Received: by 2002:a25:683:0:0:0:0:0 with SMTP id 125csp608697ybg; Tue, 9 Jun 2020 08:27:29 -0700 (PDT) X-Google-Smtp-Source: ABdhPJz9NHrMV0VihKKpd/xvgKPrBPmPXYjI3TF4GoFAveyZD4eNBeQff/eNYVc11doFk+R3kKnK X-Received: by 2002:a17:906:2e83:: with SMTP id o3mr25357037eji.312.1591716449242; Tue, 09 Jun 2020 08:27:29 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1591716449; cv=none; d=google.com; s=arc-20160816; b=yRQwLkFslEPE3FiN2ce5AhaD1DxTh1wtATLCgUYKDAoPfHuiX73QLCh5RBos7DskRU xnwZA6moKrzW8XB6sBHN/17jo4qHHiCFbm4iqhyPQiegpMzdhFLjeAQFtuXks7fWZK53 GZdg8bRNiypFIYlMKNHwfX2Jr0+qvqrlSMbfAdTKok1y/dJxkBkijZ3eD7KAumgeRO9n ICwq0+EQabKkMZtTUCBXKBSGzIcRlRfLjwU6QVx07IUllebTHEZSQLJp71+ydDXu86DN yxX5yhpYx/DYo3vHzJqvyEmpm8mt9i++wFDowadGb0YOwa6CxL2IEe65XeYrtf7kjv// pyIg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:content-transfer-encoding:in-reply-to :mime-version:user-agent:date:message-id:from:cc:references:to :subject:dkim-signature:dkim-filter; bh=slq4ZicsaXpn/WdmJBVvIQ371sS7fVL5zvg7Gxo6yTU=; b=UgvK9LeW38uJiPmku5znqP5gf2Gizq7ulu5KPXvmtELilEqKm7wnvIeP0dvMG/z8y2 E+5uo2aEDVUQ4wzW8LcoKWrRO0Vo2U+ZPLqbb0mImCQh92M/udKSlDErKNmwQsOrtMs7 rdsXAgVuQtgaRSSjthF+/diKYGeASzeTY+VU+OW2+ARmDSRvYKjlVZ7Zuc7zC90mn7ZD b7kqyEez9qRTlK0s02P6KtOhVBzi9o/ebr1ji7E4FYAzpUHx8QVhbOdAvzkMJzU5YNFF ixtHzBFCxOoP5WmVgcGEu5MN32/1qdH8SX0gaWrFoFdIMXwt2hprDYh4q/ubd9spTqK9 pIow== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@candelatech.com header.s=default header.b=WKDb5ssP; spf=pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-wireless-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=candelatech.com Return-Path: Received: from vger.kernel.org (vger.kernel.org. [23.128.96.18]) by mx.google.com with ESMTP id bf20si10953455edb.66.2020.06.09.08.27.03; Tue, 09 Jun 2020 08:27:29 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18; Authentication-Results: mx.google.com; dkim=pass header.i=@candelatech.com header.s=default header.b=WKDb5ssP; spf=pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-wireless-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=candelatech.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729027AbgFIPYL (ORCPT + 99 others); Tue, 9 Jun 2020 11:24:11 -0400 Received: from mail2.candelatech.com ([208.74.158.173]:37344 "EHLO mail3.candelatech.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726395AbgFIPYL (ORCPT ); Tue, 9 Jun 2020 11:24:11 -0400 Received: from [192.168.254.4] (unknown [50.34.197.93]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mail3.candelatech.com (Postfix) with ESMTPSA id C81C813C2B1; Tue, 9 Jun 2020 08:24:10 -0700 (PDT) DKIM-Filter: OpenDKIM Filter v2.11.0 mail3.candelatech.com C81C813C2B1 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=candelatech.com; s=default; t=1591716250; bh=2d7k1yl0VFZTgP4ppt4CWi/BOS9sY64dj84tCjTtAnw=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=WKDb5ssP+9r8iMpqeUHzJY3JE+dyGy1bxb/lmEB/xN3d1vS7e+/9c2/3zHi8NGIGN VB5rixiM4vsPYWECET4px0IjM4B4N8Ysz9yycIGIjqlsIvSfG2v0BqBYuctE4Gf/dD Kqr8aPOmPVcWTvbWsgrPfRy39k09XHuuiw/6EH6k= Subject: Re: ax200 very poor performance in stock 5.7.0 and 53.c3 firmware. To: Tim Higgins References: <3B9D5022-D95A-4CC3-B8A0-1104358F241E@timhiggins.com> <8f1f9938-2fa6-38ca-b9e8-c792078ee3c1@candelatech.com> Cc: "linux-wireless@vger.kernel.org" From: Ben Greear Message-ID: Date: Tue, 9 Jun 2020 08:24:10 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.8.0 MIME-Version: 1.0 In-Reply-To: <8f1f9938-2fa6-38ca-b9e8-c792078ee3c1@candelatech.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-wireless-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-wireless@vger.kernel.org On 06/04/2020 07:38 AM, Ben Greear wrote: > > > On 06/04/2020 04:35 AM, Tim Higgins wrote: >> >> >>> On Jun 3, 2020, at 11:54 PM, Ben Greear wrote: >>> >>> I'm testing an un-modified kernel for a change, and latest firmware that I could find, >>> but ax200 is still sucking. >>> >>> UDP upload of single ax200 station gets about 1Mbps, and station disconnects >>> very often. TCP download gets about 400Mbps on each of two radios. UDP download stalls >>> often and averages around 1Mbps throughput. TCP upload also stalls out. >>> >>> Previous testing on other kernels has shown problems with other firmware as well. >>> >>> I'm requesting 500Mbps UDP speeds. Requesting only 100Mbps does not help (tested with DL direction only). >>> >>> [root@ct523c-0b29 ~]# ethtool -i wlan2 >>> driver: iwlwifi >>> version: 5.7.0 >>> firmware-version: 53.c31ac674.0 cc-a0-53.ucode >>> expansion-rom-version: >>> bus-info: 0000:12:00.0 >>> supports-statistics: yes >>> supports-test: no >>> supports-eeprom-access: no >>> supports-register-dump: no >>> supports-priv-flags: no >>> >>> >>> Is there any way to disable OFDMA or /ax on the ax200 in case that helps somehow? >>> >>> Thanks, >>> Ben >> Hi Ben, >> Try running with no encryption. I know that is not a solution. But it could point to a cause. I've seen disabling encryption make a big difference. > > I set to open auth (was using wpa3 PSK previously), and started TCP upload on two stations. One is doing around 80Mbps, the other seems to have crashed the firmware > and will not recover. To follow up on this... I tried the backports-iwlwifi project, and that driver is significantly more stable. But, you cannot use any other mac80211 driver (ie, ath10k) with the backport-iwlwifi modules loaded since the ath10k would have been compiled against a different mac80211 and then ath10k crashes on startup. So, I ported the backports-iwlwifi driver into my 5.4 kernel, and enough of the mac80211 changes as well so it could compile. This is working OK-ish, though FW still crashes and wlans don't always restart properly after crash. Looks like I can work around this last part by bouncing things in user-space after detecting a crash... Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com