PD#SWPL-5392
Problem:
set fixpll to 1G when cpufreq init.
Solution:
set fixpll to 1G when cpufreq init.
Verify:
test pass on A311D-W400
Change-Id: I65eb70b86eca5192f8d153ac59a24a03d6018421
Signed-off-by: Hong Guo <hong.guo@amlogic.com>
PD#SWPL-4003
Problem:
VD2 keep displaying after keep last frame for mvc streaming
Solution:
check the last frame type and disable vd2 again
Verify:
verify by tl1
Change-Id: Ia2b598331e8f65c02974d41c498408bee40a407e
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
PD#SWPL-2210
Problem:
multi wifi not support
Solution:
add multi wifi support
Verify:
ampere
Change-Id: Iea32268fd081a71a1ade00e56a2dcecd975da849
Signed-off-by: Rongjun Chen <rongjun.chen@amlogic.com>
PD#SWPL-86
Problem:
Android tensorflow lite ,NN API implementation
Solution:
NN API implementation
Verify:
Local on galilei
Change-Id: I699a53b3fa2fd199cea7ef4db2f439ef49536a9c
Signed-off-by: deng.liu <deng.liu@amlogic.com>
PD#SWPL-4718
Problem:
Will have "PC=0, LR=0" issue when try to
show active cpu stacks via sysrq
Solution:
Save the addr of current interrupted task's context to the
interrupt stack first, so that we can easily obtain pt_regs
later
Verify:
Locally on Ampere
Change-Id: I1cc5095e0ec356bed90f76cdf9af1b2617ce7834
Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com>
PD#SWPL-4718
Problem:
The task stack address is not saved to irq_stack_ptr-0x08,
so when switching from irq stack to task stack, we can not
obtain the right task stack address
Solution:
save task stack address to irq_stack_ptr-0x08
Verify:
Locally on Ampere
Change-Id: I8ad9a45cee33681cc00de3a0bb264dbd9f1b859f
Signed-off-by: Jiamin Ma <jiamin.ma@amlogic.com>
PD#SWPL-2681
Problem:
If cpu support idle/hotplug, there are many vmap stack prints:
[ 74.096973@1] cpu 1, vmap stack:[e8e90000-e8e91de0]
[ 74.097084@1] cpu 1, irq stack:[e8e92000-e8e93de0]
[ 74.105901@3] cpu 3, vmap stack:[e8ee0000-e8ee1de0]
[ 74.105984@3] cpu 3, irq stack:[e8ee2000-e8ee3de0]
Solution:
Close these debug print
Verify:
P212
Change-Id: I000df1f0952328af9f359f20341eb0fe88590056
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
PD#SWPL-2681
Problem:
Kernel stack usage is large when running many tasks.
Solution:
Map kernel stack to module space and handle page-fault
for stack fault. This can save about 50% memory of stack
usage
Verify:
p212
Change-Id: Ie894bc8f00cb525ddf8ac63c6d99d9c6e937fdc0
Signed-off-by: tao zeng <tao.zeng@amlogic.com>
PD#SWPL-4597
Problem:
Probabilistic full-screen display mosaic after seek 8K H265
Solution:
alloc mem area with the cached which from the vmap. this way
can improve the efficiency of I/O.
Change-Id: Icf18b9cd57bdc54f832aa827a535399939b28611
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
PD#SWPL-3593
Problem:
the memory data might be discordant.
Solution:
1. opitimize the mem mapping and change the page type to nocache.
2. modified the way of flush mem which from lowmem or highmem.
Verify:
x301
Change-Id: I82351c235915c98a86fd201c2ff3994e4d2085ec
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
PD#OTT-1699
Problem:
not support multi-vdec user data read
Solution:
Add getting available user data vdec interface
Verify:
Verified U212
Change-Id: I9ec65367935ac42a83808db23f2fb08f694e1483
Signed-off-by: Rong Zhang <rong.zhang@amlogic.com>
PD#SWPL-5356
Problem:
when cat clkmsr node in sysfs, i2c bus does not work
Solution:
delete hard code
Verify:
verify on axg s400
Change-Id: Ib01f251f4b77685db5a57df395f1c1a75620c07b
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
PD#SWPL-5478
Problem:
system will crash rately if not init success
Solution:
remote internal_phy_remove function for chip
later then g12a
Verify:
verify on u200 board
Change-Id: Id6c3bf77c2a448aee9675deff47dc106206146d6
Signed-off-by: Zhuo Wang <zhuo.wang@amlogic.com>
PD#SWPL-4406
Problem:
SDR meta size > 0 during DoVi to SDR transition
Solution:
fixed SDR meta size to 0 for SDR
Verify:
u212
Change-Id: I92129163da0df9c81bb508ce56b78268fbd87a2e
Signed-off-by: Yi Zhou <yi.zhou@amlogic.com>
PD#172483
Problem:
To avoid A73 core speculative data loading from protected
memory, which triggers SError fault, the mapping of the
protected region should be cleared. In codec_mm driver, the
page structure pointer used to clear the mapping is wrong.
