Return-path: Received: from mail-vw0-f52.google.com ([209.85.212.52]:36509 "EHLO mail-vw0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752910Ab1ITCqA (ORCPT ); Mon, 19 Sep 2011 22:46:00 -0400 Received: by vws10 with SMTP id 10so161093vws.11 for ; Mon, 19 Sep 2011 19:46:00 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <4E77C0BF.50108@openwrt.org> References: <1316453903-75710-1-git-send-email-nbd@openwrt.org> <1316453903-75710-2-git-send-email-nbd@openwrt.org> <4E77AAFB.1080805@openwrt.org> <4E77B25D.9070907@openwrt.org> <4E77BC58.7040204@openwrt.org> <4E77C0BF.50108@openwrt.org> Date: Tue, 20 Sep 2011 10:45:59 +0800 Message-ID: (sfid-20110920_044605_694221_986FB6F4) Subject: Re: [PATCH v2 2/4] ath9k_hw: clean up tx power handling From: Adrian Chadd To: Felix Fietkau Cc: "Luis R. Rodriguez" , linux-wireless@vger.kernel.org, linville@tuxdriver.com Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: IIRC, the TP Scale stuff dates back to the previous NICs which (reportedly, from madwifi list/tickets) didn't implement TPC right, or Madwifi never ran TPC correctly on them. So instead of per-packet TPC being done as driven by the net80211 stack (which I bet would be needed for hostap mode, to use per-node TX power levels), it was likely done as a way for STA's to select a power limit, and maybe for coarse grained hostap operation (ie, if all STA's are close, use a lower TPC scale.) I bet Sam would know why. I agree with Felix; nothing in ath9k/mac80211 currently uses it, so why keep it around. All of the ath9k supported NICs should implement correct per-packet TPC anyway, right? Adrian