AM62L: Get SoC UID in bootloader

Sascha Hauer

Part Number: AM62L

Hi,

The TISCI API used to make the SOC_UID available to the bootloader and Linux via the TISCI_MSG_GET_SOC_UID call. On AM62l TISCI is no longer provided to the non secure world. What are your plans to make the SOC UID available again? Will there be a custom SIP call
to the TF-A or something like that? Will at some point the TISCI API available again?

8 months ago

0 Hong Guan64 8 months ago

TI__Guru* 77840 points

Yes, you're right that only subset of TISCI API on OTP efuse programming are currently supported via TF SiP services
https://github.com/TexasInstruments/arm-trusted-firmware/blob/ti-master/plat/ti/k3/include/k3_sip_svc.h#L17
which call into tisci driver under TF
https://github.com/TexasInstruments/arm-trusted-firmware/tree/ti-master/plat/ti/k3/common/drivers/ti_sci
Best,
-Hong

0 Sascha Hauer 8 months ago in reply to Hong Guan64

Prodigy 70 points

This doesn't answer my questions

0 Hong Guan64 8 months ago in reply to Sascha Hauer

TI__Guru* 77840 points

Due to the new AM62L SoC architecture, there's SW re-architecture on SW entity (u-boot/kernel/OPTEE... etc...) calling the TIFS secure services as well as calling the DM-services (TF-A on AM62L). In addition to the SoC_UID service, any other secure services are needed?
I'm checking internally on schedule estimate.
Best,
-Hong

0 Sascha Hauer 7 months ago in reply to Hong Guan64

Prodigy 70 points

Not that I am aware of currently. It would however be good if we could read back the fused SMPK/BMPK hashes from the bootloader or Kernel

0 Hong Guan64 7 months ago in reply to Sascha Hauer

TI__Guru* 77840 points

The SMPK-H & BMPK-H are read-accessible from MMR registers.
- SMPK-H part1: starting @0x44234800 with len=32 bytes
- SMPK-H part2: starting @0x44234850 with len=32 bytes
- BMPK-H part1: starting @0x44234900 with len=32 bytes
- BMPK-H part2: starting @0x44234950 with len=32 bytes

You'll find more details from <Table 10-4632. Security Manager Registers> in K3 Security Hardware Architecture TRM.

I'm attaching a sample log on AM62L HS-SE where I programmed the TI testing SMPK-H & BMPK-H.

Best,
-Hong

Fullscreen am62l_smpk-h_bmpk-h.log Download

NOTICE:  bl1_plat_arch_setup arch setup 
NOTICE:  Booting Trusted Firmware
NOTICE:  BL1: v2.12.0(release):11.01.14-1-g6c8ef6729
NOTICE:  BL1: Built : 17:20:39, Nov 13 2025
NOTICE:  BL1: dram_class: 11
NOTICE:  lpddr4: post start - PI training status=0x27c0a000 
NOTICE:  bl1_platform_setup DDR init done
NOTICE:  k3_bl1_handoff ENTERING WFI - end of bl1
NOTICE:  BL31: v2.12.0(release):11.01.14-1-g6c8ef6729
NOTICE:  BL31: Built : 17:20:40, Nov 13 2025
NOTICE:  SYSFW ABI: 4.0 (firmware rev 0x000b '11.1.12-v11.01.12 (Fancy Rat)')
ERROR:   Agent 0 Protocol 0x10 Message 0x7: not supported

U-Boot SPL 2025.01-gc779c758475c-dirty (Nov 13 2025 - 14:20:45 -0600)
SPL initial stack usage: 1984 bytes
Trying to boot from MMC2
ERROR:   Agent 0 Protocol 0x10 Message 0x7: not supported


U-Boot 2025.01-gc779c758475c-dirty (Nov 13 2025 - 14:20:45 -0600)

SoC:   AM62LX SR1.1 HS-SE
Model: Texas Instruments AM62L3 Evaluation Module
DRAM:  2 GiB
ERROR:   Agent 0 Protocol 0x10 Message 0x7: not supported
Core:  82 devices, 31 uclasses, devicetree: separate
MMC:   mmc@fa10000: 0, mmc@fa00000: 1
Loading Environment from nowhere... OK
In:    serial@2800000
Out:   serial@2800000
Err:   serial@2800000
Net:   eth0: ethernet@8000000port@1
Warning: ethernet@8000000port@2 (eth1) using random MAC address - fe:18:36:7d:ca:ed
, eth1: ethernet@8000000port@2
Hit any key to stop autoboot:  2  0 
=> md.b 0x44234800 0x200
44234800: 1f 60 02 b0 7c d9 b0 b7 c4 7d 9c a8 d1 aa e5 7b  .`..|....}.....{
44234810: 8e 87 84 a1 2f 63 6b 2b 76 0d 7d 98 a1 8f 18 97  ..../ck+v.}.....
44234820: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234830: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234840: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234850: 60 df d0 f2 3e 2b 0c b1 0e c7 ed c7 c6 ed ac 3d  `...>+.........=
44234860: 9b df ef e0 ed dc 3f ff 7f e9 ad 87 51 95 52 7d  ......?.....Q.R}
44234870: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234880: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234890: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442348f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234900: 38 4b e2 78 a7 a5 0e b2 5a fd ff ac 2e 8b d3 06  8K.x....Z.......
44234910: f8 2a 3b 51 a7 70 f8 05 6c 9d de b9 f3 1b 0d 3d  .*;Q.p..l......=
44234920: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234930: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234940: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234950: 3e a0 06 3e 6d e3 12 7a 47 c8 a1 44 3f c7 e1 0d  >..>m..zG..D?...
44234960: ad ff b5 16 01 ae ae b4 99 d6 07 e0 28 74 cd 80  ............(t..
44234970: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234980: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
44234990: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
442349f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
=>

Processors

Processors forum

AM62L: Get SoC UID in bootloader