From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jouni Malinen Date: Tue, 19 May 2009 17:10:06 +0300 Subject: [ath9k-devel] EEE PC with AR9280 using N mode with WPA2 In-Reply-To: <4A128DBF.5070807@openwrt.org> References: <4A11C68F.5060108@openwrt.org> <1242717636.14488.14.camel@jm-desktop> <4A128DBF.5070807@openwrt.org> Message-ID: <1242742206.14488.20.camel@jm-desktop> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ath9k-devel@lists.ath9k.org On Tue, 2009-05-19 at 03:45 -0700, Gabor Juhos wrote: > Ok, I will wait for the patches then. I just submitted the patches and they are also available from http://w1.fi/p/ps/ After implementing additional debugging functionality, I did find number of issues with ath9k touching the hardware when in sleep mode. This did include 0xdeadbeef reads, too, which seems to be the result of a register read that should not have happened in the first place.. Anyway, after these patches (on top of the previous PS fixes that are now in wireless-testing.git), I have not seen that special value (or the interrupt indicating such access) anymore. Or well, I think it is still possible to get a somewhat similar case by unloading the driver when in PS mode, but it does not seem to cause problems at least with AR9280. If you are seeing crashes with this version on an embedded board, I would be interested in getting more details of the board and wlan version. And well, even hitting the register read/write in sleep mode issue without a crash would be good thing to report since the driver should not really trigger those in general. - Jouni