• State
  • Locked Locked
  • Replies 10 replies
  • Subscribers 33 subscribers
  • Views 351 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TUSB4041I: Trouble with plastic USB drives being seen as boot disks?

tony tumminaro
Prodigy 110 points
Part Number: TUSB4041I

Hello, 

I have an interesting situation with a recent TUSB4041 design. The TUSB4041 hub works fantastic in all aspects, excluding when I try to use a plastic SanDisk USB drive as a boot device through the hub. The hub works with all other USB drives I've tried to boot from, but never the plastic SanDisk. Reads and writes to the SanDisk are fine in regular use through the hub, and the plastic SanDisk works as a boot device if plugged directly into the PC. 

I'm wondering if anyone has seen behavior like this before? I have a few thoughts I'm trying to prove/disprove now, but wanted to ask in hopes it would save some time. 

Utilizing SN75240PWR and TPS2044BD in the design as well, wondering if they are involved somehow.

Thank you for any an all replies.

  • 0
    Prodigy 110 points

    I wanted to offer one more piece of information. If I have another bootable drive AND the plastic SanDisk inserted into the hub at the same time, BIOS does see the plastic SanDisk as a possible boot device. If the SanDisk is in the hub alone, it is not seen as a possible boot device (to reiterate above, the SanDisk plugged directly into a PC is recognized as a boot device).

  • 0 in reply to tony tumminaro
    TI__Guru 60870 points

    Hi Tony,

    It sounds like the SanDisk is taking awhile to connect to the hub, maybe just a longer VBUS debounce time or some other delay feature on the Sandisk side and the device is not showing up as connected when the host polls the hub at boot.  The TUSB4041 will send an interrupt to the host when the device is connected, but the system either seems to already have passed the point where it is checking for bootable devices or it is not responding to interrupts.  It is difficult to tell without protocol traces.  I'm not sure if this is feasible in your system but can you try powering on the SanDisk device early (externally forcing power to it) and seeing if it shows up as bootable?

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 110 points

    JMMN, 

    Appreciate the reply. I like the thought, but unfortunately that doesn't seem to have effect. Tried externally powering the SanDisk, tried adjusting the TUSB4041's reset timing, none of that seems to influence it. Even rebooting only the MB after the hub is powered doesn't seem to make a difference.

    Drive is picked up without issue once in Linux or Windows. MBR or GPT formatting doesn't seem to matter either.

    Trying to get a protocol analyzer on it now. 

    Tony

  • 0 in reply to tony tumminaro
    TI__Guru 60870 points

    Hi Tony,

    I'm sending you a friend request so you can send me the protocol trace directly.  

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 110 points

    Thank you, it may be a few days though. Our protocol analyzer is on another job at the moment.

    In the meantime, I think I have a strong clue that it's SanDisk firmware/register related. Every one of this model that I plug in reports as shown below.

  • 0 in reply to tony tumminaro
    TI__Guru 60870 points

    Is that report from when the flash drive is directly connected to the host?  It shows a USB 3.0 drive which isn't possible behind a USB 2.0 hub, it should drop to the lower data rate.

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 110 points

    Indeed, that is the case as reported though. Only that model drive.

    I'm fairly convinced something isn't reporting right on that model flash drive. It would have to be something the motherboard doesn't care about, but the hub hiccups on.

  • 0 in reply to tony tumminaro
    TI__Guru 60870 points

    Oh, that is strange.  The hub doesn't actually read the data reported by the downstream device, but it could be that the device is taking time attempting the USB 3.0 connection before dropping to USB 2.0 and that is causing a delay during boot.

    Regards,

    JMMN

  • 0 in reply to JMMN
    Prodigy 110 points

    That is a fantastic thought.

    Traces should tell that story if it is the case. May even be able to se it on a scope. I can try that tomorrow.

  • 0 in reply to tony tumminaro
    TI__Guru 60870 points

    Ok, sounds good.