new: regression iwl3945/mac80211 endless after suspendassociate/deassociate loop

[Michael S. Tsirkin]

Starting with 2.6.27-rc5, my T60p sometimes fails to associate
with an access point after suspend to ram/resume.

This does not seem to ever happen on 2.6.27-rc4.

When this happens, I observe the following messages in dmesg:
(AP mac masked out with XX):


[16433.458084] iwl3945 0000:03:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
[16433.502490] PM: Adding info for No Bus:iwl-phy4:radio
[16433.502666] Registered led device: iwl-phy4:radio
[16433.502764] PM: Adding info for No Bus:iwl-phy4:assoc
[16433.502822] Registered led device: iwl-phy4:assoc
[16433.502870] PM: Adding info for No Bus:iwl-phy4:RX
[16433.503006] Registered led device: iwl-phy4:RX
[16433.503055] PM: Adding info for No Bus:iwl-phy4:TX
[16433.503111] Registered led device: iwl-phy4:TX
[16433.505268] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16433.522344] ADDRCONF(NETDEV_UP): eth1: link is not ready
[16433.704125] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16433.705986] eth1: authenticated
[16433.705997] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16433.708406] eth1: RX AssocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16433.708412] eth1: associated
[16433.709659] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
[16443.712280] eth1: disassociating by local choice (reason=3)
[16443.712902] eth1: disassociating by local choice (reason=3)
[16444.692059] eth1: no IPv6 routers present
[16445.997384] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16445.999304] eth1: authenticated
[16445.999312] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16446.001784] eth1: RX ReassocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16446.001790] eth1: associated
[16446.008222] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16446.009539] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16446.012433] eth1: authenticated
[16446.012440] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16446.015149] eth1: RX ReassocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16446.015155] eth1: associated
[16456.017223] eth1: disassociating by local choice (reason=3)
[16456.019751] eth1: disassociating by local choice (reason=3)
[16458.298346] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16458.300319] eth1: authenticated
[16458.300339] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16458.305557] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16458.314765] eth1: authenticated
[16458.314781] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16458.318666] eth1: RX ReassocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16458.318673] eth1: associated
[16468.321210] eth1: disassociating by local choice (reason=3)
[16470.604081] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16470.606122] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16470.608330] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16470.608367] eth1: authenticated
[16470.608374] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16470.614629] eth1: RX ReassocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16470.614648] eth1: associated
[16480.617268] eth1: disassociating by local choice (reason=3)
[16480.619939] eth1: disassociating by local choice (reason=3)
[16482.896840] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16482.899593] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16482.899626] eth1: authenticated
[16482.899633] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16482.906006] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16482.906048] eth1: authenticated
[16482.906055] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16482.907625] eth1: authenticate with AP XX:XX:XX:XX:XX:XX
[16482.909446] eth1: authenticated
[16482.909453] eth1: associate with AP XX:XX:XX:XX:XX:XX
[16482.918553] eth1: RX ReassocResp from XX:XX:XX:XX:XX:XX (capab=0x411 status=0 aid=1)
[16482.918564] eth1: associated
[16492.920224] eth1: disassociating by local choice (reason=3)
[16492.920986] eth1: disassociating by local choice (reason=3)


Reverting 8ab65b03b7893da4a49009e7e356e36e27b0c407 seems to make
the card associate properly after suspend, for me.

-- 
MST

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/