From owner-freebsd-questions@freebsd.org Wed Apr 15 02:17:53 2020 Return-Path: Delivered-To: freebsd-questions@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 6CEB02CFCB5; Wed, 15 Apr 2020 02:17:53 +0000 (UTC) (envelope-from pete@nomadlogic.org) Received: from mail.nomadlogic.org (mail.nomadlogic.org [174.136.98.114]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) server-signature RSA-PSS (4096 bits) client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "mail.nomadlogic.org", Issuer "Let's Encrypt Authority X3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4925c02c9Gz46hW; Wed, 15 Apr 2020 02:17:51 +0000 (UTC) (envelope-from pete@nomadlogic.org) Received: from [192.168.1.206] (cpe-23-243-162-239.socal.res.rr.com [23.243.162.239]) by mail.nomadlogic.org (OpenSMTPD) with ESMTPSA id 3e7e28ee (TLSv1.3:TLS_AES_256_GCM_SHA384:256:NO); Wed, 15 Apr 2020 02:17:45 +0000 (UTC) Subject: Re: drm-i915kms + x11-intel eats out all of the ram and swap but not with x11-scfb To: Theron , Tomasz CEDRO , Niclas Zeising Cc: freebsd-x11@freebsd.org, FreeBSD Questions Mailing List References: <19d0dbdc-f1ac-22f1-6934-076bb5733ab7@daemonic.se> <24c43065-527d-499b-282a-43df2a947a1d@gmail.com> From: Pete Wright Message-ID: Date: Tue, 14 Apr 2020 19:17:44 -0700 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:68.0) Gecko/20100101 Thunderbird/68.7.0 MIME-Version: 1.0 In-Reply-To: <24c43065-527d-499b-282a-43df2a947a1d@gmail.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Content-Language: en-US X-Rspamd-Queue-Id: 4925c02c9Gz46hW X-Spamd-Bar: --- Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=pass (mx1.freebsd.org: domain of pete@nomadlogic.org designates 174.136.98.114 as permitted sender) smtp.mailfrom=pete@nomadlogic.org X-Spamd-Result: default: False [-3.56 / 15.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; RCVD_TLS_ALL(0.00)[]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; R_SPF_ALLOW(-0.20)[+mx]; NEURAL_HAM_LONG(-1.00)[-1.000,0]; TAGGED_RCPT(0.00)[freebsd]; MIME_GOOD(-0.10)[text/plain]; DMARC_NA(0.00)[nomadlogic.org]; RCPT_COUNT_FIVE(0.00)[5]; RECEIVED_SPAMHAUS_PBL(0.00)[239.162.243.23.khpj7ygk5idzvmvt5x4ziurxhy.zen.dq.spamhaus.net : 127.0.0.10]; TO_MATCH_ENVRCPT_SOME(0.00)[]; NEURAL_HAM_MEDIUM(-1.00)[-0.998,0]; IP_SCORE(-2.76)[ip: (-9.22), ipnet: 174.136.96.0/20(-4.11), asn: 25795(-0.42), country: US(-0.05)]; FREEMAIL_TO(0.00)[gmail.com]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:25795, ipnet:174.136.96.0/20, country:US]; MID_RHS_MATCH_FROM(0.00)[]; SUSPICIOUS_RECIPS(1.50)[]; RCVD_COUNT_TWO(0.00)[2] X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Apr 2020 02:17:53 -0000 On 4/14/20 5:40 PM, Theron wrote: > Please forgive the following for being not exactly related to the > memory leak, but very relevant to scfb vs. modesetting vs. intel: > > On 2020-04-14 19:44, Tomasz CEDRO wrote: >> I forgot that xf86-video-intel is for the old cards only > So, new cards (newer than skylake at least) don't encounter > frame-tearing problem (completely unacceptable to me) even with > modesetting driver and no compositor ?  But the software design bug is > still there either way if it can happen on some hardware. Sorry, I > have no idea whether it's a hardware timing being configured wrong or > a software data race, but when I've heard "use a compositor" proposed > as a solution, it smells of underlying ill-conceived implementation.  > I wouldn't be surprised if Linux graphics has this same problem, but I > don't run that. > > Shame to see xf86-video-intel being treated as if it's deprecated > software meanwhile modesetting driver lags behind it on this basic > level of image quality. to be clear the decision to move to modesetting instead of chip specific DDX's seems to have been made upstream on the Xorg side. having said that there are still updates landing upstream to bot the intel and amdgpu DDX's, but i'm not certain on the long term plan anymore tbh.  i periodically run the intel DDX and it works well enough with our drm-kmod (and forcing the "tear-free" option in xorg works with SNA iirc).  I also run the amdgpu with our drm-kmod daily and have no issues with xorg there - perf is quite good imho. -p -- Pete Wright pete@nomadlogic.org @nomadlogicLA