BOT driver: filter invalid/unsupported peripheral qualifier and peripheral device type values from Inquiry data. about libusbhsfs HOT 12 CLOSED

@ginkuji Hi there, mate. Nice to meet you, and thanks for opening this issue.

This sounds like a BOT driver failure. I'll need you to provide a logfile so I can further analyze the issue and fix it. Please run the attached application, connect your USB HDD for ~20 seconds, disconnect it and exit by pressing the + button.

libusbhsfs-example.zip

The logfile will be saved to sdmc:/libusbhsfs.log. Please attach it here.

from libusbhsfs.

Here are the log files. I tried twice to make sure. It hanged even though I present the + button so I pressed home. Thank you very much for considering my concern.
https://filebin.net/l2vb65fu0jtgq6h0

from libusbhsfs.

@ginkuji Please test the attached build. Let me know how it goes.

libusbhsfs-example.zip

The real issue here is that LUN #1 from your USB HDD actually represents an Enclosure Services (SES) device, not a block device.

from libusbhsfs.

@ginkuji It's all looking good now. It didn't get stuck this time around, though - FatFs just started reading free cluster info to determine the free space in the FAT volume.

This is usually a fast operation (<= 10 seconds), but since you're currently using a debug build of the library that logs every single read/written logical block, your time expectation is exceeded.

I'll create a normal build and send it to you so you can compare. The problem should be fixed now.

from libusbhsfs.

Thank you! I will wait for the latest build. Glad I was able to contribute by reporting a bug.

from libusbhsfs.

@ginkuji And I'm glad you reported it. :)

Here's the normal build. It should be much faster.

libusbhsfs-example.zip

from libusbhsfs.

Thank you. It now works.
I also noticed that the file copy sdmc -> ums0 failed (5)
Other than that + now works.

from libusbhsfs.

@ginkuji Thanks, that helps a lot.

This is indeed an issue with the BOT driver. Your USB HDD reports two logical units - the first one is properly initialized, and its FAT volume is properly mounted, but the second one fails to be initialized.

Even though a possible STALL status is cleared from both endpoints when this occurs, it seems it's not enough. That's why the application hangs as soon as the library attempts to send the next command to your USB HDD.

I'll look into it.

from libusbhsfs.

Here is the latest log. The hardrive is now blinking indicating that it is reading. The exit button however is not working. I pressed home again after 20 seconds.
https://filebin.net/3jjsrtmzz88zxwdu

from libusbhsfs.

@ginkuji Excellent. Thanks a lot for helping test the changes. :)

from libusbhsfs.

That error at the end is expected behaviour - it tries to perform a r/w benchmark by copying a test.file file from the SD card to the USB drive.

from libusbhsfs.

Fixed in 404545e.

This will be part of the next public release.

from libusbhsfs.