Re: [PATCH 2/4] KVM: x86: separate pending and injected exception

[Sean Christopherson <seanjc@google.com>]

On Thu, Apr 01, 2021, Maxim Levitsky wrote:
> Use 'pending_exception' and 'injected_exception' fields
> to store the pending and the injected exceptions.
> 
> After this patch still only one is active, but
> in the next patch both could co-exist in some cases.

Please explain _why_.  

> Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
> ---
>  arch/x86/include/asm/kvm_host.h |  25 ++++--
>  arch/x86/kvm/svm/nested.c       |  26 +++---
>  arch/x86/kvm/svm/svm.c          |   6 +-
>  arch/x86/kvm/vmx/nested.c       |  36 ++++----
>  arch/x86/kvm/vmx/vmx.c          |  12 +--
>  arch/x86/kvm/x86.c              | 145 ++++++++++++++++++--------------
>  arch/x86/kvm/x86.h              |   6 +-
>  7 files changed, 143 insertions(+), 113 deletions(-)
> 
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index a52f973bdff6..3b2fd276e8d5 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -547,6 +547,14 @@ struct kvm_vcpu_xen {
>  	u64 runstate_times[4];
>  };
>  
> +struct kvm_queued_exception {
> +	bool valid;
> +	u8 nr;

If we're refactoring all this code anyways, maybe change "nr" to something a
bit more descriptive?  E.g. vector.

> +	bool has_error_code;
> +	u32 error_code;
> +};
> +
> +
>  struct kvm_vcpu_arch {
>  	/*
>  	 * rip and regs accesses must go through
> @@ -645,16 +653,15 @@ struct kvm_vcpu_arch {
>  
>  	u8 event_exit_inst_len;
>  
> -	struct kvm_queued_exception {
> -		bool pending;
> -		bool injected;
> -		bool has_error_code;
> -		u8 nr;
> -		u32 error_code;
> -		unsigned long payload;
> -		bool has_payload;
> +	struct kvm_queued_exception pending_exception;
> +
> +	struct kvm_exception_payload {
> +		bool valid;
> +		unsigned long value;
>  		u8 nested_apf;
> -	} exception;
> +	} exception_payload;

Hmm, even if it's dead code at this time, I think the exception payload should
be part of 'struct kvm_queued_exception'.  The payload is very much tied to a
single exception.

> +
> +	struct kvm_queued_exception injected_exception;

Any objection to keeping the current syntax, arch.exception.{pending,injected}?
Maybe it's fear of change, but I like the current style, I think because the
relevant info is condensed at the end, e.g. I can ignore "vcpu->arch.exception"
and look at "pending.vector" or whatever.  E.g.

	struct {
		struct kvm_queued_exception pending;
		struct kvm_queued_exception injected;
	} exception;
>  
>  	struct kvm_queued_interrupt {
>  		bool injected;