bdm USB drivers still fail to recognize some devices

[DKWDRV]

@rickgaiser same issues.
MBR and GPT are being checked but the final fallback to fatfs driver is not happening. It should only happen there once MBR and GPR have been processed in that order and they have both have failed.

Here only MBR and GPT are registered.

[DKWDRV]

@rickgaiser what you mention here is not happening. MBR and GPT do run but default to exFAT driver does not.

This is getting annoying and I have no idea how to mess with bdm to add it without breaking bdm design. Any solution please?

[rickgaiser]

You need to test your PR next time.
Also you need to provide more information so we can help you.

Lets start with the BDM debug log. Most easy way to get it is using ps2link and ps2client. Just load the IOP drivers in the correct order, then plug the troubled USB stick and provide the log.

something like:

// BDM + MBR + GPT + fatfs (exFat)
ps2client -h 192.168.1.10 execiop host:bdm.irx
ps2client -h 192.168.1.10 execiop host:bdmfs_fatfs.irx

// USB drivers
ps2client -h 192.168.1.10 execiop host:usbd.irx
ps2client -h 192.168.1.10 execiop host:usbmass_bd.irx

This will show things like:

BDM: connecting device usb1p0 id=0x0
Found MBR disk
Found partition type 0x07
BDM: connecting device usb1p1 id=0x7
BDM: usb1p0 mounted to MBR

That could help figure out what's going on.

[DKWDRV]

@rickgaiser thanks for the help! I thought it did work because it was a fluke. Let me get a log really quick.

[DKWDRV]

@rickgaiser

`Block Device Manager (BDM) v1.1
BDM: connecting fs MBR
BDM: connecting fs GPT
BDM: BDM event thread running
BDM FatFs driver (FAT/exFAT) v1.4
BDM: connecting fs fatfs

usbmass_bd: USB MASS Driver v1.1
usbmass_bd: connect: devId=2
usbmass_bd: Vendor: USB
usbmass_bd: Product: SanDisk 3.2Gen1
usbmass_bd: Revision: 1.00
usbmass_bd: 0 120176639-byte logical blocks: (0MB / 512MiB)
BDM: connecting device usb0p0 id=0x0
Found MBR disk
Found partition type 0x0c
BDM: connecting device usb0p1 id=0xc
BDM: usb0p0 mounted to MBR
`

Fails when trying to access "mass0/1".

[DKWDRV]

For this device.

[DKWDRV]

MBR disk valid_partitions=%d
Is not showing at all despite parts area being invalid!
Logs shows Found partition type 0x0c but there is no 0xC byte at all in the invalid area.

[DKWDRV]

@rickgaiser with no USB connected at all, connecting the above USB device gives this:

usbmass_bd: connect: devId=2
usbmass_bd: Vendor:  USB
usbmass_bd: Product:  SanDisk 3.2Gen1
usbmass_bd: Revision: 1.00
usbmass_bd: 0 120176639-byte logical blocks: (0MB / 512MiB)
BDM: connecting device usb0p0 id=0x0
Found MBR disk
Found partition type 0x0c
BDM: connecting device usb0p1 id=0xc
BDM: usb0p0 mounted to MBR

It is passing MBR signature checks which means at least the area has MBR signature bytes. Partitions seems to be passing the check too when it should fail!

[israpps]

Btw, since you're making DKWDRV, it might be worth considering using the PowerPC UART as a logging solution. That will allow you to get rid of the network dependency (or ps2link) from whatever software you're testing on.

PS2SDK already has an IRX that pipes all stdout to the PPC UART, ppctty.irx

[rickgaiser]

This seems to be printed wrong: usbmass_bd: 0 120176639-byte logical blocks: (0MB / 512MiB). But is just a printing issue.

For this device.

How did you create this MBR? Perhaps I can recreate the issue?

MBR disk valid_partitions=%d

Is not showing at all despite parts area being invalid!
Logs shows Found partition type 0x0c but there is no 0xC byte at all in the invalid area.

Are you sure the screenshot and the disk are still the same? Becouse I don't see any 0xc as well in the screenshot. But it's not just some random value, 0xc = "FAT32 with LBA".

[DKWDRV]

@rickgaiser most likely printing issue yeah! I think adding more prefix to logs would be a better idea? Just like usbmass_bd: is?

Try this(maybe it will format the same). Very popular format tool.

https://download.cnet.com/hp-usb-disk-storage-format-tool/3000-2094_4-10974082.html

Yes they are the same! Only one USB and I rechecked the first sector in HxD. It makes no sense...

[DKWDRV]

@rickgaiser why does bdm show this?

BDM: connecting device usb0p0 id=0x0
Found MBR disk
Found partition type 0x0c
BDM: connecting device usb0p1 id=0xc
BDM: usb0p0 mounted to MBR

Why is it trying to connect both usb0p0 and usb0p1?

[DKWDRV]

Other port connect

usbmass_bd: connect: devId=1
usbmass_bd: Vendor:  USB
usbmass_bd: Product:  SanDisk 3.2Gen1
usbmass_bd: Revision: 1.00
usbmass_bd: 0 120176639-byte logical blocks: (0MB / 512MiB)
BDM: connecting device usb1p0 id=0x0
Found MBR disk
Found partition type 0x0c
BDM: connecting device usb1p1 id=0xc
BDM: usb1p0 mounted to MBR

[DKWDRV]

Another semi "random working" device after disconnecting the above.

usbmass_bd: disconnect: devId=1

BDM: disconnecting device usb1p0 id=0x0
BDM: disconnecting device usb1p1 id=0xc
BDM: usb1p0 unmounted from MBR


usbmass_bd: connect: devId=1
usbmass_bd: Vendor: Verbatim
usbmass_bd: Product: STORE N GO      
usbmass_bd: Revision: 1.00
usbmass_bd: ERROR: scsi_cmd_test_unit_ready 1
usbmass_bd: Sense Data key: 06 code: 28 qual: 00
usbmass_bd: 0 31299583-byte logical blocks: (0MB / 512MiB)
BDM: connecting device usb1p0 id=0x0
Found MBR disk
Found partition type 0x0c
BDM: connecting device usb1p1 id=0xc
BDM: usb1p0 mounted to MBR
BDM: usb1p1 mounted to fatfs

[DKWDRV]

This other device has same corrupted partitions type.
Maybe a race condition between MBR and fatfs?
If MBR was mounted above, why did fatfs mount too? MBR, GPT, fatfs driver last?

[DKWDRV]

@rickgaiser how was HP tool?