stmmac: don't stop NAPI processing when dropping a packet

PD#TV-11316

[ Upstream commit 07b3975352374c3f5ebb4a42ef0b253fe370542d ]

Currently, if we drop a packet, we exit from NAPI loop before the budget
is consumed. In some situations this will make the RX processing stall
e.g. when flood pinging the system with oversized packets, as the
errorneous packets are not dropped efficiently.

If we drop a packet, we should just continue to the next one as long as
the budget allows.

Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Change-Id: Ib13615c2e413af86519c1cec0a6638b88ceaf4de
Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com>
Signed-off-by: zihuan.ling <zihuan.ling@amlogic.com>
5 years agodts: km3: Device ID attestation for ac214 [1/1]
Mingyen Hung [Thu, 31 Oct 2019 04:26:48 +0000 (12:26 +0800)]
dts: km3: Device ID attestation for ac214 [1/1]

PD#SWPL-12224

Problem:
Need to add support for device ID attestation

Solution:
1. Add new key slot, attestationdevidbox, to unify keys

Verify:
1. Android Q + Newton↵
2. run cts -m CtsKeystoreTestCases -t \
   android.keystore.cts.KeyAttestationTest#testDeviceIdAttestation

Change-Id: If16a35af4b63d19dcecf1319d6f0ea974ce64dd2
Signed-off-by: Mingyen Hung <mingyen.hung@amlogic.com>
5 years agoaudio: external tdm-loopback [1/1]
Shuai Li [Tue, 29 Oct 2019 06:27:20 +0000 (14:27 +0800)]
audio: external tdm-loopback [1/1]

PD#SH-1626

Problem:
Can not record ADC's i2s output from Loopback

Solution:
1. Fix the wrong reg config in TDM OE pin.
2. move the lr/sclk pad configuration to probe.

Verify:
SM1.

Change-Id: I01f419e4b0ba72fb7295641c6e7d9a189754c9d7
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
5 years agotm2: dts: clean lcd config [1/1]
Sandy Luo [Thu, 31 Oct 2019 06:43:16 +0000 (14:43 +0800)]
tm2: dts: clean lcd config [1/1]

PD#SWPL-15948

Problem:
T962e2 hasn't lcd out

Solution:
delete lcd config in dts

Verify:
T962e2 platform

Change-Id: Ibc090f948c9cba2073e17b17ccbf706d154445ef
Signed-off-by: Sandy Luo <sandy.luo@amlogic.com>
5 years agoaudio: auge: improve compatibility for eARC RX/TX connection [1/1]
Xing Wang [Wed, 23 Oct 2019 12:19:16 +0000 (20:19 +0800)]
audio: auge: improve compatibility for eARC RX/TX connection [1/1]

PD#SWPL-15189

Problem:
1. earc rx fails to connect to a sony tv
2. earc tx fails to connect to AVR-X4500H

Solution:
1. reset earc_rx pll
2. reduce comma_th

Verify:
ab311

Change-Id: I162b9697d151b682df01093cdc086330a97f8fab
Signed-off-by: Xing Wang <xing.wang@amlogic.com>
5 years agoavsync: Clear video pts & pcr pts when ts' playback stops [1/1]
hui an [Mon, 28 Oct 2019 04:38:53 +0000 (12:38 +0800)]
avsync: Clear video pts & pcr pts when ts' playback stops [1/1]

PD#TV-11272

Problem:
Switch to the ATV channel from DTV, the picture is stuck.

Solution:
Clear video pts & pcr pts.

Verify:
x30a4

Change-Id: Id265bfdfbd7e09d1423c9349de0596d4d742474c
Signed-off-by: hui an <hui.an@amlogic.com>
5 years agoavsync: tsync: fix avsync issues when dtv play [2/2]
Xiaoming Sui [Sat, 21 Sep 2019 09:36:28 +0000 (17:36 +0800)]
avsync: tsync: fix avsync issues when dtv play [2/2]

PD#TV-10253

Problem:
PD#TV-10253, SWPL-12816
dtv play when pcr is invalid, avsync is lost sometimes.

