Comment 63 for bug 1315221

After a week on 3.13.0-32-generic #57+lp1349572v201408111147 on several wireless networks, I believe I can confidently say that the disassociation problem is not present in this kernel. I still get dragging speeds from time to time consistent with the rate-related problems discussed previously, but I do not lose association. In general the experience is much improved. (I've not had to reboot the machine or reload the wireless drivers at all, for example!)

I'm not sure where that puts us with regards to "fixing" this bug, or if there's anything else you'd still like to see (different kernel versions, specific debug dumps, etc.). Please let me know if there is.