From owner-freebsd-net@freebsd.org Wed Sep 18 03:26:55 2019 Return-Path: Delivered-To: freebsd-net@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 2F4FBEFA9D for ; Wed, 18 Sep 2019 03:26:55 +0000 (UTC) (envelope-from rrs@netflix.com) Received: from mail-pl1-x644.google.com (mail-pl1-x644.google.com [IPv6:2607:f8b0:4864:20::644]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) server-signature RSA-PSS (4096 bits) client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1O1" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 46Y54Z1W9pz3H0p for ; Wed, 18 Sep 2019 03:26:53 +0000 (UTC) (envelope-from rrs@netflix.com) Received: by mail-pl1-x644.google.com with SMTP id x6so961094plv.6 for ; Tue, 17 Sep 2019 20:26:53 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=HOt8BPQrv3WKG/Zpe3yRat4EFQ5/HIG1xsKQugU9ftE=; b=GGVV01qxPER4nyLi/SeQ20u5lsKg88Fw/hYlYvHd8it6hWpuPzbyuSpQH6LrYPksXY Nysdm412yCnQbqt3oiqQQF4N5NRBTR1XHlO/2Ak91i2tHKizpvTvPQYUX1tSozR7tcLb z1c3tW+eHUVAZZFaRD9JZtMWxGESBeXIG6NpXsgThDrmVk+SvrRHC55J1e3Ar6SdtpkZ +8U7xDEmzsuSqDpbFmkyTZ2mfibzCEsI7d7TpO/f1Q0sUG6zu8Em1GFQmkTrZWw01Fqa Lnxj4kq1Bc5FYMlNltJ7X0mG4k938RmQuWfTQUtDhTCk1t0gsPno87tbF5Fxh+wzsDcR OAJQ== X-Gm-Message-State: APjAAAVb27IINoOLFX5G56W1RlIBkivYtMqshXJKkP8p+IMJMhTTsoCi bbmlUGJLA961+9r5Jy0rjFCLUg== X-Google-Smtp-Source: APXvYqxMhuxff2Hk5q8Cmo4sMz6wRWV+IArTTny9H5rpitAGZWBlLAtGz85KDMyU4YRelL+EVHAAEg== X-Received: by 2002:a17:902:6b84:: with SMTP id p4mr2068924plk.168.1568777212204; Tue, 17 Sep 2019 20:26:52 -0700 (PDT) Received: from ?IPv6:2607:fb10:7021:1::c2ea? ([2607:fb10:7021:1::c2ea]) by smtp.gmail.com with ESMTPSA id u65sm4493951pfu.104.2019.09.17.20.26.50 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 17 Sep 2019 20:26:51 -0700 (PDT) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 11.5 \(3445.9.1\)) Subject: Re: BBR patches? From: Randall Stewart In-Reply-To: Date: Tue, 17 Sep 2019 20:26:49 -0700 Cc: Jamie Landeg-Jones , Ryan Stone , freebsd-net Content-Transfer-Encoding: quoted-printable Message-Id: <84C20AED-44B1-4673-A29E-1A6FA61D8F27@netflix.com> References: <201909091232.x89CWl4b047025@donotpassgo.dyslexicfish.net> <27AD43DA-3237-490A-B3A6-C1792F9674E5@netflix.com> <5B630235-ABC7-4122-9082-F59BCCCDE14A@netflix.com> To: vm finance X-Mailer: Apple Mail (2.3445.9.1) X-Rspamd-Queue-Id: 46Y54Z1W9pz3H0p X-Spamd-Bar: ---- X-Spamd-Result: default: False [-4.06 / 15.00]; RCVD_VIA_SMTP_AUTH(0.00)[]; R_SPF_ALLOW(0.00)[+ip6:2607:f8b0:4000::/36]; MV_CASE(0.50)[]; RCVD_COUNT_THREE(0.00)[3]; TO_DN_ALL(0.00)[]; DKIM_TRACE(0.00)[netflix.com:+]; DMARC_POLICY_ALLOW(0.00)[netflix.com,reject]; FREEMAIL_TO(0.00)[gmail.com]; FROM_EQ_ENVFROM(0.00)[]; SUBJECT_ENDS_QUESTION(1.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:15169, ipnet:2607:f8b0::/32, country:US]; MID_RHS_MATCH_FROM(0.00)[]; ARC_NA(0.00)[]; NEURAL_HAM_MEDIUM(-0.96)[-0.964,0]; R_DKIM_ALLOW(0.00)[netflix.com:s=google]; FROM_HAS_DN(0.00)[]; RCPT_COUNT_THREE(0.00)[4]; NEURAL_HAM_LONG(-0.99)[-0.995,0]; TAGGED_RCPT(0.00)[]; PREVIOUSLY_DELIVERED(0.00)[freebsd-net@freebsd.org]; MIME_GOOD(-0.10)[text/plain]; BAD_REP_POLICIES(0.10)[]; TO_MATCH_ENVRCPT_SOME(0.00)[]; WHITELIST_DMARC(-7.00)[netflix.com:D:+]; RCVD_IN_DNSWL_NONE(0.00)[4.4.6.0.0.0.0.0.0.0.0.0.0.0.0.0.0.2.0.0.4.6.8.4.0.b.8.f.7.0.6.2.list.dnswl.org : 127.0.5.0]; URIBL_BLACK(7.50)[osboxes.org.multi.uribl.com]; WHITELIST_SPF_DKIM(-3.00)[netflix.com:d:+,netflix.com:s:+]; RCVD_TLS_ALL(0.00)[]; IP_SCORE(-1.10)[ip: (-0.55), ipnet: 2607:f8b0::/32(-2.69), asn: 15169(-2.23), country: US(-0.05)] X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 18 Sep 2019 03:26:55 -0000 There have been several patches pre-this one that provide the infrastructure to support BBR. Release 12.0 will *not* have these patches and will *not* compile it. I have no intention at this point in doing a MFC of this work.. so if = you want to run BBR you need to run Head