Return-path: Received: from mx1.redhat.com ([209.132.183.28]:39572 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751428AbeACLpi (ORCPT ); Wed, 3 Jan 2018 06:45:38 -0500 Date: Wed, 3 Jan 2018 12:45:37 +0100 From: Stanislaw Gruszka To: Enrico Mioso Cc: linux-wireless@vger.kernel.org, Johannes Berg , Daniel Golle , Arnd Bergmann , John Crispin , nbd@nbd.name Subject: Re: ieee80211 phy0: rt2x00queue_write_tx_frame: Error - Dropping frame due to full tx queue...? Message-ID: <20180103114536.GB10306@redhat.com> (sfid-20180103_124542_471881_3C0A551D) References: <20171213152017.GA3554@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, Dec 26, 2017 at 06:20:22PM +0100, Enrico Mioso wrote: > Ok, observing a stall. Unfortunately, it still happens. > rt2800_wmac 10180000.wmac: loaded eeprom from mtd device "factory" > ieee80211 phy0: rt2x00_set_rt: Info - RT chipset 2872, rev 0200 detected > ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 0005 detected Is possible that this is the FW problem for this particular RT2872 device or we do not correctly program registers for it. I think the issue is that at some point HW/FW is not capable to sent frames and do not report to driver failure of sending frames, hence driver queue can not be moved forward. The solution to this will be timeout mechanism for TX, similar like for rt2800usb, however this will require substantial rt2800pci/mmio driver rewrite. I'm going to look at this, but not sure when. Cheers Stanislaw