Date: Fri, 7 Feb 2020 21:23:29 -0800 From: Mark Millard <marklmi@yahoo.com> To: bob prohaska <fbsd@www.zefox.net> Cc: freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: RPi3 not using SMP? Message-ID: <85FD469B-C87E-4D88-AFF4-BFA9219AE3F6@yahoo.com> In-Reply-To: <20200208042430.GA8903@www.zefox.net> References: <20200208011940.GA8570@www.zefox.net> <6B6CCB8F-B56A-4758-BEEC-6418718C95CB@yahoo.com> <9F1B762C-D1DA-40F6-A2D6-451B40A39E4A@yahoo.com> <20200208042430.GA8903@www.zefox.net>
next in thread | previous in thread | raw e-mail | index | archive | help
[I've dropped Jeff from this note not having additional evidence.] On 2020-Feb-7, at 20:24, bob prohaska <fbsd at www.zefox.net> wrote: > On Fri, Feb 07, 2020 at 07:10:58PM -0800, Mark Millard wrote: >>> >>> psci0: <ARM Power State Co-ordination Interface Driver> on ofwbus0 >>> psci0: PSCI version number 0 mismatched with DT, default 2 >>> device_attach: psci0 attach returned 6 >>> > > Is the "DT" in the boot message a reference to the Device Tree (blob)? > If not then my question is on the wrong foot. I meant to wonder if the > problem is simply using a wrong dtb file. > > It does seem clear that one or more things are somewhat amiss. > Are they adequately advertised in existing bug reports, or should > more noises be made? A search for psci on bugs.freebsd.org finds > https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=243118 > but to the untrained eye it seems unrelated. If you want a working system, use a system based on head -r356767 (and an appropriately matching world). No device tree change required. Nothing about -r356776 is supposed to involve determining which Device Tree is used. The changes are not to architecture specific code. But I'm not making claims about possibilities like an inaccurate Excluded Memory Regions list or the code ignoring that list in some way and allocating with an overlap with such a region. Mostly I'm looking at this in case there is a FreeBSD problem that the RPi3/4's just happen to expose. If it proves to be a RPi3/4 error and FreeBSD should be okay as it is in -r356776 and later, then I'm not sure there are many folks with the appropriate skills that would want to deal with making things work. But, if it turns out that FreeBSD has a problem that RPi3/4's just happen to expose, I'm sure folks with the skills would deal with that. I'm unlikely to figure out which it is. But some of the evidence I provide might prove useful. === Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?85FD469B-C87E-4D88-AFF4-BFA9219AE3F6>