[PATCH v7 12/41] KVM: arm64: gic-v5: Sanitize ID_AA64PFR2_EL1.GCIE

[Marc Zyngier]

On Tue, 24 Mar 2026 12:47:24 +0000,
Mark Brown <broonie at kernel.org> wrote:
> 
> [1  <text/plain; us-ascii (7bit)>]
> On Thu, Mar 19, 2026 at 03:52:50PM +0000, Sascha Bischoff wrote:
> > Add in a sanitization function for ID_AA64PFR2_EL1, preserving the
> > already-present behaviour for the FPMR, MTEFAR, and MTESTOREONLY
> > fields. Add sanitisation for the GCIE field, which is set to IMP if
> > the host supports a GICv5 guest and NI, otherwise.
> 
> We're also seeing an issue with this in the aarch32_id_regs test:
> 
> # selftests: kvm: aarch32_id_regs
> # Random seed: 0x6b8b4567
> # ==== Test Assertion Failure ====
> #   arm64/aarch32_id_regs.c:25: read_sysreg_s((((3) << 19) | ((0) << 16) | ((0) << 12) | ((1) << 8) | ((1) << 5))) == 0
> #   pid=3111 tid=3111 errno=4 - Interrupted system call
> # sh: 1: addr2line: not found
> #   0x10000000 != 0x0 (read_sysreg_s((((3) << 19) | ((0) << 16) | ((0) << 12) | ((1) << 8) | ((1) << 5))) != 0)
> not ok 9 selftests: kvm: aarch32_id_regs # exit=254
> 
> which for some reason only manifests on TX2 of the platforms I've
> seen.

Is your TX2 the only machine you have that is AArch64 only at all ELs?

	M.

-- 
Without deviation from the norm, progress is not possible.