Hide Forgot
Description of problem: System crash apparently provoked by wireless driver rtl8192se. The system crashes without warning (and without specific logs) a few minutes to a few hours after boot when the realtek 8192se wireless chip using the F16 3.1.0-7.fc16.x86_64 kernel with driver module rtl8192se. The driver works fine until the crash. However with a linksys USB wireless adapter card and the rtl8192se driver rmmod-ed the crash does not happen. With rtl8192se from F15 kernel 2.32? on the same Toshiba L510 laptop, the crash did not occur either. Version-Release number of selected component (if applicable): rtl8192se driver in kernel 3.1.0-7.fc16.x86_64 How reproducible: after every boot, if the Realtek wireless chip and linux 3. kernel are used. It may take up to 4 hours for the crash to occur, but it will occur. NetworkManager is used in all cases. Steps to Reproduce: 1.You need a network controller: Realtek Semiconductor Co., Ltd. RTL8191SEvB Wireless LAN Controller (rev 10) (as installed on Toshiba L510 Satellite laptop, for example). 2.Boot Fedora 16 in standard configuration, using wireless Ethernet (wifi). 3.Wait for crash Actual results: System crashes without warning, sometime between just after boot and 4 hours later. Screen freezes, keyboard and mouse are impotent. No special logs are generated. Expected results: The system would stay up. ;-) Additional info:
Apparently this bug has already been discovered upstream: http://groups.google.com/group/linux.kernel/browse_thread/thread/88279467b40b7f58 I don't know when a solution will be incorporated in the kernel, however. Just loaded kernel 3.1.1-1.fc16.x86_64 and monitoring....
as expected, the bug remains active in 3.1.1-1.
With the help of a manual patch (http://permalink.gmane.org/gmane.linux.kernel.wireless.general/78707) I got the compat-wireless rtl8192se driver for linux 3.1.1 from wireless.kernel.org to compile and am now running it. After a couple of days I should be able to say if it is stable, at least for me. Less time if it is not.
Unfortunately, the bug is still there.
This bug may be being addressed under bug 755154. If so, a fix should be coming in new kernel packages.
The bug is fixed since Fedora kernel 3.1.5 . I am closing this bug report.