[PATCH 2/2] KVM: arm64: Map hyp text as RO and dump instr on panic

[Mostafa Saleh]

Hi Ben,

On Fri, Jul 18, 2025 at 11:16:18AM +0100, Ben Horgan wrote:
> Hi Mostafa,
> 
> On 18/07/2025 00:47, Mostafa Saleh wrote:
> > Map the hyp text section as RO, there are no secrets there
> > and that allows the kernel extract info for debugging.
> > 
> > As in case of panic we can now dump the faulting instructions
> > similar to the kernel.
> > 
> > Signed-off-by: Mostafa Saleh <smostafa at google.com>
> > ---
> >   arch/arm64/kvm/handle_exit.c    |  4 +---
> >   arch/arm64/kvm/hyp/nvhe/setup.c | 12 ++++++++++--
> >   2 files changed, 11 insertions(+), 5 deletions(-)
> > 
> > diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
> > index de12b4d4bccd..d59f33c40767 100644
> > --- a/arch/arm64/kvm/handle_exit.c
> > +++ b/arch/arm64/kvm/handle_exit.c
> > @@ -566,9 +566,7 @@ void __noreturn __cold nvhe_hyp_panic_handler(u64 esr, u64 spsr,
> >   	kvm_nvhe_dump_backtrace(hyp_offset);
> >   	/* Dump the faulting instruction */
> > -	if (!is_protected_kvm_enabled() ||
> > -	    IS_ENABLED(CONFIG_NVHE_EL2_DEBUG))
> > -		dump_instr(panic_addr + kaslr_offset());
> > +	dump_instr(panic_addr + kaslr_offset());
> This makes the dumping in nvhe no longer conditional on
> CONFIG_NVHE_EL2_DEBUG. A change from what you introduced in the patch.
> Perhaps it makes sense to reorder the patches; do the preparatory work for
> instruction dumping before the enabling.>

Yes, I thought about squashing both patches, but I was worried this patch
might be more controversial, so I split the code into 2 patches, where the
first one can be merged separately if needed. But no strong opinion.

Thanks,
Mostafa

> >   	/*
> >   	 * Hyp has panicked and we're going to handle that by panicking the
> > diff --git a/arch/arm64/kvm/hyp/nvhe/setup.c b/arch/arm64/kvm/hyp/nvhe/setup.c
> > index a48d3f5a5afb..90bd014e952f 100644
> > --- a/arch/arm64/kvm/hyp/nvhe/setup.c
> > +++ b/arch/arm64/kvm/hyp/nvhe/setup.c
> > @@ -192,6 +192,7 @@ static int fix_host_ownership_walker(const struct kvm_pgtable_visit_ctx *ctx,
> >   	enum pkvm_page_state state;
> >   	struct hyp_page *page;
> >   	phys_addr_t phys;
> > +	enum kvm_pgtable_prot prot;
> >   	if (!kvm_pte_valid(ctx->old))
> >   		return 0;
> > @@ -210,11 +211,18 @@ static int fix_host_ownership_walker(const struct kvm_pgtable_visit_ctx *ctx,
> >   	 * configured in the hypervisor stage-1, and make sure to propagate them
> >   	 * to the hyp_vmemmap state.
> >   	 */
> > -	state = pkvm_getstate(kvm_pgtable_hyp_pte_prot(ctx->old));
> > +	prot = kvm_pgtable_hyp_pte_prot(ctx->old);
> > +	state = pkvm_getstate(prot);
> >   	switch (state) {
> >   	case PKVM_PAGE_OWNED:
> >   		set_hyp_state(page, PKVM_PAGE_OWNED);
> > -		return host_stage2_set_owner_locked(phys, PAGE_SIZE, PKVM_ID_HYP);
> > +		/* hyp text is RO in the host stage-2 to be inspected on panic. */
> > +		if (prot == PAGE_HYP_EXEC) {
> > +			set_host_state(page, PKVM_NOPAGE);
> > +			return host_stage2_idmap_locked(phys, PAGE_SIZE, KVM_PGTABLE_PROT_R);
> > +		} else {
> > +			return host_stage2_set_owner_locked(phys, PAGE_SIZE, PKVM_ID_HYP);
> > +		}
> >   	case PKVM_PAGE_SHARED_OWNED:
> >   		set_hyp_state(page, PKVM_PAGE_SHARED_OWNED);
> >   		set_host_state(page, PKVM_PAGE_SHARED_BORROWED);
> -- 
> Thanks,
> 
> Ben
>