Solution:
add some specific marginal checks.

Verify:
X301

Change-Id: I2115bc934cd87d8a93c77180f4f37f1d23fa23ba
Signed-off-by: Xiaoming Sui <xiaoming.sui@amlogic.com>
5 years agoaudio: optimize TL1 codec init sequence [1/1]
Shuai Li [Thu, 19 Sep 2019 13:02:28 +0000 (21:02 +0800)]
audio: optimize TL1 codec init sequence [1/1]

PD#TV-9458

Problem:
Audio codec output has a ditch.

Solution:
Optimize TL1 codec init sequence [1/1]

Verify:
TL1.

Change-Id: Ib494895b96cb0083d0bdaa9ffa23c67bf57f7aa7
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
5 years agovdin: add RGB mode rgb full to yuv full [1/2]
Yong Qin [Tue, 15 Oct 2019 11:37:19 +0000 (19:37 +0800)]
vdin: add RGB mode rgb full to yuv full [1/2]

PD#SWPL-15366

Problem:
HDMI 256 gray will lose data info

Solution:
RGB full mode vdin matrix use RGB to YUV706F

Verify:
TL1

Change-Id: Idfcafefcc31cc2fc2a6a21368598eae8a5159ec4
Signed-off-by: Yong Qin <yong.qin@amlogic.com>
5 years agolcd: tl1: tcon chpi phy setting update [2/2]
shaochan.liu [Fri, 25 Oct 2019 05:24:26 +0000 (13:24 +0800)]
lcd: tl1: tcon chpi phy setting update [2/2]

PD#SWPL-13250

Problem:
tcon chpi phy setting update

Solution:
add tcon chpi phy setting update

Verify:
x301_t962x2

Change-Id: I84b6f4298f64697ee33376ec866e5ec34174227e
Signed-off-by: shaochan.liu <shaochan.liu@amlogic.com>
5 years agodeinterlace: pq: add save db value interface [1/2]
Jihong Sui [Sun, 29 Sep 2019 10:56:32 +0000 (18:56 +0800)]
deinterlace: pq: add save db value interface [1/2]

PD#SWPL-15705

Problem:
VLSI finetue TL1 1080i mtn setting.

Solution:
a. add di_gmv in vframe type;
b. add interface for save db setting and switch;
-api:void di_patch_mov_setreg(unsigned int nub, unsigned int *preg)
-for set register that you want to save from db;
-preg: register list that you want to save from db; < 5
-api:bool di_api_mov_sel(unsigned int mode, unsigned int *pdate)
-to set register that you save;
-mode:0:value from db;1: setting from pq
-*pdate: data from pq, value/mask
- pdate need keep same order from preg;
c. add interface for switch 1080i comb setting
-void di_set_comb_mode(unsigned int mode)
-mode:0: set nothing; 1:fix_tl1_1080i_sawtooth_patch;
2: new 1080i setting;

Verify:
tl1

Change-Id: I8e0f084d33757ab0e48fb49aececc020dae5c62d
Signed-off-by: Jihong Sui <jihong.sui@amlogic.com>
5 years agohdmitx: sync HDR/VSIF from A [1/1]
Zongdong Jiao [Fri, 25 Oct 2019 03:29:53 +0000 (11:29 +0800)]
hdmitx: sync HDR/VSIF from A [1/1]

PD#SWPL-13206

Problem:
Need to sync HDR/HLG/DV callback

Solution:
Sync HDR/HLG/DV callback

Verify:
G12/U212

Change-Id: Ibe93c3e538ad2f3cbb254c4ce731473b6ac1562b
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
5 years agoMTD: slc-nand: fix boot system failed due to pointer out-bounds operating [1/1]
xianjun.liu [Mon, 28 Oct 2019 08:05:39 +0000 (16:05 +0800)]
MTD: slc-nand: fix boot system failed due to pointer out-bounds operating [1/1]

