Working on EVM:
#1. Boot from eMMC, execute reset from U-boot, or restart from Linux, hangs as below snapshot:
#2. boot from MMC, it is OK.
#3. Push reset button to reboot is OK.
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Working on EVM:
#1. Boot from eMMC, execute reset from U-boot, or restart from Linux, hangs as below snapshot:
#2. boot from MMC, it is OK.
#3. Push reset button to reboot is OK.
Tony,
from the screenshots I can see that you are using an older SDK, and I know there have been issues with software-initiated device resets on AM654x on earlier builds of the SDK that have since gotten resolved. Can you please re-try using the latest processor SDK and see if this fixes your issues:
http://software-dl.ti.com/processor-sdk-linux/esd/AM65X/latest/index_FDS.html
Regards, Andreas
Hi Tony,
Tony Tang said:It is same with latest Processor SDK with booting from eMMC.
Can you post a log from this attempt with the latest SDK.
Also what hardware revision is your EVM?
E3? E4?
(should be marked on the base board, also also get printed to the U-Boot console when using the latest SDK).
Regards, Andreas
Hi Tony,
I can re-create this issue on our Rev E3 EVM with the current SDK performing eMMC raw boot. I can also re-create it with the latest-latest ti-u-boot-2019.01 development branch. Going through my archives it seems like that U-Boot leaves the eMMC in a state that prevents ROM code from booting. Will follow-up internally to see if this is something we can fix easily in (U-Boot) software.
One workaround however I just found and successfully tested is booting from eMMC UDA. Basically, treat the eMMC UDA area just like an SD card, have a FAT32 boot partition there, and a partition for Linux, etc. In this case you can reboot without any issue as in this case the ROM has less dependencies on the state the eMMC was left in it seems.
=> reset resetting ... U-Boot SPL 2019.01-g350f3927b8 (Jan 04 2020 - 17:16:58 +0000) SYSFW ABI: 2.8 (firmware rev 0x0013 '19.10.1-v2019.10a (Terrific Lla') Trying to boot from MMC1 Starting ATF on ARM64 core... NOTICE: BL31: v2.2(release):ti2019.04-447-ge516a38 NOTICE: BL31: Built : 17:05:40, Jan 4 2020 I/TC: I/TC: OP-TEE version: ti2019.04-190-g5208e5c-dev (gcc version 8.3.0 (GNU Toolchain for the A-profile Architecture 8.3-2019.03 (arm-rel-8.36))) #1 Sat Jan 4 17:07:58 UTC 2020 aarch64 I/TC: Initialized U-Boot SPL 2019.01-g350f3927b8 (Jan 04 2020 - 17:19:14 +0000) Detected: AM6-IDKAPPEVM rev E3 Detected: SER-PCIE2LEVM rev E3 Trying to boot from MMC1 U-Boot 2019.01-g350f3927b8 (Jan 04 2020 - 17:19:14 +0000) SoC: AM654 PG 1.0 Model: Texas Instruments AM654 Base Board Board: AM6-COMPROCEVM rev E3 DRAM: 4 GiB MMC: sdhci@04FA0000 - probe failed: -123 sdhci@4f80000: 0 Loading Environment from MMC... *** Warning - bad CRC, using default environment In: serial Out: serial Err: serial Detected: AM6-IDKAPPEVM rev E3 Detected: SER-PCIE2LEVM rev E3 Net: eth3: pruss1_eth, eth2: pruss0_eth Warning: cpsw_nuss@046000000 using MAC address from ROM , eth0: cpsw_nuss@046000000, eth1: pruss2_eth Hit any key to stop autoboot: 0 => reset resetting ... U-Boot SPL 2019.01-g350f3927b8 (Jan 04 2020 - 17:16:58 +0000) SYSFW ABI: 2.8 (firmware rev 0x0013 '19.10.1-v2019.10a (Terrific Lla') Trying to boot from MMC1 Starting ATF on ARM64 core... NOTICE: BL31: v2.2(release):ti2019.04-447-ge516a38 NOTICE: BL31: Built : 17:05:40, Jan 4 2020 I/TC: I/TC: OP-TEE version: ti2019.04-190-g5208e5c-dev (gcc version 8.3.0 (GNU Toolchain for the A-profile Architecture 8.3-2019.03 (arm-rel-8.36))) #1 Sat Jan 4 17:07:58 UTC 2020 aarch64 I/TC: Initialized U-Boot SPL 2019.01-g350f3927b8 (Jan 04 2020 - 17:19:14 +0000) Detected: AM6-IDKAPPEVM rev E3 Detected: SER-PCIE2LEVM rev E3 Trying to boot from MMC1 U-Boot 2019.01-g350f3927b8 (Jan 04 2020 - 17:19:14 +0000) SoC: AM654 PG 1.0 Model: Texas Instruments AM654 Base Board Board: AM6-COMPROCEVM rev E3 DRAM: 4 GiB MMC: sdhci@04FA0000 - probe failed: -123 sdhci@4f80000: 0 Loading Environment from MMC... *** Warning - bad CRC, using default environment In: serial Out: serial Err: serial Detected: AM6-IDKAPPEVM rev E3 Detected: SER-PCIE2LEVM rev E3 Net: eth3: pruss1_eth, eth2: pruss0_eth Warning: cpsw_nuss@046000000 using MAC address from ROM , eth0: cpsw_nuss@046000000, eth1: pruss2_eth Hit any key to stop autoboot: 0 =>
Regards, Andreas
Andreas,
Thanks, How to write to UDA partition, I did not find guide in the readme:
Hi Tony,
Tony Tang said:How to write to UDA partition,
I made a previous post that shows example steps together with some explanation, please see here:
https://e2e.ti.com/support/processors/f/791/p/855749/3166126#3166126
Obviously the flow/usage is a bit different in terms of how the image should get build and deployed (and updated later) compared to eMMC raw boot, however since UDA-based boot is essentially just like using an SD card it probably means things will get simpler.
Regards, Andreas