From 82cdc0a2e5054c5dd3037055ab9a0cc1ac8b7ef6 Mon Sep 17 00:00:00 2001 From: Luke Curley Date: Thu, 29 Oct 2026 16:05:23 -0700 Subject: [PATCH 1/2] chore(quest): plan refusing VM checks without KVM Co-Authored-By: Claude Opus 5.5 --- quest/m0/ci/README.md | 1 + quest/m0/ci/vm-kvm.md | 21 +++++++++++++++++++++ 2 files changed, 22 insertions(+) create mode 100644 quest/m0/ci/vm-kvm.md diff --git a/quest/m0/ci/README.md b/quest/m0/ci/README.md index e219161897e..400235e8d52 100644 --- a/quest/m0/ci/README.md +++ b/quest/m0/ci/README.md @@ -138,3 +138,4 @@ The historical samples locate candidates; they are not controlled benchmarks. - [tracing-test filter](/quest/m0/ci/tracing-filter.md) - moq-tokio's tests stop forcing every crate to format at TRACE - [Stats oversize](/quest/m0/ci/stats-oversize.md) - moq-stats refuses an oversized plain snapshot before parsing it back - [Cargo aliases in nested checkouts](/quest/m0/ci/cargo-alias-strings.md) - `cargo build-ffi` and `cargo moq` work from a worktree nested inside another checkout +- [VM checks refuse a host without KVM](/quest/m0/ci/vm-kvm.md) - `check flake` fails fast with a clear message instead of an emulated-boot timeout diff --git a/quest/m0/ci/vm-kvm.md b/quest/m0/ci/vm-kvm.md new file mode 100644 index 00000000000..f6c46a35610 --- /dev/null +++ b/quest/m0/ci/vm-kvm.md @@ -0,0 +1,21 @@ +# [XS] VM checks refuse a host without KVM + +## Goal + +`just ci`'s `check flake` fails at once with a message naming the missing +`/dev/kvm` when it would run the NixOS VM tests (such as +`vm-test-run-github-runner-vm`) on a host without KVM, instead of booting the +VMs under software emulation until they time out with "Shell did not start +in time". + +## Plan + +Decided by the maintainer on 2026-10-29, from the follow-ups of +[moq-dev/moq#7175](https://github.com/moq-dev/moq/pull/7175), whose local gate +spent about 19 minutes before failing this way on a host without KVM. + +- Refuse, don't skip: a skipped check would let a local gate read as passing. + The message says the check needs KVM and that CI runs it. +- The other follow-up from #7175, a nested Codex shell's `BASH_ENV` snapshot + inflating `PATH` past mbx's exec limit, is reported upstream to Codex + rather than planned here. From 51cfce025fa840e40a54fa8e4215581cfee51a77 Mon Sep 17 00:00:00 2001 From: Luke Curley Date: Thu, 29 Oct 2026 16:06:31 -0700 Subject: [PATCH 2/2] chore(quest): the Codex PATH growth did not reproduce Co-Authored-By: Claude Opus 5.5 --- quest/m0/ci/vm-kvm.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/quest/m0/ci/vm-kvm.md b/quest/m0/ci/vm-kvm.md index f6c46a35610..1849bc677e9 100644 --- a/quest/m0/ci/vm-kvm.md +++ b/quest/m0/ci/vm-kvm.md @@ -17,5 +17,6 @@ spent about 19 minutes before failing this way on a host without KVM. - Refuse, don't skip: a skipped check would let a local gate read as passing. The message says the check needs KVM and that CI runs it. - The other follow-up from #7175, a nested Codex shell's `BASH_ENV` snapshot - inflating `PATH` past mbx's exec limit, is reported upstream to Codex - rather than planned here. + inflating `PATH` past mbx's exec limit, did not reproduce with codex-cli + 0.162.1 (`BASH_ENV` unset, `PATH` stable across nested shells), so it is + neither planned nor reported.