Re: [PATCH v2 1/6] KVM: x86: Add dedicated storage for guest RIP

[Chang S. Bae]

On 4/9/2026 3:42 PM, Sean Christopherson wrote:
> Add kvm_vcpu_arch.rip to track guest RIP instead of including it in the
> generic regs[] array.  Decoupling RIP from regs[] will allow using a
> *completely* arbitrary index for RIP, as opposed to the mostly-arbitrary
> index that is currently used.  That in turn will allow using indices
> 16-31 to track R16-R31 that are coming with APX....
> -	unsigned long regs[NR_VCPU_REGS];
> +	unsigned long regs[NR_VCPU_GENERAL_PURPOSE_REGS];
> +	unsigned long rip;

Digging the history, this effectively reverts part of changes in

  commit 5fdbf9765b7b ("KVM: x86: accessors for guest registers")

which had

-       unsigned long regs[NR_VCPU_REGS];
-       unsigned long rip;      /* needs vcpu_load_rsp_rip() */
+       /*
+        * rip and regs accesses must go through
+        * kvm_{register,rip}_{read,write} functions.
+        */
+       unsigned long regs[NR_VCPU_REGS];

But its changelog didn't go into much detail about this change. I could 
only relate to vcpu_load_rsp_rip() which might establish perception 
coupling RSP with RIP back then.

In any case, it doesn't matter. I think this patch makes a clear 
improvement - for example, now aligns with _regs[NR_EMULATOR_GPRS] in 
struct x86_emulate_ctxt for general consistency.

Indeed, this and the whole series paves the way for APX. Appreciate for 
the time and effort!

Reviewed-by: Chang S. Bae <chang.seok.bae@xxxxxxxxx>