Michael Buesch wrote: > On Thursday 02 October 2008 16:51:42 Rafał Miłecki wrote: >> 2008/10/1 Vaidas Mikalauskas <[EMAIL PROTECTED]>: >>> b43-phy0 ERROR: PHY transmission error >>> wlan0: No ProbeResp from current AP 00:18:e7:32:a0:a4 - assume out of range >> I guess it was Larry who was talking about PHY transmission error few >> days ago? Talking that this was introduced some time ago but it's >> extremly hard to bisect as it doesn't happen often. >> >> Vaidas, Kurt: do you get "PHY transmission error" on every disconnecting? >> >> If you do -> Larry: wouldn't it be nice to try bisecting in this >> situation, when they have PHY transmission error reproductable? >> >> I may write something stupid however ;) >> > > Guys, you are taking this way too seriously. > These PHY transmission errors are _harmless_. They don't cause any > breakage, if they only appear once in a while. > The message does only mean that we failed to transmit a packet. > That is not fatal. The 802.11 protocol does recover from it > without any problems. > > I think we should probably silence the warning and only print it > when it happens like 20 times per second. > > We have more important things to do than to waste our time on > a packet lost out of one thousand.
Michael, On occasion, the PHY errors get much more serious. In dmesg, you get the notation that 600 or 700 have been skipped. In addition, the interface will reset. I think that if anyone can reproduce these regularly, then bisection is desirable. If it happens only once out of 20 boots, or more, then skip it. Larry _______________________________________________ Bcm43xx-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/bcm43xx-dev
