Re: Coverity: handle_vmclear(): Error handling issues

[Sean Christopherson]

On Thu, Dec 01, 2022, coverity-bot wrote:
> Hello!
> 
> This is an experimental semi-automated report about issues detected by
> Coverity from a scan of next-20221201 as part of the linux-next scan project:
> https://scan.coverity.com/projects/linux-next-weekly-scan
> 
> You're getting this email because you were associated with the identified
> lines of code (noted below) that were touched by commits:
> 
>   Fri Dec 14 17:59:46 2018 +0100
>     55d2375e58a6 ("KVM: nVMX: Move nested code to dedicated files")
> 
> Coverity reported the following:
> 
> *** CID 1527765:  Error handling issues  (CHECKED_RETURN)
> arch/x86/kvm/vmx/nested.c:5269 in handle_vmclear()
> 5263     	 */
> 5264     	if (likely(!guest_cpuid_has_evmcs(vcpu) ||
> 5265     		   !evmptr_is_valid(nested_get_evmptr(vcpu)))) {
> 5266     		if (vmptr == vmx->nested.current_vmptr)
> 5267     			nested_release_vmcs12(vcpu);
> 5268
> vvv     CID 1527765:  Error handling issues  (CHECKED_RETURN)
> vvv     Calling "kvm_vcpu_write_guest" without checking return value (as is done elsewhere 7 out of 8 times).
> 5269     		kvm_vcpu_write_guest(vcpu,
> 5270     				     vmptr + offsetof(struct vmcs12,
> 5271     						      launch_state),
> 5272     				     &zero, sizeof(zero));
> 5273     	} else if (vmx->nested.hv_evmcs && vmptr == vmx->nested.hv_evmcs_vmptr) {
> 5274     		nested_release_evmcs(vcpu);
> 
> If this is a false positive, please let us know so we can mark it as
> such, or teach the Coverity rules to be smarter. If not, please make
> sure fixes get into linux-next. :) For patches fixing this, please
> include these lines (but double-check the "Fixes" first):
> 
> Reported-by: coverity-bot <keescook+coverity-bot@xxxxxxxxxxxx>
> Addresses-Coverity-ID: 1527765 ("Error handling issues")
> Fixes: 55d2375e58a6 ("KVM: nVMX: Move nested code to dedicated files")

Nit on the bot, if it's going to provide a Fixes without human verification, then
it should probably try to bisect (which I realize could get quite nasty).  Both
this VMCLEAR issue and the SMM issue report bogus Fixes due to code movement.  If
the blamed commit on this won't hadn't been so obviously wrong I likely would have
copy+pasted without ever verifying.

Maybe just omit the Fixes entirely and rely on the above "touched by commits" to
provide the developer with the hint?