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.

OMAP3530 EHCI PORT

Other Parts Discussed in Thread: OMAP3530, TUSB1210

Can please someone of TI explain what is the problem of the usb HOST port in the OMAP3530, i have tried beagle board, and other boards containing the omap3530 and all of the drops the ehci under load.....

I never seen so much problems with a standart port usb port...

Can the problem be mitigated using a software patch ?

is a new revision of the OMAP needed ?

 

Please answer...

Nuno

  • The debugging is on going and as of now it seems the issue is with noise on 1.8V supply to PHY line. There is no software workaround for this issue. We may get final fix for this very soon.

    Regards,
    Ajay

  • Hi Ajay,

    Is the fix to this issue available now ?

     

    Regards,

    CM.

  • OMAP3EVM REV-G has got the fix for this and so far EHCI seems to be very stable. The change was to use VAUX-2 supply got EHCI PHY's VIO_1V8 line from TWL4030 power chip.

    Regards,
    Ajay

  • Dear Ajay,

    Is this post still alive?

    I am also facing the same issue with EVM 3530 Rev G board.

    Following is my setup

    1. Configured EHCI for ULPI - 12 pin port2

    2. Connected a SMSC 9500 HUB (It has two USB port and one ethernet)

    3. Running a camera application streaming at 30 or 60 fps (i also checked in linux )

    4. Video overlay 1 is used by camera for streaming

    USB host is working if i did not run the camera application. But if i run the camera application after few minutes USB is not responding. I could see the PHY clock (60 Mhz)

    What could be the issue?

    Jethin

  • sekhar,

     

    You are hitting known hardware issue of EHCI with heavy DSS transfers. refer http://focus.ti.com.cn/cn/lit/er/sprz278f/sprz278f.pdf

    This issue has been fixed in OMAP3630 ES1.2 (or AM37x ES2.1)

    Please use the patch as shared in below link,

    http://marc.info/?l=linux-usb&m=129472850014890&w=2

    This patch restarts teh EHCI controller after hitting the issue,

    Regards,

    Ajay

  • HI Ajay,

    Thanks for your inputs, Can i get the updated code for Windows CE 6.0, As i am working in Wince 6.0?

     

     

    Regards,

    Sekar

  • Hi Ajay,

    I have some questions about your patch. 

    Apart from building the ehci as "in-built" what else are the prerequisites for the patch to be active. It looks like, queuing the "recover work" doesn't happening. I added several printks but I couldn't see anything on the console. I am pretty sure that part of code was not executed. 

    Any ideas?

     

    -Kalai

  • We are running into some issues as described at the end of this thread:

     

    http://e2e.ti.com/support/dsp/omap_applications_processors/f/447/t/56733.aspx?PageIndex=1

     

    On rev 2 of our design, we used SMSC phys with the OMAP3530 EHCI controller, and the ports won't suspend/resume, but with these phys, and the batch of OMAPs we used on this revision, the data communication was always reliable.

    With rev3, we used TUSB1210 phys, and we are getting data errors on port 1.  Port 2 seems to be working OK.

    We are hitting the timeout condition described in the above patch (http://marc.info/?l=linux-usb&m=129472850014890&w=2).  I then turned off the DSS subsystem, and we're still getting errors, so perhaps there are multiple conditions that can cause timeouts.

    Appreciate any other suggestions.

  • Cliff,

    As i have used SMSC 9500 and DSS in turned off state, doesn't find any such issues. and successfully entered suspend and resumed.

    Regards

    Jethin