PD#SWPL-15894

Problem:
The audio 113D platform + TC58NVG2S0HTA00 boot system failed due
to pointer out-bounds operating

Solution:
reduce oobavail value to avoid pointer access overbounds

Verify:
AXG-S400

Change-Id: I737d16b937b39b0b03c9248e6688c1158face3b5
Signed-off-by: xianjun.liu <xianjun.liu@amlogic.com>
5 years agodts: km4: Device ID attestation for ac214 [1/1]
Mingyen Hung [Thu, 31 Oct 2019 03:02:02 +0000 (11:02 +0800)]
dts: km4: Device ID attestation for ac214 [1/1]

PD#SWPL-12224

Problem:
Need to add support for device ID attestation

Solution:
1. Add new key slot, attestationdevidbox, to unify keys

Verify:
1. Android Q + Newton
2. run cts -m CtsKeystoreTestCases -t \
   android.keystore.cts.KeyAttestationTest#testDeviceIdAttestation

Change-Id: Ia07bcc6c05fc5376e2f029d5d4bc2781d7fa5618
Signed-off-by: Mingyen Hung <mingyen.hung@amlogic.com>
5 years agoHDR: HDR luminace tune to 316, 31 [1/1]
Shen Liu [Sat, 14 Sep 2019 02:14:29 +0000 (10:14 +0800)]
HDR: HDR luminace tune to 316, 31 [1/1]

PD#SWPL-13766

Problem:
HDR Luminance 285.5 exceeded recommendation range 300~400

Solution:
increase graphic target max luminuance for HDR output
DV -> HDR 316
HDR -> HDR 31,31...
value 316 and 31 are confirmed w/ Netflix

Verify:
Verified on u212

Change-Id: I1fbde36e1a15d31b6187ca1714f6e79ae1482301
Signed-off-by: Shen Liu <shen.liu@amlogic.com>
5 years agovideosync: add pattern support [1/1]
jintao xu [Wed, 23 Oct 2019 09:33:03 +0000 (17:33 +0800)]
videosync: add pattern support [1/1]

PD#SWPL-15586

Problem:
When hdmi is 59.94 and 30fps and 60fps video pts is not linearly increasing
video display in vd2 video will shake.

Solution:
Videosync add pattern support.

Verify:
U212

Change-Id: I47416681dffb6e5eddef39e9c1dfcd29cfc12b92
Signed-off-by: jintao xu <jintao.xu@amlogic.com>
5 years agohdmirx: add phy eq debug function [1/1]
yicheng shen [Thu, 10 Oct 2019 12:06:09 +0000 (20:06 +0800)]
hdmirx: add phy eq debug function [1/1]

PD#SWPL-12881

Problem:
Need to add eq debug function

Solution:
Add phy eq debug function,use auto-eq as default.

Verify:
TL1

Change-Id: I723d74d1bad226267b60ffb66d3670d01e6a4e1b
Signed-off-by: yicheng shen <yicheng.shen@amlogic.com>
5 years agomm: change lmk/cma using policy [1/1]
Tao Zeng [Wed, 16 Oct 2019 08:44:11 +0000 (16:44 +0800)]
mm: change lmk/cma using policy [1/1]

PD#TV-10462

Problem:
Memory will allocation fail if play secure vedio source. Usually
seen by zram/wifi driver

Solution:
1, wake up kswapd earlier if water mark without free cma is not ok;
2, using zone-filecache to increace active of lmk. Which can be more
accurate than using global page status;
3, remove some restrict of using cma when allocate movable page by
zram or migrate from cma pool;
4, try allocate hard for atomic request in soft IRQ

Verify:
T950L

Change-Id: Ibf03f3c11a32175e9983ee8a61a14ae4b2436f1e
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
5 years agodv: fix gxm dv bug [3/7]
yao liu [Fri, 11 Oct 2019 07:33:43 +0000 (15:33 +0800)]
dv: fix gxm dv bug [3/7]

PD#SWPL-13526

Problem:
1.watchdog reboot:
hdr_func->set_hdr_matrix is called
when disable dv, which leads to
access 38a0 illegally

2.no dolby_ctrl_backup and err
viu_misc_ctrl_backup lead to
green screen after disable dv

3.Switch resolution, display highlights
and some pixels are overexposed

4.First frame shift when playing video.

