Linux 5.7
On the Power ISA side, other than refactoring the exception code, the most noteworthy changes (to me) are support for fast reboot, which might fix that obnoxious issue I keep seeing with failing over to the serial port on regular reboots; a means for discovery of secure guests under KVM-HV so that those of you using the DD2.3 Ultravisor and the Protected Execution Facility can keep secure guests only on systems that support them; and one of particular historical interest, where compatibility with 32-bit PowerPC binaries is now disabled on ppc64le by default (it can still be enabled with a configuration flag). Big-endian ppc64 does still contain this support by default, presumably because 32-bit PowerPC binaries are more common on those systems, but to be sure, none of my little-endian Power systems have any 32-bit binaries on them. Do yours? It'll be interesting to hear what some of our boutique OpenPOWER distros will do with this going forward.
As a matter of fact, my ppc64le Blackbird is the staging ground for some experimentation with 32-bit compatibility. Fun fact: it hasn't worked out of the box for a long time :) To be more precise, it does work, but I don't think anyone considers "the shell dies after running a command and the date is wrong" to comprise a functional system.
ReplyDeleteAs it turns out, though, there's a neat little project called box86 that confers significant speed gains for x86 emulation upon anything with a 32-bit little-endian mode, hence the seemingly bizarre venture. I've been working with a couple others in #talos-workstation on freenode (including the voidlinux-ppc maintainer) and just posted to some mailing lists the other day to gather more feedback, so you can read the summary of the whole thing here: https://lists.ozlabs.org/pipermail/linuxppc-dev/2020-May/211669.html
I'm surprised that the compat setting on LE has been left enabled for this long, considering (afaik) it hasn't really been usable since 2006. A funny coincidence, too, since the aforementioned project has been in the background for a while before getting to a point where the path forward was clear (and then I had to gather enough focus to write an email to an unfamiliar group...). Very excited to see where it goes, though, especially if it solidifies enough to see some changes to the upstream kernel ^^
Wow, there's a real rabbit hole here. Interesting to hear of your experiments. I figured there had to be 32-bit LE systems out there, but I'd never run into one. I'm familiar with box86 but didn't look much further into it exactly *because* of the 32-bit requirement.
DeleteYour 2006 note threw me, since there wasn't OpenPOWER in 2006, but I see from your message what you're actually referring to. :)