imx8mp win10iot strange patterns on display

[jiangyaqiang]

HI :

I found an issue on imx8mp win10iot(version1.3.1): the display sometimes show mess thing.

But I found a fix in version 1.5.0, it says: " Occasional freezes or strange patterns on display i.MX 8M fixed by the DDR timing fix in the U-Boot component based on HW errata."

Could someone tell me what exactly need to modify for the "fix" ? Because I need to use my own lpddr_timing.c , and the fix seems need to modify lpddr_timing.c.

Thanks .

[michal_hromadka]

Hello @jiangyaqiang ,

The fix you mentioned above is not for the i.MX8M Plus, but for SoC i.MX8M (Quad). So the cause of your issue may be other than DDR timing.
Are you able to describe your issue in more detail? What does the screen look like or how often does it happen?

Thank you.

Best Regards,

Michal

[jiangyaqiang]

Hi michal_hromadka:

I attached a video shows the display on the screen. Please take a look at it.

Thanks.

[michal_hromadka]

Hi @jiangyaqiang ,

Thank you for the video.

Please, what version of BSP is used in the video? 1.5.0 or older ? 

Best Regards, 

Michal

[jiangyaqiang]

Hi  michal_hromadka:

My bsp version is: W21H2-1-4-0

So I want to know how version-1-5 fix "strange patterns on display".

Thanks.

[michal_hromadka]

Hello @jiangyaqiang,

I've tried to scroll the same page as you on the i.MX8MP EVK board with BSP v1.5.0 and no patterns were visible.

Regarding your question: Every release of Win IoT BSP contains some updates or improvements to the GPU driver. For testing, you can try using a driver from a newer version and update your older version.

Best Regards,

Michal

[jiangyaqiang]

Hi michal_hromadka:

I'm trying to upgrade the version to 1.5.0, but it fail to bootup, it shows up the first window picture as bellow and stop.

After research , when I use uefi.fit from 1.4.0, it can bootup success (only change uefi.fit is enough, anther source are from 1.5.0).  

And there is WARNING: " WARNING: Test Key is used" 

My board's DDR is 3GB only.

Any suggestion ?

Thanks.

pasted the log:

U-Boot SPL 2022.04-00001-g43b55151f1c-dirty (Aug 27 2024 - 13:56:05 +0800)
write error to device: 95f0f0 register: x!write error to device: 95f0f0 register: x!write error to device: 95f0f0 register: x!write error to device: 95f0f0 register: x!write error to device: 95f0f0 register: x!write error to device: 95f0f0 register: x!DDRINFO: start DRAM init
DDRINFO: DRAM rate 3600MTS<Enter>=Select Entry
DDRINFO:ddrphy calibration done
DDRINFO: ddrmix config done
SEC0: RNG instantiated
Normal Boot
Trying to boot from BOOTROM
Boot Stage: Primary boot
image offset 0x0, pagesize 0x200, ivt offset 0x0
NOTICE: BL31: v2.6(release):automotive-13.0.0_1.1.0-1-g2e577a2c0
NOTICE: BL31: Built : 03:31:56, Aug 21 2024

U-Boot 2022.04-00001-g43b55151f1c-dirty (Aug 27 2024 - 13:56:05 +0800)

CPU: i.MX8MP[8] rev1.1 1600 MHz (running at 1200 MHz)
CPU: Industrial temperature grade (-40C to 105C) at 36C
Reset cause: POR
Model: NXP i.MX8MPlus LPDDR4 EVK board
DRAM: ssssssssss size:c0000000
ssssssssss size:c0000000
ssssssssss size:c0000000
3 GiB
Core: 93 devices, 29 uclasses, devicetree: separate
MMC: FSL_SDHC: 1, FSL_SDHC: 2
Loading Environment from MMC... *** Warning - bad CRC, using default environment

