Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756031AbdGKPBW (ORCPT ); Tue, 11 Jul 2017 11:01:22 -0400 Received: from mail-wr0-f193.google.com ([209.85.128.193]:36488 "EHLO mail-wr0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756011AbdGKPBU (ORCPT ); Tue, 11 Jul 2017 11:01:20 -0400 Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\)) Subject: Re: [PATCH] brcmfmac: added LED triggers for transmit/receive From: Russell Joyce In-Reply-To: <60de1301-b4a3-95a7-117f-a953576655d9@broadcom.com> Date: Tue, 11 Jul 2017 16:01:16 +0100 Cc: =?utf-8?B?UmFmYcWCIE1pxYJlY2tp?= , Alan Millard , Franky Lin , Hante Meuleman , Chi-Hsien Lin , Wright Feng , Kalle Valo , Marc Kleine-Budde , Marcel Holtmann , "Michael S. Tsirkin" , Pieter-Paul Giesberts , =?utf-8?B?UmFmYcWCIE1pxYJlY2tp?= , "mhiramat@kernel.org" , James Hughes , Tobias Klauser , Linux Kernel Mailing List , "linux-wireless@vger.kernel.org" , "open list:BROADCOM BRCM80211 IEEE802.11n WIRELESS DRIVER" , brcm80211-dev-list@cypress.com, Network Development Message-Id: References: <20170707140941.122115-1-russell.joyce@york.ac.uk> <60de1301-b4a3-95a7-117f-a953576655d9@broadcom.com> To: Arend van Spriel X-Mailer: Apple Mail (2.3273) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by nfs id v6BF1YQr017110 Content-Length: 1220 Lines: 24 Thanks for your comments. > What I think Rafał is saying is that it would be better to have this > code in cfg80211 so other drivers including mac80211 could use it. While I agree that moving all wireless LED triggers to cfg80211 would be an ideal situation, it seems a bit out of scope for what I was trying to achieve. This would probably also require removing the mac80211 LED triggers (and any other similar triggers that might be created by specific wireless drivers not using mac80211), in order to consolidate them in one place. Besides this, I'm not sure where exactly in cfg80211 this functionality would go (I assume it was originally put in mac80211 instead for a reason?), although I'm certainly no expert in this area of the kernel. > Indeed. However, the LED subsystem could/should(?) take care of mapping > "rx" and "tx" triggers to the same LED. In terms of the LED triggers, the only alternative I can see is to create a single complex trigger that exposes "rx" and "tx" parameters that can be individually enabled or disabled. This would reduce the number of triggers from three to one, but also makes things slightly more awkward for the user, and deviates from the convention set by mac80211.