Latest working kernel version:2.6.25 Earliest failing kernel version: 2.6.25 Distribution: OpenSuSE 11.0B3 Hardware Environment:Intel CeleronM 520, RealTek 8185 wireless chip (Gateway ML3109 laptop) Software Environment: 64 bit no proprietary modules loaded. Problem Description: The rtl8180 wireless driver works if the wireless radio is turned on at boot time. If the fn+f2 is used to turn off the radio and then used again to turn it on, or if the wireless is off during boot then turned on, the wireless is not able to see any AP. The only way to get wireless up is to rmmod rtl8180 then modprobe rtl8180 with the radio turned on. Steps to reproduce: Load the 2.6.25 kernel on a machine with the rtl8185 wireless chip. Boot with the wireless radio turned off. once booted, turn on the wireless radio. run iwlist scan no AP will be found. rmmod rtl8180 modprobe rtl8180 rerun iwlist scan AP will be listed.
> Latest working kernel version:2.6.25 > Earliest failing kernel version: 2.6.25 This means that 2.6.25 works, and it doesn't work. Please clarify?
The driver works, but it does not see the wireless radio turn on. The radio has to be on when booting for the driver to work. If the radio is off or is turned off, the driver never sees the state change when the radio is turned on for the wireless card to work. boot with fn+f2 so that the radio is on --driver sees AP(s) fn+f2 radio off -- no AP(s) seen fn+f2 radio on -- no AP(s) seen driver believes radio is off rmmod rtl8180 modprobe rtl8180 driver now sees that the radio is on and AP(s) will be seen
Could you include the contents of /var/log/messages and/or the output of running dmesg shortly after booting with the radio on, then turning it off and then back on?
Created attachment 16155 [details] message log as requested output from dmesg. booted with radio on, then turned off radio, then turned radio back on. Also showed rmmod/modprobe of rtl8180.
I don't see any indication that the driver is even aware of the rfkill button. Do you have see something to indicate that the "driver believes radio is off"? Or are you just saying that frames are not being received? Is the device built-in to your machine? Or is it a cardbus device? If it is built-in, what machine is it? At this point my guess is that the device is built-in, your machine's BIOS is physically disconnecting the antenna on your Fn+F2 keypress, and the driver (or possibly mac80211) is not coping well with the resulting loss of frames.
Gateway laptop - ML3109 - CeleronM 520 (1.6Ghz 64bit) 2GB memory, ATI chipset, RTl8185 mini-pci wireless card. Driver does not see any state change with the fn+f2. The driver appears not be be written to support rtl chips in laptops.
John , Clark, Is this bug resolved, if not i could try to help.
Wow, this is an old one... As indicated in comment 5, the rtl8180 driver has no notion of rfkill support. However, since then the mac80211 component has grown "soft rfkill" support. So if your laptop has an appropriate platform device that support your rfkill button, mac80211 will keep the device IFF_DOWN so long as rfkill is activated. I'm going to close this as INSUFFICIENT_DATA based on its age and the unfulfilled information request from comment 5. Please reopen if the problem persists and the information becomes available.