Date: Thu, 23 Apr 2020 15:24:57 -0700 From: Mark Millard <marklmi@yahoo.com> To: bob prohaska <fbsd@www.zefox.net> Cc: freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: Booting from USB on RPI3 Message-ID: <328B7083-1119-43F1-A6E4-61F1FB9E922E@yahoo.com> In-Reply-To: <20200423212614.GA3996@www.zefox.net> References: <mailman.61.1587470402.80084.freebsd-arm@freebsd.org> <trinity-4938b1d4-f29f-4907-bedd-65be21112e48-1587489497227@3c-app-webde-bs65> <20200421181224.GC96994@www.zefox.net> <trinity-19081201-3024-4046-817a-48321c51a515-1587587309088@3c-app-webde-bap64> <20200423162124.GA3583@www.zefox.net> <CAJuc1zMHL6ke3LH67iV8%2BogKFO18xwW_v7bRAUsj8CbxD2EOTw@mail.gmail.com> <8D1F6A8D-4910-4C1E-8EB5-2F5F89E31120@yahoo.com> <CAJuc1zOTnoRmeaKXHMTH9po=6ipBYDxvxAoiVmow-sNgwUERVQ@mail.gmail.com> <20200423212614.GA3996@www.zefox.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2020-Apr-23, at 14:26, bob prohaska <fbsd@www.zefox.net> wrote: > On Fri, Apr 24, 2020 at 08:32:54AM +1200, Jonathan Chen wrote: >> On Fri, 24 Apr 2020 at 08:22, Mark Millard <marklmi@yahoo.com> wrote: >> [...] >>> The RPi3 will not start to boot from a gpt partitioned >>> media. So picking gpt labeling as the example is somewhat >>> misleading for single-media booting. glabel based >>> labeling would be more realistic for the context. >> >> The OP is attempting to boot off an external USB drive via loader.env. >> So it's the external drive's partitioning system that is of interest. > > The USB drive on my system was written with dd using > FreeBSD-13.0-CURRENT-arm64-aarch64-RPI3-20200409-r359731.img > It's been resized to fill the hard disk and given a swap partition. > Gpart reports > bob@www:~ % gpart show da0 > => 63 78140097 da0 MBR (37G) > 63 2016 - free - (1.0M) > 2079 102312 1 fat32lba [active] (50M) > 104391 78035769 2 freebsd (37G) > > gpart show da0s2 > => 0 78035769 da0s2 BSD (37G) > 0 57 - free - (29K) > 57 71303168 1 freebsd-ufs (34G) > 71303225 6732544 2 freebsd-swap (3.2G) > > To my surprise, > gpart show da0s1 reports > gpart: No such geom: da0s1. > even though there is a /dev/da0s1. I thought it would at least report > the size. da0s1 is analogous ot da0s2a, not da0s2: # gpart show da0s2a gpart: No such geom: da0s2a. It is the "BSD" that makes da0s2 special, much like "MBR" being special for da0. Both MBR and BSD contain other slices. fat32lba and freebsd-ufs and freebsd-swap do not. That makes the "geom" distinction involved. > No attempt has been made to alter the msdos partition, > That does mean the u-boot on the USB drive isn't quite > the same as the u-boot on the microSD, which uses > https://github.com/raspberrypi/firmware/raw/master/boot/bootcode.bin > and modified to turn on its serial console. Could that matter? Since > it appears the USB installation is getting all the way to the loader > prompt on USB I'm tempted to think the trouble's on the FreeBSD side > of the USB setup. > Note: bootcode.bin may have fixes of RPi* problems for the boot code that is burned into the part involved. For USB-only or for microsd card also in use, have you tried the equivalent of: OK lsdev and then used the naming convention shown in the output for any loader commands/settings instead of using FreeBSD-stage da0 based naming? (You might want to report what lsdev shows.) Note: the disk0 and disk1 like prefixes are copied from the u-boot notations that the loader gets from u-boot, if I understand right. Does your loader.conf do anything to control what is booted? As Joanathan asked: what does your /etc/fstab have listed? (/etc/fstab should have FreeBSD notations, not loader device names.) Are there any other boot-contributing files that would be appropriate to show together in one message for review, including any on the msdos file system? === 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?328B7083-1119-43F1-A6E4-61F1FB9E922E>