Re: Installer loop on apu6b4
On Wed, Aug 2, 2023 at 9:04 PM Chris Cappuccio wrote: > > This should be an FAQ entry. > > You need to setup the serial console in the boot blocks: > boot> stty com0 115200 > boot> set tty com0 > > Chris That solved the problem. Thank you, Chris. Apologies to the list. > > -- RD
Re: Installer loop on apu6b4
Ronald Dahlgren [ronald.dahlg...@gmail.com] wrote: > I have a PC Engines apu6b4 that is acting up during the installation > process. I want to see if this group has any insight before I try an RMA > with the seller. > > entry point at 0x8100100PC Engines apu6 This should be an FAQ entry. You need to setup the serial console in the boot blocks: boot> stty com0 115200 boot> set tty com0 Chris
Installer loop on apu6b4
I have a PC Engines apu6b4 that is acting up during the installation process. I want to see if this group has any insight before I try an RMA with the seller. The apu6b4 (https://www.pcengines.ch/apu6b4.htm) runs an AMD processor and has the standard ports. It uses a micro-USB serial port rather than an RS-232 interface. When loading the 7.3 release installer, I get to a point where "entry point at 0x8100100" is printed before the device loads the BIOS and starts the POST / boot process again. There isn't much output to use for clues. I run current on an apu2, so I'm wondering if it's the apu6b4 model generally or just my example that is the problem. For comparison, I tried loading Debian stable as well and found that it got partially into the text installer before appearing to hang when I selected a "video" mode. Note that I've never done a serial console installation of Debian. It may be expected behavior. Does anyone have guidance on how to proceed with investigating the problem? Has anyone run OpenBSD on the apu6b4? Here are some relevant screenshots: https://star.sw.gy/pub/boot-loop-1.png https://star.sw.gy/pub/boot-loop-2.png https://star.sw.gy/pub/debian-installer-1.png https://star.sw.gy/pub/debian-installer-2.png Here's the captured output from the boot loop I mentioned: SeaBIOS (version rel-1.12.1.3-0-g300e8b70) Press F10 key now for boot menu Booting from Hard Disk... Using drive 0, partition 3. Loading.. probing: pc0 com0 com1 mem[639K 3325M 752M a20=on] disk: hd0+ hd1+* >> OpenBSD/amd64 BOOT 3.55 boot> cannot open hd0a:/etc/random.seed: No such file or directory booting hd0a:/7.3/amd64/bsd.rd: 3924676+1647616+3886216+0+704512 [109+440424+293 778]=0xa667f0 entry point at 0x8100100PC Engines apu6 coreboot build 20202509 BIOS version v4.12.0. -- RD
Re: ksh bug or just normal behaviour?
On Wed, Aug 02, 2023 at 12:14:51PM +, Thomas Schweikle wrote:
>
>
> Am Mi., 02.Aug..2023 um 13:45:26 schrieb Peter N. M. Hansteen:
> > On Wed, Aug 02, 2023 at 11:35:39AM +, Ioan Samarul wrote:
> > > Can you please tell me if this is a bug or it is considered normal?
> > >
> > > $ set -A test a b c d e f g h i
> > > $ echo ${test[07]}
> > > h
> > > $ echo ${test[08]}
> > > ksh: 08: bad number `08'
> > > $ echo ${test[8]}
> > > i
> >
> > I strongly suspect you stumbled on to a case of the old convention
> > "numerals with
> > leading zeroes are interpreted as octal notation" (but do check the
> > underlying
> > code to make sure).
>
> Yes, that is it. It is considered octal notation.
And here is the proof:
$ echo ${test[010]}
i
Re: Installing openBSD
On Wed, Aug 2, 2023 at 7:31 AM David Demelier wrote: > On Tue, 2023-08-01 at 01:00 +0800, ykla wrote: > > Actually, I think it's a bug that OpenBSD cannot create EFI > > partitions manually. > > I've installed OpenBSD many times in dualboot with linux (for some > things we can't do right now on OpenBSD such as ESP32 development). And > my take is to install Linux first, actually quite simple to go for a > dual boot afterwards. I go back to linux and configure the bootloader > (i.e. efibootmgr, grub, whatever you like). > > macOS and Windows are not really friendly in that area either. I can't > blame an OS to not spend effort for those topics, being able to boot > any OS from EFI is already good enough. > > -- > David > > Note that multiboot with refind is trivial (UEFI only), it automatically detects windows, linux and openbsd. Just install them in that order, and refind at the end.
Re: unhibernate failed: original kernel changed
On Tue, Aug 01, 2023 at 07:22:04AM -, Piotr Isajew wrote:
> Dnia 31.07.2023 Mike Larkin napisał/a:
>
> > The message explained exactly what happened. What is unclear?
>
> I understand the message. What I don't undestand is the reason
> for it. The message is due to this comparison not returning 0:
>
> if (bcmp(mine->kern_hash, disk->kern_hash, SHA256_DIGEST_LENGTH) != 0) {
>
> but the same kernel image was used when booting the system before
> hibernation and on unhibernate.
>
Something changed on disk otherwise that bcmp would be the same.
Try reproing with a GENERIC/GENERIC.MP kernel and see if that fixes it.
Re: ksh bug or just normal behaviour?
Am Mi., 02.Aug..2023 um 13:45:26 schrieb Peter N. M. Hansteen:
On Wed, Aug 02, 2023 at 11:35:39AM +, Ioan Samarul wrote:
Can you please tell me if this is a bug or it is considered normal?
$ set -A test a b c d e f g h i
$ echo ${test[07]}
h
$ echo ${test[08]}
ksh: 08: bad number `08'
$ echo ${test[8]}
i
I strongly suspect you stumbled on to a case of the old convention "numerals
with
leading zeroes are interpreted as octal notation" (but do check the underlying
code to make sure).
Yes, that is it. It is considered octal notation.
--
Thomas
OpenPGP_0x27AE2304B4974851.asc
Description: OpenPGP public key
Re: Installing openBSD
On Tue, 2023-08-01 at 01:00 +0800, ykla wrote: > Actually, I think it's a bug that OpenBSD cannot create EFI > partitions manually. I've installed OpenBSD many times in dualboot with linux (for some things we can't do right now on OpenBSD such as ESP32 development). And my take is to install Linux first, actually quite simple to go for a dual boot afterwards. I go back to linux and configure the bootloader (i.e. efibootmgr, grub, whatever you like). macOS and Windows are not really friendly in that area either. I can't blame an OS to not spend effort for those topics, being able to boot any OS from EFI is already good enough. -- David
Re: ksh bug or just normal behaviour?
On Wed, Aug 02, 2023 at 11:35:39AM +, Ioan Samarul wrote:
> Can you please tell me if this is a bug or it is considered normal?
>
> $ set -A test a b c d e f g h i
> $ echo ${test[07]}
> h
> $ echo ${test[08]}
> ksh: 08: bad number `08'
> $ echo ${test[8]}
> i
I strongly suspect you stumbled on to a case of the old convention "numerals
with
leading zeroes are interpreted as octal notation" (but do check the underlying
code to make sure).
- Peter
--
Peter N. M. Hansteen, member of the first RFC 1149 implementation team
https://bsdly.blogspot.com/ https://www.bsdly.net/ https://www.nuug.no/
"Remember to set the evil bit on all malicious network traffic"
delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
ksh bug or just normal behaviour?
Hello to you all!
Can you please tell me if this is a bug or it is considered normal?
$ set -A test a b c d e f g h i
$ echo ${test[07]}
h
$ echo ${test[08]}
ksh: 08: bad number `08'
$ echo ${test[8]}
i
Thank you all!
