I forgot just how crap pci ide controllers and disks can be.
I installed using Qemu 0.9 and having it drive the disk directly. Â Naturally, MS-DOS works fine.
I forgot just how crap pci ide controllers and disks can be.
I installed using Qemu 0.9 and having it drive the disk directly. Â Naturally, MS-DOS works fine.
You must be logged in to post a comment.
Is thero no SCSI emulation available in qemu for PC platforms?
There is, I haven’t tried it with later versions. The P4 I was transplanting this to doesn’t have a SCSI controller, and I don’t have any SCSI disks..
(E)IDE drivers from Nextstep don’t like everything that’s younger than Pentium II.
I thought it was more so to load the ‘beta’ drivers, the Dual channel Intel PII chipset or something like that one.. . I’ve used it on a P4 board in the past, although obviously not a 40GB disk jumpered to support only ‘8GB’ .. I may have to see how tricky it is to try other drivers.
I should add, it’s basically the install process I’ve been using for years as I wrote down here, on bsdnexus.com, before I started my blog.
I do things like this with VirtualBox, raw disk access works pretty well. I’m pretty sure I installed NeXTSTEP once like this – although I didn’t use a “real” disk, I installed it on a CF-Card that I accessed via raw disk access.
Cheers,
-F
I’m just not sure if it doesn’t like my ide controllers or my disk…
This reminds me of the strange behavior exhibited by modern versions of qemu. I installed NeXTSTEP 3.3 on qemu 2.7.0 and the same ‘interrupt timeout’ errors. In that case, it was that edge triggered interrupts weren’t being handled properly. It was fixed by applying the first half of this patch: https://patchwork.ozlabs.org/patch/208252/
Maybe whatever Intel did in this i8xx chipset interrupt controller wasn’t standard?
It could be a hint, that is for sure! I never thought my old p4 ‘white box’ was such a special thing as it installed NS without any issues, but obviously it was.