On Wed, 2012-05-23 at 10:27 +0800, Gregory Hosler wrote: > On 05/23/2012 10:02 AM, JD wrote: > > On 05/22/2012 02:56 PM, John Horne wrote: > >> On Mon, 2012-05-21 at 17:19 -0700, JD wrote: > >>> When I take my laptop to starbucks or McDonalds hot spot, even when > >>> sitting outside, bitrate never is below 54mbps, sometime is as high as > >>> 72mb/s > >>> > >> Hi, > >> > >> Reading through your messages, you have now said two things: > >> > >> 1) You used an RTL card and got the same problem; 2) Using the laptop > >> elsewhere (starbucks) and it works fine. > >> > >> To me both of those tend to indicate that the WLAN card is okay, but > >> perhaps your router (or at least something else) is causing a problem. > >> The RTL card would have used a different driver than the Atheros. > > Of course. The external rtl usb wifi adapter uses the rtl8187 driver. So, I > > really doubt the linux driver is behind this fixation of bit rate. All I am > > saying is that I did not have this problem when I was on F14. Perhaps > > something above the low lever driver is at work here, either wext driver or > > wpa_supplicant application. I found no settings on the router which would > > set the bitrate per wifi client. The other 3 clients are all at 54 Mb/s. > > > >> > >> You also said that the card worked fine with F14. Although the ath9k > >> driver will most likely have changed between F14 and F16, that wouldn't > >> explain why it works fine at starbucks or why the RTL had the same > >> problem. > > Right. > >> > >> I would say take a look at the router to see if anything unusual is > >> happening there. > >> > >> I have heard before that sometimes dropping the router/card to 802.11b or > >> g can resolve odd problems with 802.11n cards. (Something like 'iwconfig > >> wlan0 rate 54M') Obviously the speed will not be the best, but possibly > >> better than 18Mb/s :-) > >> > >> Finally, if the driver really looks like a possibility, then you could > >> run 'modinfo ath9k' to see what options the driver takes. It may have a > >> debug option to cause further messages to be written to dmesg. Actually, > >> checking dmesg may be a good step anyway (something like 'dmesg | grep > >> ath'). > >> > >> (FYI, I have an Atheros USB card which uses the carl9170 driver. The card > >> is 802.11bgn, but my router is only 'b' or 'g'. I get a fairly consistent > >> maximum 54Mb/s :-) ) > > I will look to see if the ath9k and the rtl8187 drivers have debug options > > that can be turned on. If they do, then I will have to recompile my > > kernel. > > if they do, then there is NO need to recompile the kernel. > > modinfo ath9k > > shows ... > > parm: debug:Debugging mask (uint) > > looking at the source would show you the value of the bits (and how many) in > the bit mask. Turning all the bits on might be done like this: > > put the line: > > options ath9k debug=0xFFFF > > into > > /etc/modprobe.d/ath9k.conf > > and reload the module. > Hi, No need to use modprobe.d. Just run from the command line: modprobe -r ath9k modprobe ath9k debug=0xFFFF John. -- John Horne Tel: +44 (0)1752 587287 Plymouth University, UK Fax: +44 (0)1752 587001 -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
