It seems that my initial reply did not made it to the mailing list. I'm reposting it below with links to scribd for the logs. Would like to note, that the flooding of debug messages continue's (indefinitly??) after the first USB connection attempt. So, the log messages of the file containing the boot continue into the second attempt on where I tried to prevent the log from wrapping. Not optimal, I know. But this is almost the best I can do now. And maybe this is enough?? USB log + boot + attempt (possibly wrapped): http://www.scribd.com/doc/126396934/USB-debug-boot USB log + attempt (prevented wrapping): http://www.scribd.com/doc/126396676/USB-debug-connection-attempt 2013/2/20 Ronald <ronald645@xxxxxxxxx>: > Patch applied cleanly. Kernel log says v3.8-rc7-dirty but it's v3.8 > final. I pulled linus' tree, then nouveau and then linus' tree again > to get the v3.8 tag. I *guess* that explains it since it isn't really > sensible to mention the v3.8 tag commit there. > > Furthermore, I had problems capturing the dmesg logs. It's flooding > like crazy. So I supply you with two files (attachments). > > One containing a connect with boot procedure (as requested if I read > correctly). However, I'm almost sure the log wrapped due to a small > dmesg buffer. I set CONFIG_LOG_BUF_SHIFT=21, but it still wraps. This > is with a device connect and disconnect plus boot log. The attachment > is called: usb_log_boot.txt.gz > > So I have added another file with debug, in which I did something like > this in bash: > while true; do dmesg -c >> log.txt; done . Then I connected and > disconnected the device (with looong timeout). > The attachment is called: usb_log_connectonly.txt.gz > > Will do some more testing, or refine it, if requested. I think this > will give a slight hint of what is happenning. > > Good luck! -- 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