5.OSD jitter when play video

Solution:
1.Dont call hdr_vd1_off and hdr_osd_off for s912
2.dolby_ctrl_backup for s912 and update afbc bits
 for viu_misc_ctrl
3.Dont set core reg if regs no changed
4.It is hardware bug, need set run_mode_delay=3,
  that means the first three frames should be blank.
5.put osd_path_enable after enable_rgb_to_yuv_matrix_for_dvll

Verify:
Q201

Change-Id: I50440c5b3c6554c28f587c0542bea3773a2c9b9c
Signed-off-by: yao liu <yao.liu@amlogic.com>
5 years agodts: Add new dts files for T962E2 SBR [1/1]
bing.jiang [Fri, 18 Oct 2019 08:56:42 +0000 (16:56 +0800)]
dts: Add new dts files for T962E2 SBR [1/1]

PD#SWPL-15141

Problem:
    adapt Soundbar solution to T962E2 SBR platform

Solution:
    new dts file tm2_t962e2_ab311_sbr.dts changed from tm2_t962e2_ab311.dts
    tm2_t962e2_ab311.dts file version
        'commit 952ac4c1d709 ("hdmirx: pinmux later so as to
                               prevent abnormal hpd pulse [1/1]")'

    add amp tas5782m
    add amlogic-remote-4

Verify:
    T962E2

Change-Id: I5b4ad1bc526167970de2a85aadc4089ef5eeb233
Signed-off-by: bing.jiang <bing.jiang@amlogic.com>
5 years agombox: g12a: add m4 mbox driver [1/3]
Shunzhou Jiang [Fri, 7 Dec 2018 08:45:05 +0000 (16:45 +0800)]
mbox: g12a: add m4 mbox driver [1/3]

PD#SWPL-12873

Problem:
add m4 mailbox driver

Solution:
add m4 mailbox driver

Verify:
test pass on sm1 skt

Change-Id: Ied77949482179014d9e695ffc606f2ac939c82ca
Signed-off-by: Shunzhou Jiang <shunzhou.jiang@amlogic.com>
5 years agoemmc: optimize sdio tuning print [1/1]
Nan Li [Fri, 25 Oct 2019 11:20:48 +0000 (19:20 +0800)]
emmc: optimize sdio tuning print [1/1]

PD#SWPL-15797

Problem:
commit a5c3731f5d46f4104876b8c7ddf84f7a1b6ba071
this patch has err, the print buf will be out of bounds.

Solution:
Reposition the pointer before each loop.

Verify:
tl1

Change-Id: I8f44269c375147c5290af9c6d84fd5d85be271f0
Signed-off-by: Nan Li <nan.li@amlogic.com>
5 years agoosd: support 4K UI worked in 1080p output [1/1]
Pengcheng Chen [Wed, 23 Oct 2019 01:06:06 +0000 (09:06 +0800)]
osd: support 4K UI worked in 1080p output [1/1]

PD#SWPL-15649

Problem:
4K UI display abnormal with 1080p output

Solution:
adjust freescale to 2line mode when 4K UI with 1080p output

Verify:
g12a

Change-Id: I636b1d8634beabd3f356e63b9d7a9a05b1a556bb
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
5 years agodts: sync w400_a form W400 [1/1]
Gongwei Chen [Mon, 28 Oct 2019 03:17:40 +0000 (11:17 +0800)]
dts: sync w400_a form W400 [1/1]

PD#SWPL-15005

Problem:
sync w400_a form W400

Solution:
sync w400_a form W400

Verify:
Verify by Android P w400_a

Change-Id: I0ace7496f26bacac6343fd772ec2ab8c87721e76
Signed-off-by: GongWei Chen <gongwei.chen@amlogic.com>