Although the memory is allocated from CMA, because we are
using CMA_res mode the address from CMA_res_pool is actually
a physical address, similar as reserved memory case. As a
result, the mapping to protected address is not cleared.
Solution:
1.Add "mem->from_flags ==
AMPORTS_MEM_FLAGS_FROM_GET_FROM_CMA_RES" case processing.
Verify:
Run YTTV and there is no SError crash after the fix.
Change-Id: Ia912337c1d7d5a55800fc55ad1d61016f9710f4d
Signed-off-by: Kai Song <kai.song@amlogic.com>
PD#SWPL-1738
Problem:
support frame crc32 check and dump yuv.
Solution:
support frame check and yuv dump in media_module, add interface
to get canvas width/height.
Verify:
u212
Change-Id: Ieadc95521c4a21bec5685f8c8926d3bb2f92637b
Signed-off-by: shihong.zheng <shihong.zheng@amlogic.com>
PD#SWPL-1885
Problem:
We need a function to statistics drop frame.
Solution:
Add a function to statistics drop frame.
Verify:
Verified U212
Change-Id: I30b429c2f3710c5c8d0da6ec083060c9cc3e61ad
Signed-off-by: Peng Yixin <yixin.peng@amlogic.com>
PD#SWPL-4932
Problem:
The vd/osd blend ctrl is not correct when disable video
Solution:
Refresh the vd/osd blend ctrl for that case
Verify:
Verified on tl1
Change-Id: I6a7bfc58c810279cdfbf55f3d6a51e94890b4899
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
PD#SWPL-2367
Problem:
building err and issue of compatibility.
Solution:
1. fixed building err without merging media_modules
2. add a new api tee_load_video_fw_swap
Verify:
p212,u212
Change-Id: Ifeedf2f3fa04b2427dccf2ac3940bad6860362d6
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
PD#SWPL-2367
Problem:
can not playback hevc after update video fw.
Solution:
add swapping of the fw for the tee loading.
Verify:
p212,u212
Change-Id: I698e43ba8e71060f85be4b468c4c61d625f46d4a
Signed-off-by: Nanxin Qin <nanxin.qin@amlogic.com>
PD#SWPL-5148
Problem:
bootvideo stuck when starting to play
Solution:
wait hw vsync for afbc when phy_addr==0 only before chip g12A
Verify:
tl1
Change-Id: I883c8f6699729854a852fc70c22b5629aa025c60
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
PD#SWPL-4900
Problem:
printk_prefix() passes NULL buf to print_time() to get the length of
the time prefix; when printk times are enable, the current code just
return "snprint(NULL, 0, "[%5lu.000000]", (unsigned long)ts)"
but in the SMP case, we add cpu core id at the end of time prefix, the
format is "[%5lu, %06lu@%d]", this gives a wrong size(shorter).
this will make buf get out of boundary, when run dmesg.
Solution:
replace [%5lu.000000] with [%5lu.000000@0]
Verify:
g12a-u200
Change-Id: Ic74b9f466b5af5aac7e6569a1afc7a46d9594050
Signed-off-by: Jianxiong Pan <jianxiong.pan@amlogic.com>
PD#SWPL-3744
Problem:
For the previous Amlogic CPU series(before TXLX), the CPU booting vector
locates at physical address starting from 0x00. To avoid influencing cpu
booting procedure, we have to reserve the first 1MB physical memory. But
it has changed since TXLX, the normal text section can safely use 0~1M
physical address. So it's no need to reserve the first 1MB in DTS.
Solution:
Change the starting address of usable memory from 0x100000 to 0x0 for CPU
series coming after TXLX.
