Re: [PATCH v11 14/18] KVM: arm64: Handle guest_memfd-backed guest page faults

[David Hildenbrand <david@redhat.com>]

>>> +       write_fault = kvm_is_write_fault(vcpu);
>>> +       exec_fault = kvm_vcpu_trap_is_exec_fault(vcpu);
>>> +
>>> +       if (write_fault && exec_fault) {
>>> +               kvm_err("Simultaneous write and execution fault\n");
>>> +               return -EFAULT;
>>> +       }
>>> +
>>> +       if (is_perm && !write_fault && !exec_fault) {
>>> +               kvm_err("Unexpected L2 read permission error\n");
>>> +               return -EFAULT;
>>> +       }
>>
>> I think, ideally, these above checks should be put into a separate
>> function and shared with user_mem_abort(). (The VM_BUG_ON(write_fault
>> && exec_fault) that user_mem_abort() does seems fine to me, I don't see a
>> real need to change it to -EFAULT.)
> 
> I would like to do that, however, I didn't want to change
> user_mem_abort(), and regarding the VM_BUG_ON, see David's feedback to
> V10:
> 
> https://lore.kernel.org/all/ed1928ce-fc6f-4aaa-9f54-126a8af12240@redhat.com/

Worth reading Linus' reply in [1], that contains a bit more history on 
BUG_ON() and how it should not be used. (VM_BUG_ON we'll now likely get 
rid of completely)

[1] https://lkml.kernel.org/r/20250604140544.688711-1-david@redhat.com

-- 
Cheers,

David / dhildenb