View Issue Details

IDProjectCategoryView StatusLast Update
0001531T99X171.00 SKB EagleSW Issuepublic2022-11-04 13:36
Reporter(ALTech) Younkwang Jung Assigned To(ALTech) Younkwang Jung Due Date
PriorityhighSeveritys3-moderateReproducibilityhave not tried
Status closedResolutionfixed 
Summary0001531: [Smart3][VoC] Ramoops issue (LR is at hdmitx_edid_check_valid_mode+0x40/0x338)
DescriptionHi Kerwin

About 880k of Smart3 v15.534.18 version was distributed.

The problem is that the following error was collected from the log collection system. ( 7 second after booting )
============================== dmesg-ramoops-0 ====================================
<4>[ 7.451302@2] task: 00000000903221c6 task.stack: 0000000045ebbc44
<4>[ 7.451312@2] PC is at strcmp+0x18/0x160
<4>[ 7.451322@2] LR is at hdmitx_edid_check_valid_mode+0x40/0x338
<6>[ 7.451327@2] R3 : ffffff8009117b80, PFN: 1117
<6>[ 7.451333@2] R14 : ffffff800ac0e8b8, PFN: 2c0e
..
<4>[ 7.453450@2] fb00: 0000000000000001 0000000000000000 ffffff800ac0e8b8 ffffffffffffffff
<4>[ 7.453456@2] fb20: ffffff8009239240 0000000000000000
<4>[ 7.453466@2] [<000000001f7fe90e>] strcmp+0x18/0x160
<4>[ 7.453474@2] [<00000000ab09efa7>] show_valid_mode+0xcc/0x158
<4>[ 7.453481@2] [<000000003b039058>] dev_attr_show+0x3c/0x78
<4>[ 7.453489@2] [<000000004e2d8b10>] sysfs_kf_seq_show+0xc0/0x158
<4>[ 7.453495@2] [<000000008b8cca21>] kernfs_seq_show+0x44/0x50
<4>[ 7.453502@2] [<0000000025111e5a>] seq_read+0x1ac/0x440
<4>[ 7.453508@2] [<00000000e5414c19>] kernfs_fop_read+0x154/0x1d8
<4>[ 7.453517@2] [<00000000bd5f7e5a>] __vfs_read+0x48/0x120
<4>[ 7.453523@2] [<0000000062f7452b>] vfs_read+0x94/0x150
<4>[ 7.453529@2] [<0000000049a76134>] SyS_read+0x54/0xb0
<4>[ 7.453536@2] [<00000000f90bd6c4>] el0_svc_naked+0x34/0x38
<0>[ 7.453544@2] Code: f24008ff 540002e1 f2400807 54
=============================================================================

And I updated the LR code location of the ramoops in jira below.
https://jira.skbroadband.com/browse/BTFAML-1231

In common, this issue occurred as soon as it was booted (within 1-2 minutes).
but this issue is not observed in INTEK or Innopia.

Please check this issue

Thank you.
YK.Jung
TagsNo tags attached.
Attach Tags

Users monitoring this issue

User List (ALTech) JunGyu Kim , (ALTech) SY Yoon

Activities

(ALTech) Younkwang Jung

2022-08-30 17:34

developer  

(SW) Kerwin Chen

2022-08-31 13:17

developer   ~0010756

Hi YK,

1. It looks there is a potential bug in AML HDMI driver.
You can try to reproduce by command below:
echo "640x480p60hz" > /sys/class/amhdmitxamhdmitx0/valid_mode
cat /sys/class/amhdmitxamhdmitx0/valid_mode

The root cause is some hdmi_format_para doesn't set 'sname' string in hdmi_parameters.c. (VESA formats)
So, para->sname will be zero.
I think AML should review the driver codes.

2. It happens on Foxconn Smart3 because we use "valid_mode" to check HDMI format while generating hdmi_edid.xml.

Thank you !

(ALTech) Younkwang Jung

2022-09-05 13:17

developer   ~0010785

Hi Kerwin

This issue only occurs on the FXN.
Amlogic suspects code modified by FXN.
https://jira.skbroadband.com/browse/BTFAML-1231

Please remove the modified part from FXN and check if the issue occurs.
or, please check it in the amlogic reference sdk.

Thank you!!
YK.Jung

(SW) Kerwin Chen

2022-09-05 15:40

developer   ~0010789

Hi YK,

Foxconn don't modify "hdmi_parameters.c" file. (drivers/amlogic/media/vout/hdmitx/hdmi_common/hdmi_parameters.c)
BTW, we have an Intek's Smart3 and the result is the same.
System reboot after command below:
echo "640x480p60hz" > /sys/class/amhdmitxamhdmitx0/valid_mode
cat /sys/class/amhdmitxamhdmitx0/valid_mode

The support for VESA mode was introduced by patch:
https://bitbucket.skbroadband.com/projects/AMANDROIDQ/repos/kernel-common/commits/c7d8f2df92f65ed467e1f1a6baf97e034f35bf58

Thank you !

(ALTech) Younkwang Jung

2022-11-04 13:36

developer   ~0011163

HI Kerwin

I received the patch from AML and there was no problem when I tested it.
I will close this issue.

Thank you
YK.Jung

Issue History

Date Modified Username Field Change
2022-08-30 17:34 (ALTech) Younkwang Jung New Issue
2022-08-30 17:34 (ALTech) Younkwang Jung Status new => assigned
2022-08-30 17:34 (ALTech) Younkwang Jung Assigned To => (SW) Kerwin Chen
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_2.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_3.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_4.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_7.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_5.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_6.zip.zip
2022-08-30 17:34 (ALTech) Younkwang Jung File Added: output__BFX_AT100_534.18_1.zip.zip
2022-08-31 08:27 (ALTech) Younkwang Jung Issue Monitored: (ALTech) SY Yoon
2022-08-31 08:27 (ALTech) Younkwang Jung Issue Monitored: (ALTech) JunGyu Kim
2022-08-31 13:17 (SW) Kerwin Chen Note Added: 0010756
2022-08-31 13:17 (SW) Kerwin Chen Assigned To (SW) Kerwin Chen => (ALTech) Younkwang Jung
2022-08-31 13:17 (SW) Kerwin Chen Status assigned => acknowledged
2022-09-01 10:07 (SW) Kerwin Chen Severity s2-severe => s3-moderate
2022-09-05 13:17 (ALTech) Younkwang Jung Note Added: 0010785
2022-09-05 15:40 (SW) Kerwin Chen Note Added: 0010789
2022-11-04 13:36 (ALTech) Younkwang Jung Note Added: 0011163
2022-11-04 13:36 (ALTech) Younkwang Jung Status acknowledged => closed
2022-11-04 13:36 (ALTech) Younkwang Jung Resolution open => fixed