From nobody Thu Dec 2 16:03:11 2021 X-Original-To: freebsd-virtualization@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 9489518AAD9D for ; Thu, 2 Dec 2021 16:03:26 +0000 (UTC) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: from gndrsh.dnsmgr.net (br1.CN84in.dnsmgr.net [69.59.192.140]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 4J4gjR0cCpz3vQD; Thu, 2 Dec 2021 16:03:22 +0000 (UTC) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: from gndrsh.dnsmgr.net (localhost [127.0.0.1]) by gndrsh.dnsmgr.net (8.13.3/8.13.3) with ESMTP id 1B2G3DRZ049479; Thu, 2 Dec 2021 08:03:13 -0800 (PST) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: (from freebsd-rwg@localhost) by gndrsh.dnsmgr.net (8.13.3/8.13.3/Submit) id 1B2G3B9U049478; Thu, 2 Dec 2021 08:03:11 -0800 (PST) (envelope-from freebsd-rwg) From: "Rodney W. Grimes" Message-Id: <202112021603.1B2G3B9U049478@gndrsh.dnsmgr.net> Subject: Re: bhyve vCPU limit In-Reply-To: To: bsdlists@jld3.net Date: Thu, 2 Dec 2021 08:03:11 -0800 (PST) CC: Oleg Ginzburg , Miroslav Lachman <000.fbsd@quip.cz>, rgrimes@FreeBSD.org, jbo@insane.engineer, freebsd-virtualization@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL121h (25)] List-Id: Discussion List-Archive: https://lists.freebsd.org/archives/freebsd-virtualization List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-virtualization@freebsd.org X-BeenThere: freebsd-virtualization@freebsd.org MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII X-Rspamd-Queue-Id: 4J4gjR0cCpz3vQD X-Spamd-Bar: ---- Authentication-Results: mx1.freebsd.org; none X-Spamd-Result: default: False [-4.00 / 15.00]; REPLY(-4.00)[] X-ThisMailContainsUnwantedMimeParts: N > On Wed 2021-12-01 11:13 AM MST -0700, wrote: > > > As far as I remember, it is not enough to simply change one parameter > > in .h file: https://bhyvecon.org/bhyveconOttawa2019-Rodney.pdf > > I may be wrong but my understanding is that prior to the work described > in that presentation, some of the bhyve data structures were such that > vCPU counts above 21 would not be possible. That work resolved that but > the VM_MAXCPU value was left at 16. That is correct, the work described was committed, and raised the usable limit of VM_MAXCPU well into the 200 range. Note that raising this ABOVE the physcial core count is neither desirable or performant. > > On my system, I'm sure that all I did was edit vmm.h, make buildworld, > and make installworld. Yes, that would do all that is needed. There well be some ipi interrupt stats that dont work right, but that is the only known issue. > > I did some other stuff while sort of stumbling toward that eventual > solution, described here: > > > > Once what I was doing wrong was pointed out, it seemed to me that after > editing that file, "make buildworld" and "make installworld" would be > the simplest way to do things, so that's what I did. It's been working > fine for six months or so. In this message of that thread: https://lists.freebsd.org/pipermail/freebsd-virtualization/2021-May/009510.html I correct the error that caused your procedure to not work and explain why it did not work (compiled with old /usr/include file). -- Rod Grimes rgrimes@freebsd.org