Bug 13605
Summary: | usb-storage: broken with new Nokia E75 (Symbian S60 9.3 FP2) | ||
---|---|---|---|
Product: | Drivers | Reporter: | Dave Lister (foceni) |
Component: | USB | Assignee: | Greg Kroah-Hartman (greg) |
Status: | RESOLVED INVALID | ||
Severity: | high | CC: | foceni, stern |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 2.6.30 | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
kern.log: USB connect - disconnect
Windows XP usbsnoop output, gzipped Linux usbmon output |
Description
Dave Lister
2009-06-22 20:55:51 UTC
Created attachment 22066 [details]
Windows XP usbsnoop output, gzipped
Begins with phone USB connect. Windows initiates fs scan, etc, so parts of that are unfortunately present as well. I didn't know how to grab just the mass storage conversation and didn't want to remove something important.
Created attachment 22067 [details]
Linux usbmon output
Complete USB conversation. Seems kinda small, as if the data packets were truncated. This is my first usbmon capture, you're welcome to let me know how to do it properly.
I did just this (#1 = E75):
cat /sys/kernel/debug/usbmon/1u > tmp/usbmon1.log
I found a viable workaround, I guess it might come handy to other people searching the web for these error messages. Use external card reader and reformat your card, removing all partitions, with the filesystem directly on the device. This is not a solution, mass storage in Linux is still broken with valid single-partition cards in Symbian S60 9.3 FP2 devices, while Windows work fine. Can you provide a usbmon log with a reformatted card, for comparison with the non-working card? It's interesting to note that the Windows usbsnoop log also shows a lot of "Medium not present" errors. In fact, essentially nothing worked until URB 145. Just prior to that the device reported "Not-ready to ready transition", which suggests that the previous error codes were all wrong -- they should have been "Logical unit is in process of becoming ready" instead of "Medium not present". It's also interesting to see that the usbmon log shows the device responded properly to a 32-sector read starting at sector 0, but then started reporting errors when asked to do an 8-sector read starting at sector 7729024 (out of 7959552). But when the kernel stopped trying to read sectors near the end, the device worked properly again. In the usbmon log I count a total of no more than 40 of those errors, lasting for about 1-1/2 seconds. That's a little better than "repeated infinitely". And at the end of the log, everything appeared to be working correctly. So are you sure this is really a problem? It appears that Windows had very similar or even worse problems, but the error messages weren't so readily visible. And have you tried reading sectors near the end of the card under Windows? All USB bugs should be sent to the linux-usb@vger.kernel.org mailing list, and not entered into bugzilla. Please bring this issue up there, if it is still a problem in the latest kernel release. |