PD#SWPL-141905
Problem:
kernel 5.15 have no wifi comm API
Solution:
kernel 5.15 add wifi comm API
Verify:
ohm
Change-Id: I14b3443d7ce7894c019cd76f0891f157f15f1f5d
Signed-off-by: jiabin.chen <jiabin.chen@amlogic.com>
PD#SWPL-136709
Problem:
pre_vsync rdma not work
Solution:
1. fix pre vsync rdma not work
2. add pre_vsync video process
Verify:
t3x_revb
Change-Id: Idea48258140934374b2be6e951d7128f3081604f
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
PD#SWPL-139554
PD#SWPL-138558
Problem:
colorspace,colordepth,hdr_priority were controlled by sysfs node
hdr_priority definition was changed
Solution:
change colorspace,colordepth,hdr_priority control method to drm
add new hdr_priority definition to support multi strategy
update the RX hdr/hdr10+/dv capabilities in the end of EDID parse
Verify:
ohm
Test:
DRM-TX-40
Change-Id: Id9989d542ac780e358263ace67e7f13370498285
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
Signed-off-by: zongdong.jiao <zongdong.jiao@amlogic.com>
PD#SWPL-104009
Problem:
drm hdmitx call hook interface
Solution:
use common api to replace hook interface
Verify:
ohm
Test:
DRM-TX-37
Change-Id: Ia29c7f9d6aea9b358053460cdbc6ab99738c984e
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
PD#SWPL-104023
Problem:
1.hdmitx_event flag can be set/cleared in
both top/bottom half of hpd interrupt, so
there may be write concurrency of this flag,
subsequent plug in event may be ignored
if the first plugin handler clear the
this flag of the second plugin event
2.hpd change event during bootup may be
lost to process, as there's opportunity
there's hpd event come between hpd level
read/handler and hpd irq enable
Solution:
optimise hpd flow
1.use hw hpd state to indicate the hpd
intr state and only change in top half
of hpd irq handler. use sw hpd flag to
indicate if plugout/in bottom half done
2.lock mutex-> enable hpd irq-> do hpd
top half handler-> do hpd bottom half
handler->unlock mutex, so that all hpd
irq will be handled.
Verify:
ohm
Test:
hotplug
Change-Id: I5347a37f022dd37c5a5a436101af440aa49dd662
Signed-off-by: hang cheng <hang.cheng@amlogic.com>
PD#SWPL-137140
Problem:
cur_VIC is not needed
hdmitx20 and hdmitx21 has the same vout api
Solution:
remove cur_VIC and move vout server to hdmitx common
remove pr_fmt
Verify:
ohm
Test:
DRM-TX-33
Change-Id: I981a7769fe1f9f6f562f1814cf19cd5f57535fdc
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
Signed-off-by: zongdong.jiao <zongdong.jiao@amlogic.com>
PD#SWPL-104015
Problem:
tx20/tx21 have similar functions for
sysfs and hwop.
Solution:
move to hdmitx_hw_common and hdmitx_sysfs_common
Verify:
ohm/t7
Test:
DRM-TX-36
Change-Id: I0638bfe263c07589716da467f251a279582bcda6
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
PD#SWPL-138821
Problem:
need to sync hdmitx20 with kernel5.4
Solution:
sync hdmitx20 with kernel5.4
Verify:
verify on ah212
Test:
DRM-TX-31
Change-Id: I667ff3df6d58552feb8972dc5c76b764713f61ac
Signed-off-by: hang cheng <hang.cheng@amlogic.com>
PD#SWPL-98813
Problem:
Need to unify the edid 2.0 and 2.1
Solution:
Unify the edid 2.0 and 2.1
Verify:
t7/sc2
Test:
DRM-TX-34
Change-Id: Id145c0343279489b82316f22322d48a41c2df33c
Signed-off-by: zongdong.jiao <zongdong.jiao@amlogic.com>
PD#SWPL-104011
Problem:
hdev has some hw related members and
hdev is still be used in hdmitx_hw level.
