Description of problem: The USB 2.0 specification requires a 10ms reset recovery time (TRSTRCY) after port resets before commands are issued to USB devices. However linux kernels < 2.6.11 do not implement this recovery time leading to failures on USB2.0 devices. This failure is intermittent and seen on some devices (If other system activity causes there to be a 10ms delay between port resume and device access, then it works, else it would fail). It also depends on the Host controller implementation. The EHCI controller in the Broadcom HT1000 (BCM5785) SouthBridge does not like this violation of the spec. and fails to initialize devices. Version-Release number of selected component (if applicable): 2.6.9-11.EL and all older versions as well. How reproducible: Intermittent on RHEL4. Always on RHEL3. Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: This was fixed via a patch - http://marc.theaimsgroup.com/?l=linux-usb-devel&m=110859443620747&w=2 GregKH accepted this patch into the 2.6.11 kernel. [PATCH] USB Hub driver: Add reset recovery-time delay This patch is clearly needed for us to be in compliance with the USB spec. It adds the mandated recovery-time delay following a port reset. Regardless of anything else we do to alter the device initialization sequence, this is necessary. Signed-off-by: Alan Stern <stern.edu> Signed-off-by: Greg Kroah-Hartman <gregkh> drivers/usb/core/hub.c | 3 +++ 1 files changed, 3 insertions(+) diff -Nru a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c --- a/drivers/usb/core/hub.c 2005-02-16 14:47:36 -08:00 +++ b/drivers/usb/core/hub.c 2005-02-16 14:47:36 -08:00 @@ -1382,6 +1382,9 @@ /* return on disconnect or reset */ switch (status) { case 0: + /* TRSTRCY = 10 ms */ + msleep(10); + /* FALL THROUGH */ case -ENOTCONN: case -ENODEV: clear_port_feature(hub->hdev,
This could explain some of the failures I am seeing with a HP printer. Sometimes printing works, sometimes things get confused, and sometimes (eg. right now) khubd just sits in D state and I need to reboot to (hopefully) print again.
Created attachment 123830 [details] Candidate #1 - cosmetically different for 2.6.9 of RHEL
committed in stream U4 build 34.12. A test kernel with this patch is available from http://people.redhat.com/~jbaron/rhel4/
I have not been able to verify the bug itself with the USB devices I have, however I have confirm the patch is in 40.1.EL, so marking as verified.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2006-0575.html