R > On Sep 17, 2019, at 7:26 PM, vm finance wrote: >=20 > Hi Randall, >=20 > Could you please provide a pointer to the latest patch. I had applied = the one you published Sep-10. >=20 > Following is what I have done: > 1. Picked FreeBSD 12.0 RELEASE VM from osboxes.org > 2. Got VM up under VMPlayer on x86 laptop > 3. Checked out latest codebase from freebsd repo: > svn co svn://svn.freebsd.org/base/head /usr/src* > cd /usr/src; > make buildworld buildkernel =20 > [this is where make fails when patch is applied] >=20 > I could try to build with your latest patch - please send me a = pointer. >=20 > Thanks for your help! >=20 >=20 > On Tue, Sep 17, 2019 at 1:33 PM Randall Stewart = wrote: > Looking at your make file log I can=E2=80=99t really tell what you are = doing. >=20 > Its not the BBR or Rack code that is blowing up=E2=80=A6=20 >=20 > Are you cross compiling? >=20 > I have done the old fashioned kernel make >=20 > i.e.=20 >=20 > cd src/sys/amd64/config > config headvm > cd ../compile/headvm >=20 > make cleandepend ; make depend; make -j3 >=20 > I have done >=20 > cd src > make buildkernel KERNCONF=3Dheadvm >=20 > And even >=20 > make buildkernel KERNCONF=3DGENERIC >=20 > to make sure that the build works without bbr. >=20 > I attach my headvm config.. >=20 > What exactly are you building and how? kernel config too please? >=20 > R >=20 >=20 >=20 >=20 > > On Sep 17, 2019, at 1:11 PM, Randall Stewart = wrote: > >=20 > > looking > >=20 > > I was at 352408.. let me update and try it > >=20 > > R > >=20 > >> On Sep 17, 2019, at 1:10 PM, Randall Stewart = wrote: > >>=20 > >> Hmm > >>=20 > >> Did you get the patch I updated too this am? > >>=20 > >> I have built it both with and without the bbr stack and had no = issue.. there was > >> an issue with KTLS before the update though. > >>=20 > >> I don=E2=80=99t recognize what you have below there though=E2=80=A6 > >>=20 > >> R > >>=20 > >>> On Sep 17, 2019, at 11:47 AM, vm finance = wrote: > >>>=20 > >>> Got it - thank you! > >>>=20 > >>> btw, I tried to build the patch but its giving an error. Following = is SVN info + make error mesg. > >>>=20 > >>> Please let me know what am I missing here? > >>>=20 > >>> Thanks! > >>>=20 > >>> #svnlite revision > >>> Path: . > >>> Working Copy Root Path: /usr/src > >>> URL: svn://svn.freebsd.org/base/head > >>> Relative URL: ^/head > >>> Repository Root: svn://svn.freebsd.org/base > >>> Repository UUID: ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f > >>> Revision: 352436 > >>> Node Kind: directory > >>> Schedule: normal > >>> Last Changed Author: jah > >>> Last Changed Rev: 352434 > >>> Last Changed Date: 2019-09-17 03:39:31 +0000 (Tue, 17 Sep 2019) > >>>=20 > >>> root@osboxes:/usr/src # > >>>=20 > >>> -------------------------------------------- snip = ---------------------------- > >>>=20 > >>> ad_elf64_obj.llo reloc_elf64.llo disk.llo part.llo vdisk.llo = dev_net.llo bcache.llo interp_simple.llo zfs_cmd.llo > >>> rm -f .depend .depend.* GPATH GRTAGS GSYMS GTAGS > >>> =3D=3D=3D> sys (cleandir) > >>> rm -f export_syms machine x86 tcp_bbr.ko tcp_bbr.kld bbr.o = sack_filter.o rack_bbr_common.o opt_inet.h opt_inet6.h opt_ipsec.h = opt_tcpdebug.h opt_kern_tls.h > >>> rm: x86: is a directory > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[3]: stopped in /usr/src/sys > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[2]: stopped in /usr/src > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[1]: stopped in /usr/src > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make: stopped in /usr/src > >>>=20 > >>>=20 > >>> On Tue, Sep 17, 2019 at 6:41 PM vm finance = wrote: > >>> Got it - thank you! > >>>=20 > >>> btw, I tried to build the patch but its giving an error. Following = is SVN info + make error mesg. > >>> I have also attached the entire build log...snippet is below > >>>=20 > >>> Please let me know what am I missing