Solution:
abstract tx20_hw/tx21_hw
Verify:
verify on ohm/t7
Test:
DRM-TX-33
Change-Id: Ic0e3867af7894fcdcd50a3a4fe234156e2ef821d
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
PD#SWPL-139231
Problem:
meson_drm.h in 5.15 and 5.4 have discrepancies
Solution:
eliminate the difference between 5.15 and 5.4
Verify:
sc2
Test:
DRM-OSD-103
Change-Id: I806cf16490945bc3f34c1504b95a597c34d0b6b8
Signed-off-by: mingyang.he <mingyang.he@amlogic.com>
Signed-off-by: Luan Yuan <luan.yuan@amlogic.com>
PD#SWPL-130171
Problem:
add use tsync flag for 5.15
Solution:
add use tsync flag for 5.15
Verify:
t5d
Change-Id: Ia2a4cc742a9213c5b40b16ce5a46502f9a857374
Signed-off-by: dehong.chen <dehong.chen@amlogic.com>
PD#SWPL-139231
Problem:
meson_drm.h in 5.15 and 5.4 have discrepancies
Solution:
eliminate the difference between 5.15 and 5.4
Verify:
sc2
Test:
DRM-OSD-103
Change-Id: I2f7432dcafc0976fbc67dbeed684280b17953284
Signed-off-by: mingyang.he <mingyang.he@amlogic.com>
PD#SWPL-138534
Problem:
tx20/tx21 have different valid mode flow and api
Solution:
use common api instead
Verify:
verify on T7
Test:
DRM-TX-19, DRM-TX-20,DRM-TX-22, DRM-TX-25, DRM-TX-27
Change-Id: I008e5931adb223391e79149f908d5ad1cb186889
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
PD#SWPL-126494
Problem:
hdmi_format_para is the struct for hdmi setting,
but no api defined.
Solution:
abstract format_para api.
Verify:
verify on ohm
Test:
DRM-TX-19, DRM-TX-20,DRM-TX-22, DRM-TX-25, DRM-TX-27
Change-Id: Id14fd0186254963361f0df3d869aceab6704377b
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
PD#SWPL-140403
Problem:
Missing SM4 support in aml-crypto-device on S5
Solution:
Add SM4 support to aml-crypto-device
Verify:
S5
Change-Id: I0933bbcefa46daa5b00bd41d1e68fc9441bd3185
Signed-off-by: Matthew Shyu <matthew.shyu@amlogic.com>
PD#SWPL-132959
Problem:
mode validation is in many functions and
checked many times.
Solution:
abstract the mode validate api and cleanup
the flow in tx20.
Verify:
verify on ohm.
Test:
DRM-TX-27, DRM-TX-26,DRM-TX-28,DRM-TX-25,DRM-TX-20
Change-Id: I9ddd17718fa43ac7a670192ba72a48d9bc0a6dc5
Signed-off-by: sky zhou <sky.zhou@amlogic.com>
PD#SWPL-115660
Problem:
not enough time for avmute when suspend
Solution:
add delay for avmute
Verify:
T7
Change-Id: I105e2485bea918b8350fec16369b243de9782b75
Signed-off-by: xiang.wu1 <xiang.wu1@amlogic.com>
PD#SWPL-137475
Problem:
process_checkin find ref has released
Solution:
add mutex for process and release check in
Verify:
SC2
Change-Id: I98af46e1488d534503b9984fc7866b9523522a49
Signed-off-by: jintao xu <jintao.xu@amlogic.com>
PD#SWPL-132145
Problem:
for "invalid" or "null" mode, need reset format para
Solution:
add reset format para api for hdmitx21
Verify:
t7c-an400
Test:
DRM-TX-28
Change-Id: I12c77bb463048a0bd71aa309f1dbae118f196917
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
PD#SWPL-132145
Problem:
tx20 timing para has been moved to common timing
tx21 also need use the common timing
Solution:
tx21 use common timing api
Verify:
t7c
Change-Id: I6397eafc98aa8be1e86e5f3db7ccabd24c448554
Signed-off-by: Ao Xu <ao.xu@amlogic.com>
PD#SWPL-131211
Problem:
kernel crash in check_page_to_cma
Solution:
The problem is that page_mapping function returns a invalid mapping
pointer, when a compound page is building while a compaction is doing
at same time, there will have a risk of compound page status checking.
page->mapping is set to TAIL_MAPPING first then set compound header.
