Closed agners closed 1 year ago
Tested with the latest beta firmware pieeprom-2022-11-04.bin
, no success.
SIG pieeprom.sig 9221b44ba796ba4cd2f60a27990afc66c74d7e2f72b2f7be02e9f8ab13f39481 1668678754
Reading EEPROM: 524288
Writing EEPROM
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++*************************....+
Verify BOOT EEPROM
Reading EEPROM: 524288
BOOT-EEPROM: UPDATED
When enabling UART, I get the following:
...
Trying partition: 0
type: 32 lba: 8192 oem: 'mkfs.fat' volume: ' boot '
rsc 32 fat-sectors 1020 c-count 130554 c-size 4
root dir cluster 2 sectors 0 entries 0
FAT32 clusters 130554
Trying partition: 0
type: 32 lba: 8192 oem: 'mkfs.fat' volume: ' boot '
rsc 32 fat-sectors 1020 c-count 130554 c-size 4
root dir cluster 2 sectors 0 entries 0
FAT32 clusters 130554
Read config.txt bytes 2075 hnd 0x163
Read start4.elf bytes 2249280 hnd 0x3cdb
Read fixup4.dat bytes 5399 hnd 0x169
0x00b03140 0x00000000 0x00001fff
MEM GPU: 76 ARM: 948 TOTAL: 1024
Firmware: 102f1e848393c2112206fadffaaf86db04e98326 Aug 26 2022 14:03:16
Starting start4.elf @ 0xfec00200 partition 0
NVME off
+
So it seems start4.elf crashing actually?
Firmware: 102f1e848393c2112206fadffaaf86db04e98326 Aug 26 2022 14:03:16
Any chance you can try a recent version of start4.elf? Something after "Oct 5 2022" It sounds a bit like https://github.com/RPi-Distro/repo/issues/309
You can get the most recent here https://github.com/raspberrypi/firmware/tree/master/boot
If you add enable_uart=1 and uart_2ndstage=1 to config.txt it'll tell us if it's getting as far as firmware.
After realizing that the EEPROM is actually handing off too start4.elf, that is exactly what I was start doing :smile:
I manually replaced the files on the NVMe with the ones from the 1.20221104
tag, and it boots! The firmware which is currently shipped with the Raspberry Pi OS seems to be the culprit (release date 2022-09-22 according to RPi Imager, start4.elf seems to be from 1.20220830
tag).
It seems that 1.20221028
already fixed the problem. From the git log no change between 1.20220830..1.20221028
really jumps into my eye which seems like it would address an NVMe boot issue, but maybe the git log is incomplete?
In any case, booting works with the latest firmware hence this can be closed.
but maybe the git log is incomplete
Yes - my fault. The change was to turn nvme off before switching from bootloader to firmware and firmware to kernel. That "NVME off" line is showing the fix working in the bootloader. I failed to mark the change as affecting the firmware so there's no comment in the git log - apologies for that. Thanks for testing.
Ok, I see. I was actually wondering if the "NVME off" log entry could be a problem. Thanks for the insight!
Describe the bug
I am trying to boot from a SK hynix BC711 NVMe but it seems the CM4 is not able to boot from that device. The device gets successfully detected in Linux (when booting the system from another media). Also the same setup boots fine from a Samsung 970 EVO plus.
Steps to reproduce the behaviour
Device (s)
Raspberry Pi CM4 Lite
Bootloader configuration.
Updated to latest version using rpiboot
System
No response
Bootloader logs
No response
USB boot
No response
NVMe boot
Network (TFTP boot)
No response