[*]-Video Link 0adv7535_mipi2hdmi adv7535@3d: Can't find cec device id=0x3c
fail to probe panel device adv7535@3d
fail to get display timings
probe video device failed, ret -19

[0] lcd-controller@32e80000, video
[1] mipi_dsi@32e60000, video_bridge
[2] adv7535@3d, panel
adv7535_mipi2hdmi adv7535@3d: Can't find cec device id=0x3c
fail to probe panel device adv7535@3d
fail to get display timings
probe video device failed, ret -19
In: serial
Out: serial
Err: serial
SEC0: RNG instantiated

BuildInfo:
- ATF 2e577a2

switch to partitions #0, OK
mmc2(part 0) is current device
flash target is MMC:2
Net: Could not get PHY for FEC0: addr 1
Could not get PHY for FEC0: addr 1
eth1: ethernet@30bf0000 [PRIME]
Fastboot: Normal
Normal Boot
starting USB...
Bus usb@38100000: Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.10
Bus usb@38200000: Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.10
scanning bus usb@38100000 for devices... 1 USB Device(s) found
scanning bus usb@38200000 for devices... 1 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
## Loading kernel from FIT Image at 40480000 ...
Using 'conf-1' configuration
Verifying Hash Integrity ... OK
Trying 'uefi' kernel subimage
Description: UEFI
Created: 2024-08-26 7:11:12 UTC
Type: Kernel Image
Compression: gzip compressed
Data Start: 0x404800a4
Data Size: 852761 Bytes = 832.8 KiB
Architecture: AArch64
OS: Linux
Load Address: 0x50004000
Entry Point: 0x50004000
Hash algo: sha256
Hash value: bd58faaccbcba0d7575d80ff18b18c050eb76e703b989b0a62c9fe53390bea33
Verifying Hash Integrity ... sha256+ OK
Uncompressing Kernel Image
adv7535_mipi2hdmi adv7535@3d: Can't find cec device id=0x3c
fail to probe panel device adv7535@3d
fail to get display timings
probe video device failed, ret -19

Starting kernel ...

UEFI Debug Serial Port Init
UEFI firmware (version 2023-10-21 built at 15:09:41 on Aug 26 2024)
iMX8RTC Initialized with sample time 1.7.2023 9:00:00 UTC, DST.

Tianocore/EDK2 firmware version 2023-10-21
Press ESCAPE for boot options ** WARNING: Test Key is used. **

[jiangyaqiang · Accepted solution]

Hi:

upgrade to version1.5.0 seems no display pattern issue.

And the bootup issue can be fixed by:

mu_platform_nxp/NXP/MX8M_PLUS_EVK/MX8M_PLUS_EVK.dsc 
change : giMX8TokenSpaceGuid.PcdBank1MemorySize|0x00000000C0000000 to giMX8TokenSpaceGuid.PcdBank1MemorySize|0x000000000000000

Thanks.

[michal_hromadka]

Hi @jiangyaqiang 

I was just finishing up some steps to update just the GPU driver, but if you were able to run the whole BSP v1.5.0, that's better.

Best Regards,

Michal

[jiangyaqiang]

HI michal:

Got it.

Thanks.

[Bio_TICFSL]

Hello,

This has to be with the errata:

ERR050711 DDR: Activate Commands blocked in Per-Bank Refresh Mode for LPDDR4.
ERR050712 DDR: Register corruption possible when software triggered mode register (MR) operations performed in DDR4 mode.

Regards

[jiangyaqiang]

HI Bio_TICFSL:

Thanks for reply.

But where  the errata fix ? 

I have download it newest errata file, and it says the workaround need to modify " RFSHCTL.refresh_burst = 0".

I search file : drivers/ddr/imx/imx8m/ddr_init.c and lpddr4_timging.c in "imx-windows-bsp-w21h2-1-5-0", and no where to find the modify.

[Bio_TICFSL]

Hi,

Yes this works still pending by our developer and the Microsoft developers since the 1.5v of the BSP has the issue.

Regards