From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from wolverine01.qualcomm.com ([199.106.114.254]:3651 "EHLO wolverine01.qualcomm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751615Ab1IVHYj (ORCPT ); Thu, 22 Sep 2011 03:24:39 -0400 Message-ID: <4E7AE2B0.3060905@qca.qualcomm.com> (sfid-20110922_092443_122530_BD51463E) Date: Thu, 22 Sep 2011 10:24:32 +0300 From: Kalle Valo MIME-Version: 1.0 To: Jouni Malinen CC: Subject: Re: [PATCH 0/9] ath6kl: Use cfg80211 BSS table without local node table References: <1316448907-10166-1-git-send-email-jouni@qca.qualcomm.com> In-Reply-To: <1316448907-10166-1-git-send-email-jouni@qca.qualcomm.com> Content-Type: text/plain; charset="ISO-8859-1" Sender: linux-wireless-owner@vger.kernel.org List-ID: On 09/19/2011 07:14 PM, Jouni Malinen wrote: > This patch set fixes number of issues in how scan results were being > reported to cfg80211 by properly using the cfg80211 BSS inform > mechanism. There is no point in maintaining the internal node table in > ath6kl since it is not really used for anything and the duplicated table > implementation was just trying to handle functionality that is already > present in cfg80211. In addition to fixing various roaming cases that > resulted in errors due to incorrect information in the cfg80211 BSS > table, this series removes the internal node table implementation that > had its own issues that apparently could cause crashes in some cases. This is really cool. I hated the node table right from the beginning. I noticed that there were two magic constants used few times, 8 + 2 + 2 for the IE length and 95 for RSSI conversion. I would prefer to use a define for those, but that can be changed later. I have applied all nine patches. Thank you! Kalle