https://bugzilla.kernel.org/show_bug.cgi?id=213081 --- Comment #19 from Alan Stern (stern@xxxxxxxxxxxxxxxxxxx) --- The usbmon trace shows that the original problem of not reporting the STALL has been fixed. But there is a new problem. Here's where it shows up in the trace for bus 6: ffff88810248e0c0 67313744 S Bi:6:004:1 -115 13 < ffff88810248e0c0 97843800 C Bi:6:004:1 -104 13 = 55534243 04000000 00000000 00 This was the second packet received from ep1in after the halt was cleared. The values in the received packet are correct (i.e., what I expect them to be), but the timestamp and the -104 error code show that the URB timed out after 30 second and was unlinked. This should not have happened; the URB should have completed normally in less than one millisecond. After this error, the device was reset and the same sequence of packets was exchanged, with the same result as before (apparently -- the trace ends during the second 30-second timeout window). This indicates that the patch does not fully fix all the issues. Evidently xhci-hcd has trouble seeing when an URB for a formerly halted endpoint has completed. -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
