Re: [PATCH v5 20/21] KVM: selftests: Make number of vCPUs configurable in IRQ test

[Sean Christopherson <seanjc@google.com>]

On Thu, Jun 04, 2026, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] Missing bounds check on user-supplied `nr_vcpus` allows stack buffer overflows and out-of-bounds global array accesses.
> --
> 
> commit 115e22303a8d11509c805d6bfdcdbd39b670c16d
> Author: David Matlack <dmatlack@google.com>
> 
> KVM: selftests: Make number of vCPUs configurable in IRQ test
> 
> This commit adds a '-v' flag to tools/testing/selftests/kvm/irq_test.c,
> allowing users to configure the number of vCPUs to run in the test.
> 
> > diff --git a/tools/testing/selftests/kvm/irq_test.c b/tools/testing/selftests/kvm/irq_test.c
> > index 1d66293b6d0b8..41c50caa928e2 100644
> > --- a/tools/testing/selftests/kvm/irq_test.c
> > +++ b/tools/testing/selftests/kvm/irq_test.c
> [ ... ]
> > @@ -213,6 +214,9 @@ int main(int argc, char **argv)
> >  		case 'n':
> >  			use_nmi = true;
> >  			break;
> > +		case 'v':
> > +			nr_vcpus = atoi_positive("Number of vCPUS", optarg);
> 
> [Severity: Medium]
> Does this code need a bounds check for nr_vcpus against KVM_MAX_VCPUS?

Yep.  It also needs to configure the test to enable x2APIC format and disable
KVM's broadcast quirk, otherwise running with 255+ vCPUs will fail due to KVM
blasting an IRQ to all vCPUs when trying to target vCPU 255.