here? > >>>=20 > >>> Thanks! > >>>=20 > >>> #svnlite revision > >>> Path: . > >>> Working Copy Root Path: /usr/src > >>> URL: svn://svn.freebsd.org/base/head > >>> Relative URL: ^/head > >>> Repository Root: svn://svn.freebsd.org/base > >>> Repository UUID: ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f > >>> Revision: 352436 > >>> Node Kind: directory > >>> Schedule: normal > >>> Last Changed Author: jah > >>> Last Changed Rev: 352434 > >>> Last Changed Date: 2019-09-17 03:39:31 +0000 (Tue, 17 Sep 2019) > >>>=20 > >>> root@osboxes:/usr/src # > >>>=20 > >>> -------------------------------------------- snip = ---------------------------- > >>>=20 > >>> ad_elf64_obj.llo reloc_elf64.llo disk.llo part.llo vdisk.llo = dev_net.llo bcache.llo interp_simple.llo zfs_cmd.llo > >>> rm -f .depend .depend.* GPATH GRTAGS GSYMS GTAGS > >>> =3D=3D=3D> sys (cleandir) > >>> rm -f export_syms machine x86 tcp_bbr.ko tcp_bbr.kld bbr.o = sack_filter.o rack_bbr_common.o opt_inet.h opt_inet6.h opt_ipsec.h = opt_tcpdebug.h opt_kern_tls.h > >>> rm: x86: is a directory > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[3]: stopped in /usr/src/sys > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[2]: stopped in /usr/src > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make[1]: stopped in /usr/src > >>> *** Error code 1 > >>>=20 > >>> Stop. > >>> make: stopped in /usr/src > >>>=20 > >>>=20 > >>> On Tue, Sep 17, 2019 at 6:27 PM Randall Stewart = wrote: > >>> Pacing is provided by tcp_hpts.c. The current linux patches do not = have > >>> to have fq.. they built an alternate means of doing pacing into = bbr. > >>>=20 > >>> In either case our testing has shown that our pacing is more = accurate than > >>> either fq or the internal pacer :) > >>>=20 > >>> R > >>>=20 > >>>> On Sep 17, 2019, at 11:05 AM, vm finance = wrote: > >>>>=20 > >>>> Thanks Randall. > >>>> I was able to apply the patch - now rebuilding the kernel. Would = update on how it goes. > >>>>=20 > >>>> BTW, is there any description on how lack of tc_fq under FreeBSD = is compensated here? > >>>> The original BBR patches on Linux show that as a must-have? Is = that functionality implemented via tcp_ratelimit.[ch]? > >>>>=20 > >>>> Any pointers to existing presentations/discussions highly = appreciated? > >>>>=20 > >>>> Thanks a lot. > >>>>=20 > >>>> On Tue, Sep 17, 2019 at 5:39 AM Randall Stewart = wrote: > >>>> You should be able to compile it against the current head. I = re-doing that now (had an > >>>> issue with my machine and had to roll it back to a backup). > >>>>=20 > >>>> When I put the patch up on Sept 10th it complied with and without = BBR on whatever > >>>> was that rev.. > >>>>=20 > >>>> Looking in the commit logs that would have been around 352191 > >>>>=20 > >>>> Hope that helps > >>>>=20 > >>>> R > >>>>=20 > >>>>> On Sep 17, 2019, at 3:32 AM, vm finance = wrote: > >>>>>=20 > >>>>> Hi Randall, > >>>>>=20 > >>>>> Thanks for releasing BBR patch: > >>>>> https://reviews.freebsd.org/D21582#change-xcAWBif3E9Jq=20 > >>>>>=20 > >>>>> Could you please let me know what SVN/GIT label tag this is = based on? I would like to patch and experiment with it. I couldn't find = this info in the released patch. > >>>>>=20 > >>>>> Thanks a lot!=20 > >>>>>=20 > >>>>> On Tue, Sep 10, 2019 at 10:26 AM Ryan Stone = wrote: > >>>>> rrs@ has just posted the BBR patch to phabricator: > >>>>>=20 > >>>>> https://reviews.freebsd.org/D21582 > >>>>=20 > >>>> ------ > >>>> Randall Stewart > >>>> rrs@netflix.com > >>>>=20 > >>>>=20 > >>>>=20 > >>>=20 > >>> ------ > >>> Randall Stewart > >>> rrs@netflix.com > >>>=20 > >>>=20 > >>>=20 > >>=20 > >> ------ > >> Randall Stewart > >> rrs@netflix.com > >>=20 > >>=20 > >>=20 > >=20 > > ------ > > Randall Stewart > > rrs@netflix.com > >=20 > >=20 > >=20 >=20 > ------ > Randall Stewart > rrs@netflix.com >=20 >=20 >=20 ------ Randall Stewart rrs@netflix.com