Hi Sven, as tom said on IRC - to some extend it is that Bionic didn't have a Rome chip definition yet, so the changes to add it to Rome do not 100% make sense there.
You have as part of similar fixes there already: 1. all kind of intel chips which were faster to provide that for 2.11 (in the stable tree of qemu) root@b:~# qemu-system-x86_64 -cpu ? | grep -i ibrs x86 Broadwell-IBRS Intel Core Processor (Broadwell, IBRS) x86 Broadwell-noTSX-IBRS Intel Core Processor (Broadwell, no TSX, IBRS) x86 Haswell-IBRS Intel Core Processor (Haswell, IBRS) x86 Haswell-noTSX-IBRS Intel Core Processor (Haswell, no TSX, IBRS) x86 IvyBridge-IBRS Intel Xeon E3-12xx v2 (Ivy Bridge, IBRS) x86 Nehalem-IBRS Intel Core i7 9xx (Nehalem Core i7, IBRS update) x86 SandyBridge-IBRS Intel Xeon E312xx (Sandy Bridge, IBRS update) x86 Skylake-Client-IBRS Intel Core Processor (Skylake, IBRS) x86 Skylake-Server-IBRS Intel Xeon Processor (Skylake, IBRS) x86 Westmere-IBRS Westmere E56xx/L56xx/X56xx (IBRS update) 2. a generic EPYC chip which got what AMD first pushed for that was IBPB (mind all the characters - not the same) root@b:~# qemu-system-x86_64 -cpu ? | grep -i epy x86 EPYC-IBPB AMD EPYC Processor (with IBPB) x86 EPYC AMD EPYC Processor I rechecked if e.g. upstream stable of 2.11 since then changed things, but there is nothing in this regard. Nor did I see (in a very quick not thorough check) anyone else backporting it that far. So you might say "can't I use shiny new chips and all of their feature with Bionic :-/", but for some applications stacks this is a common issue (the balance between stability vs staying new). And for those there often is a way between the obvious first recommendation "use a newer LTS and you get the newer bits" in the form of https://wiki.ubuntu.com/OpenStack/CloudArchive or https://launchpad.net/~canonical-server/+archive/ubuntu/server-backports that allow you to get some newer stacks for older LTSes. Back then when this bug came up these patches did neither make sense nor where in any way applicable to Bionic, one could take some more time to deeply re-evaluate this, but as I said above there are options already ...; I'm tempted to say that this should be spawned into a separate bug then to keep this one as-is reflecting only the efforts of the past. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1921754 Title: Add missing cpu feature bits in EPYC-Rome model To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1921754/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