Verify:
Locally on W400
Change-Id: I2df7859a38feca48abd58459e8b231ee15ada9e2
Signed-off-by: changqing.gao <changqing.gao@amlogic.com>
Signed-off-by: Luan Yuan <luan.yuan@amlogic.com>
Conflicts:
arch/arm/boot/dts/amlogic/txlx_t962x_r311_1g.dts
arch/arm64/boot/dts/amlogic/tl1_t962x2_t309.dts
arch/arm64/boot/dts/amlogic/tl1_t962x2_x301.dts
PD#SWPL-4705
Problem:
In 61207 patch, the define HDMI_IEEE_OUI is conflicted with the kernel
head file include/linux/hdmi.h
Solution:
rename HDMI_IEEE_OUI
Verify:
GXL/P212
Change-Id: I75a12734e85478f22edf0b48636ed86e60302b58
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
PD#SWPL-4705
Problem:
Lack ALLM function
Solution:
Add ALLM function
Verify:
GXL/P212
If Rx supports ALLM, then
echo 1 > /sys/class/amhdmitx/amhdmitx0/allm_mode
otherwise it will set failed, cat allm_mode and will get 0.
Change-Id: I00233e5a5aac133b405590e7df78c7c4805ed0ef
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
PD#SWPL-2456
Problem:
Lack the latency info of RX
Solution:
Add the latency info of RX
Verify:
T962E/R321
Change-Id: I480398466753dd93ca6e908d39157acadff6879f
Signed-off-by: Zongdong Jiao <zongdong.jiao@amlogic.com>
PD#TV-1565
Problem:
After g12a chip, vpp post dummy reg was changed. The
test_screen/rgb_screen sysfs did not work.
Solution:
Use the correct vpp post dummy reg for those chip
Verify:
Verified on X301 and W400
Change-Id: I1cb718a1f7040804b63d0197de0bb6aafe233357
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
PD#SWPL-3907
Problem:
PDM controls are dynamically added or removed.
This will bring in panic.
Solution:
Remove the unused controls.
Verify:
Need stress test.
Change-Id: I69f63a0516e0812e9abb86e42f091ecb496004d4
Signed-off-by: Shuai Li <shuai.li@amlogic.com>
PD#SWPL-4673
Problem:
userspace can't access perf monitor cnt with "mrc ... c9,c13,0"
Solution:
actually userspace should check perf monitor cnt access permissions first
before use, so this is a workround.
enable pmuserenr with "mcr ... c9,c14,0" in several places:
1, perf probe
2, cpu idle (state>0) exit
3, cpu online
4, system resume after suspend
Verify:
u212
Change-Id: Ib09682d1d47545dfef8b088283ddbbf390630d3e
Signed-off-by: Hanjie Lin <hanjie.lin@amlogic.com>
PD#TV-2331
Problem:
SR insert unnecessary scale up to make scale down ratio
too larger.
Solution:
disable SR core0 which is before pps when mif vskip is > 2
Verify:
Locally on X301
Change-Id: I810fbae20f337e6eeb0a8e88d9758f78ad224254
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
PD#SWPL-4617
Problem:
CONFIG_CC_STACKPROTECTOR_STRONG will increase stack size for some large
functions and may cause stack overflow problems or increase real stack
usage after we have enabled VMAP STACK config. But direct remove it
from defconfig will cause CTS fail.
Solution:
using CONFIG_CC_STACKPROTECTOR_STRONG_AMLOGIC for real stack protect
config
Verify:
p212
Change-Id: I1ccba2ef6ab5ea6f2987af2986e0cf222da1a7c7
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
PD#SWPL-4857
Problem:
fixdiv calc Affect the use of other chips
Solution:
Adjust the usage conditions for the fixdiv calc func, witch
pdata->calc_f from dts.
Verify:
verify by g12b
Change-Id: I9598e2a24f76c76f4312c2694029fe4bb0f534d4
Signed-off-by: Nan Li <nan.li@amlogic.com>
PD#SWPL-5113
Problem:
SR core0 enable switch register is latched as default. It
will cause the screen flicker when operating this bit in vsync.
Because the frame size will be out of sync with back-end module.
Solution:
1. For g12a, no latch ctrl. So did not disable sr core2 enable bit.
2. For g12b/tl1, disable the latch function.
Verify:
Verified on U212/w400/x301
Change-Id: I54027b71ef8a6066004b3bd32ed1633b4bfa351c
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
PD#SWPL-5020
Problem:
need 1.9G for A53 core in g12b RevB.
Solution:
add 1.9G frequency for A53 core in RevB dts.
Verify:
local test.
Change-Id: Ic4085c05b3b9d18f759720e3d268d5e4629faba7
Signed-off-by: pan.yang <pan.yang@amlogic.com>
