From owner-freebsd-stable@FreeBSD.ORG Sat Sep 13 20:42:21 2014 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 89F1AC97; Sat, 13 Sep 2014 20:42:21 +0000 (UTC) Received: from isis.morrow.me.uk (isis.morrow.me.uk [204.109.63.142]) by mx1.freebsd.org (Postfix) with ESMTP id 6219486D; Sat, 13 Sep 2014 20:42:20 +0000 (UTC) Received: from anubis.morrow.me.uk (unknown [93.89.81.46]) (Authenticated sender: mauzo) by isis.morrow.me.uk (Postfix) with ESMTPSA id 6EBF745087; Sat, 13 Sep 2014 20:42:14 +0000 (UTC) DKIM-Filter: OpenDKIM Filter v2.8.3 isis.morrow.me.uk 6EBF745087 DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=morrow.me.uk; s=dkim201101; t=1410640934; bh=G+eF+9ANPZM2D6VGugzxTYN9jKGwO+tsxtmVDFrLOqI=; h=Date:From:To:Subject:References:In-Reply-To; b=uaVali5U4bWeGQ05KtWLDd75hqnip/kBKy0wzrzmUyWnnPdLaAFqDE4fv+QBkNr3f wRCRJBPUcIRmpswhp/i/SM9w2+7I14tF8uMvImgYXhbi5VOqxPr639WuajOL8OkkQ9 pwCowzNFWQbT/e6gW6IMF3R8W8m8JEj4PTOpuNfY= X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.98.1 at isis.morrow.me.uk Received: by anubis.morrow.me.uk (Postfix, from userid 5001) id 3410A1517B; Sat, 13 Sep 2014 21:42:12 +0100 (BST) Date: Sat, 13 Sep 2014 21:42:12 +0100 From: Ben Morrow To: dumbbell@FreeBSD.org, freebsd-stable@freebsd.org Subject: Re: 10.1 kernel device config for drm2 and kms drivers Message-ID: <20140913204209.GA18072@anubis.morrow.me.uk> Mail-Followup-To: dumbbell@FreeBSD.org, freebsd-stable@freebsd.org References: <20140911064731.GA15930@rwpc15.gfn.riverwillow.net.au> <54115D2A.5020802@FreeBSD.org> <20140911084624.GK2737@kib.kiev.ua> <20140911211428.GA6247@anubis.morrow.me.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <54146CCF.7020107@FreeBSD.org> X-Newsgroups: gmane.os.freebsd.stable User-Agent: Mutt/1.5.23 (2014-03-12) X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 13 Sep 2014 20:42:21 -0000 Quoth =?windows-1252?Q?Jean-S=E9bastien_P=E9dron?= : > On 11.09.2014 23:14, Ben Morrow wrote: > > I didn't realise it was possible to load radeonkms after boot. > > Presumably if you do that you lose even more of the boot messages when > > the new console loads? Currently it appears to lose everything that was > > above the top of the screen when the new driver loaded; it would be > > better (if possible) if it were to keep the scrollback as well. > > When a KMS driver is loaded, the history buffer of vt(4) grows and the > current data is preserved. Therefore you should not lose any data. > > One caveat is that the cursor is put at the top of the screen, so the > previous content is off-screen. Another issue is that the history size > (hard-coded) is admittedly very small. But data should be there. > > Could you please describe your problem in more details? Previously, with syscons, I could usually scroll back after boot had finished and the login: prompt appeared and get all the way up to the start of the kernel messages (the Copyright line). Now, with vt, and loading radeonkms from rc.conf rather than loader.conf, if I switch to the console, I can: scroll up through 8 screens of kernel and userland console messages until I come to a screen which has info: [drm] Initialised drm 1.1.0 20060810 in approximately the middle; then I can scroll up one more screen only, which takes me to somewhere in the middle of probing the USB ports. Looking at dmesg suggests there should be 3 or 4 more screens of messages above that. Obviously I can't be certain without trying more things that this isn't just because the rest has gone off the top of the buffer, but ISTR when I was first changing the machine over to vt that when I didn't load radeonkms and stayed with vt_vga I could still scroll all the way up. I may be misremembering, and/or making assumptions based on the fact the end of the scrollback is about a screenful above drm loading; I can do more experiments next time I reboot the machine if that would be helpful. Ben