Re: [PATCH v2 3/4] x86/sev: Fix broken SNP support with KVM module built-in

[Sean Christopherson <seanjc@google.com>]

On Thu, Jan 30, 2025, Ashish Kalra wrote:
> On 1/30/2025 7:41 PM, Sean Christopherson wrote:
> > On Fri, Jan 31, 2025, Ashish Kalra wrote:
> >> -nosnp:
> >> -	cc_platform_clear(CC_ATTR_HOST_SEV_SNP);
> >> -	return -ENOSYS;
> >>  }
> >>  
> >> -/*
> >> - * This must be called after the IOMMU has been initialized.
> >> - */
> >> -device_initcall(snp_rmptable_init);
> > 
> > There's the wee little problem that snp_rmptable_init() is never called as of
> > this patch.  Dropping the device_initcall() needs to happen in the same patch
> > that wires up the IOMMU code to invoke snp_rmptable_init().
> 
> The issue with that is the IOMMU and x86 maintainers are different, so i
> believe that we will need to split the dropping of device_initcall() in
> platform code and the code to wire up the IOMMU driver to invoke
> snp_rmptable_init(), to get the patch merged in different trees ?

No, that's pure insanity and not how patches that touch multiple subsystems are
handled.  This all goes through one tree, with Acks from the necessary parties.