Re: [PATCH v2 30/45] KVM: Move vcpu alloc and init invocation to common code

[Sean Christopherson]

On Fri, Dec 20, 2019 at 10:33:25AM +0100, Cornelia Huck wrote:
> On Wed, 18 Dec 2019 13:55:15 -0800
> Sean Christopherson <sean.j.christopherson@xxxxxxxxx> wrote:
> > +int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
> >  {
> > -	struct kvm_vcpu *vcpu;
> >  	struct sie_page *sie_page;
> >  	int rc;
> >  
> > -	rc = -ENOMEM;
> > -
> > -	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
> > -	if (!vcpu)
> > -		goto out;
> > -
> > -	rc = kvm_vcpu_init(vcpu, kvm, id);
> > -	if (rc)
> > -		goto out_free_cpu;
> > -
> > -	rc = -ENOMEM;
> > -
> >  	BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
> >  	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
> >  	if (!sie_page)
> > -		goto out_uninit_vcpu;
> > +		return -ENOMEM;
> >  
> >  	vcpu->arch.sie_block = &sie_page->sie_block;
> >  	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;
> > @@ -3087,15 +3070,11 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
> >  		 vcpu->arch.sie_block);
> >  	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
> >  
> > -	return vcpu;
> > +	return 0;
> > +
> >  out_free_sie_block:
> >  	free_page((unsigned long)(vcpu->arch.sie_block));
> > -out_uninit_vcpu:
> > -	kvm_vcpu_uninit(vcpu);
> > -out_free_cpu:
> > -	kmem_cache_free(kvm_vcpu_cache, vcpu);
> > -out:
> > -	return ERR_PTR(rc);
> > +	return rc;
> 
> This is getting a bit hard to follow across the patches, but I think rc
> is now only set for ucontrol guests. So this looks correct right now,
> but feels a bit brittle... should we maybe init rc to 0 and always
> return rc instead?

Yes, but only for a few patches until kvm_s390_vcpu_setup() is introduced,
at which point @rc is unconditionally set at the end.

        rc = kvm_s390_vcpu_setup(vcpu);
        if (rc)
                goto out_free_ucontrol_gmap;
        return 0;

My personal preference is to use "return 0;" when the return is known to
be zero as it makes the success path obvious at a glance.  I also didn't
want to intialize @rc at he beginning because then the sie_page allocation
would look a bit funky:

	int rc = 0;

        BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
        sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
        if (!sie_page)
                return -ENOMEM;


An alternative would be to init @rc to -ENOMEM:

        int rc = -ENOMEM;

        BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
        sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
        if (!sie_page)
                return rc;

This would be my preference if you'd prefer to init @rc right away,
especially if  __kvm_ucontrol_vcpu_init() is open coded here (discussion
in patch 35, "KVM: s390: Manually invoke vcpu setup during 
kvm_arch_vcpu_create()", e.g.:

        int rc = -ENOMEM;

        BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
        sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
        if (!sie_page)
                return rc;

	...

        if (kvm_is_ucontrol(vcpu->kvm)) {
                vcpu->arch.gmap = gmap_create(current->mm, -1UL);
                if (!vcpu->arch.gmap)
                        goto out_free_sie_block;
                vcpu->arch.gmap->private = vcpu->kvm;
        }

        VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
                 vcpu->arch.sie_block);
        trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);

        rc = kvm_s390_vcpu_setup(vcpu);
        if (rc)
                goto out_free_ucontrol_gmap;
        return 0;

out_free_ucontrol_gmap:
        if (kvm_is_ucontrol(vcpu->kvm))
                gmap_remove(vcpu->arch.gmap);
out_free_sie_block:
        free_page((unsigned long)(vcpu->arch.sie_block));
        return rc;