CIRCUITPY drive not appearing with SAM E54 Xplained Pro

[tomiahola]

CircuitPython version

Adafruit CircuitPython 8.1.0 on 2023-05-22; SAM E54 Xplained Pro with same54p20

Code/REPL

]0;🐍Done | 8.1.0\Auto-reload is off.
Running in safe mode! Not running saved code.

You are in safe mode because:
CIRCUITPY drive could not be found or created.
Press reset to exit safe mode.

Press any key to enter the REPL. Use CTRL-D to reload.
]0;🐍Done | 8.1.0\

Behavior

USB drive appearing instead of CIRCUITPY drive.

Description

Hello, could someone help me understand why after installing the latest CircuitPython to my SAM E54 Xplained Pro, the "CIRCUITPY" drive does not appear but "USB Drive" appears instead?

When trying to open the "USB Drive", a notification appears: "Please insert a disk into USB Drive (D:)."

I'm using the latest bootloader version (3.15).

storage.erase_filesystem() did not solve the issue.

Additional information

No response

[dhalbert]

Here are some things to check:

• See https://forums.adafruit.com/viewtopic.php?p=838854#p838854, and the whole thread. A user had trouble, and it turned out some board jumpers were set wrong.
• (If I remember right:) A user had trouble with a newer SAMD21 Xplained board. It turns out Microchip changed the flash chip on the board during the chip shortage without much if any notification. Perhaps they did something similar on this board. Currently the board build is only expecting an N25Q256A chip. What do the flash chip markings say?
• May be related to (SAME54) Volume "Circuitpy" not updated after board reset #6470.

[tomiahola]

Here are some things to check:

• See https://forums.adafruit.com/viewtopic.php?p=838854#p838854, and the whole thread. A user had trouble, and it turned out some board jumpers were set wrong.
• (If I remember right:) A user had trouble with a newer SAMD21 Xplained board. It turns out Microchip changed the flash chip on the board during the chip shortage without much if any notification. Perhaps they did something similar on this board. Currently the board build is only expecting an N25Q256A chip. What do the flash chip markings say?
• May be related to (SAME54) Volume "Circuitpy" not updated after board reset #6470.

Hi, the flash chip markings are: 26F064B

[tannewt]

Please post a picture of the chip. It'll likely have a manufacturer's logo too.

[tomiahola]

Here is a picture of the chip.

[tannewt]

Thanks! That looks like a MicroChip flash chip instead of micron as the board def says. Want to make a PR adding "SST26VF064B" to the list here: https://github.com/adafruit/circuitpython/blob/main/ports/atmel-samd/boards/same54_xplained/mpconfigboard.mk#L10 ? That's the full name of the MicroChip flash (nvm.toml define here.)

[tomiahola]

Made a PR: #8054

[tannewt]

Thank you! I've tweaked it so it'll hopefully build.

[Jones14021]

Just adding this here in case anyone stumbles over this issue like I did:

My SAM E54 Xplained Pro (ordered July 2024, SST26VF064B QSPI flash chip dates to sometime in 2021 so I guess this issue has been around for some time already) has the same issue as @tomiahola described. The issue can be observed for all CircuitPython builds that include the above commit from #8054.
No response from the board, neither CIRCUITPY or USB Drive appears, and no COM port is detected.
Previous builds of CircuitPython (<8.1.0) at least boot into safe-mode but the CIRCUITPY drive also does not appear.

I was not able to fix this by playing around with the nvm.toml file for the SST26VF064B.

The only thing that worked for me was to disable the external flash completely and instead enabling the internal flash file system by replacing the QSPI_FLASH_FILESYSTEM = 1 and EXTERNAL_FLASH_DEVICES = "N25Q256A,SST26VF064B" with INTERNAL_FLASH_FILESYSTEM = 1 in mpconfigboard.mk.
Since the latest 9.X.X builds of CircuitPython are too large to fit into the default FLASH_FIRMWARE area, I also had to shrink the filesystem by defining CIRCUITPY_INTERNAL_FLASH_FILESYSTEM_SIZE (FLASH_SIZE / 4) in mpconfigboard.h.

I know that this is not a solution, but maybe someone finds this workaround helpful.

[dhalbert]

Reopening since this sounds like it is still a problem.

[eightycc]

@tomiahola Please re-test with the build artifact from #10249. Thank you!