On Wed, Apr 05, 2023 at 01:30:19PM +0200, Michal Privoznik wrote:
In my previous commit v9.2.0-rc1~3 I've made virt-host-validate
You're including the commit hash below in the Fixes: pseudo-header, so mentioning it here again feels unnecessary. But it's fine if you prefer keeping this.
to report host IOMMU check pass if IORT table is present. This is not sufficient though, because IORT describes much more than just IOMMU (well, it's called SMMU in ARM world). In fact, this can be seen in previous commit which adds test cases: there are tables (IORT_virt_aarch64) which does not contain any SMMU records.
But after previous commits, we can parse the table so switch to that.
Fixes: 2c13a2a7c9c368ea81eccd4ba12d9cf34bdd331b
Personally I've grown to quite like the alternative take on this pseudo-header Fixes: 2c13a2a ("virt-host-validate: Detect SMMU support on ARMs") as seen very frequently in QEMU and other projects. I feel that it's more immediately informative than just the raw hash. But your version is perfectly okay too. Please include a reference to https://bugzilla.redhat.com/show_bug.cgi?id=2178885 though, since this is finishing the fix for that bug. Regardless of whether you decide to follow any of the suggestions above, Reviewed-by: Andrea Bolognani <abologna@redhat.com> -- Andrea Bolognani / Red Hat / Virtualization