Thanks to Ajay Kumar Gupta from TI for pointing me to a patch which I believe has resolved this issue. The patch is referenced here: http://www.spinics.net/lists/stable-commits/msg11219.html This patch was for 2.6.38, but it applies cleanly without modification to the 2.6.32 kernel we are using. Hopefully this will help others who might be experiencing this issue. Based on my previous test results, I was focusing on HCD source code which was unique to the TI EHCI, but this patch applies to the mainline EHCI code. So I would not have found this problem for a long time. Can someone explain how the root cause of this bug was discovered? In hindsight, I'm wondering if I missed something in my analysis or if there were better tools that I should have been using which would have helped me to isolate it more quickly. Any thoughts/suggestions? I'm very relieved that this patch seems to have eliminated this elusive bug. Thanks to everyone for your help. Best Regards, Ed Endejan -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