But check_page_to_cma don't check real status of page, which caused
panic
Verify:
local
Change-Id: Id44ac3d43d1385148e0527b0ba4d3421c68d825e
Signed-off-by: Tao Zeng <tao.zeng@amlogic.com>
PD#SWPL-137107
Problem:
The data in the ramdump crash file is abnormal
because the cache data has not been flushed back.
Solution:
1. Use ipi_stop hook to flush online cpu cache.
2. Use panic_notifier to flush panic cpu cache.
Verify:
sc2_ah212
Change-Id: Ifad5bd4203fdde327b8c2b19357aa2d92586063f
Signed-off-by: dongqing.li <dongqing.li@amlogic.com>
PD#SWPL-132066
Problem:
need to support video lowlatency
Solution:
video lowlatency optimize
Verify:
s1a-bg201
Signed-off-by: Jian Cao <jian.cao@amlogic.com>
Change-Id: Ie8a6f4c72314f9acf7202e7d86721403842d5499
PD#SWPL-79478
Problem:
old DVB-C blind scan is too slow
Solution:
1.new dvbc blind scan feature time usage about 11min30s,
the speed has increased by about 65%.
2.AMLDTVDEMOD_VER V3.5.4
Verify:
s905c3 t214/ap232
Change-Id: I9197ed12844dff36ccb0a79488899f8872e04fed
Signed-off-by: zhicheng.huang <zhicheng.huang@amlogic.com>
PD#SWPL-132411
Problem:
video buffer put back when in lowlatency proc
Solution:
only put video buffer in vsync_isr
Verify:
ohm
Change-Id: If78ef47a58fa93d44303d49c84c4079a1b1e321c
Signed-off-by: Pengcheng Chen <pengcheng.chen@amlogic.com>
PD#SWPL-132354
PD#OTT-43684
Problem:
when the audio hw sample freq changes, the audio sample packets
still keep send out. Compared with 3rd devices, there will stop
the sample packets, only keep the ACR send out.
Solution:
Optimize the audio setting flow
disable ACR in PREPARE
finial enable ACR
sync the aud_src_if and aud_output_ch variables
when prepare, add the temporary CT_PREPARE
Verify:
sc2
Change-Id: Ic3a589fd16124e3036d93b766d84f600bf9cb811
Signed-off-by: zongdong.jiao <zongdong.jiao@amlogic.com>
PD#SWPL-136272
Problem:
1.txhd2 need support lcd ss
2.ss support need update
Solution:
update clk ss support
Verify:
be311, ar301
Change-Id: Ia1462905734f79b6dad9bea8c25681dd596fa1a1
Signed-off-by: Evoke Zhang <evoke.zhang@amlogic.com>
PD#SWPL-136140
Problem:
need screencap for only video of point 7.
Solution:
get width and height for only video of point 7.
Verify:
TXHD2
Change-Id: I534baa5201191cc4324d8f1c914df39c548ccf40
Signed-off-by: chen.xu <chen.xu@amlogic.com>
PD#SWPL-136132
Problem:
provide api to get input size
Solution:
provide api for it
Verify:
TXHD2
Change-Id: Ifcf7f18ce6b806cc8f68cf985172e46bc2625f76
Signed-off-by: hai.cao <hai.cao@amlogic.com>
PD#SWPL-119524
Problem:
Due to the post-positioning of di,
the vb2 buffer is directly thrown from the decoder to the upper layer.
Since each vb2 buffer can only be returned by the upper layer,
it can be thrown to the upper layer again. For interlaced streams,
two fields need to be output, so the second field cannot be output.
Solution:
The capture dma buffer is allocated in the decoder.
uvm only allocates an empty shell,
which is actually allocated by the decoder and replaces the dma buffer.
Verify:
ohm
Change-Id: Ic93fc379ad35d873ed3ef2bb0c5312254a08dee5
Signed-off-by: miaohong chen <miaohong.chen@amlogic.com>
PD#SWPL-128700
Problem:
When prelink and dct, do not support crop function
Solution:
1. add alignment for dct and prelink case
2. When grid crop did not match for dct post, bypass dct
3. add crop function in grid and dct post
4. when afbc and crop source, adjust nr write afbce and mem afbcd scope
Verify:
verified on t5m
Change-Id: Id4cb63bb91e168f540dce22a3834ba70b052f128
Signed-off-by: Brian Zhu <brian.zhu@amlogic.com>
jiabin.chen