Re: [Regression v5.19-rc1] crash kexec fails to boot the 2nd kernel (Re: [PATCH v12 38/46] x86/sev: Add SEV-SNP feature detection/setup)

From: Michael Roth <michael.roth@amd.com>
To: "NOMURA JUNICHI(野村　淳一)" <junichi.nomura@nec.com>
Cc: Borislav Petkov <bp@alien8.de>,
	LKML <linux-kernel@vger.kernel.org>,
	"x86@kernel.org" <x86@kernel.org>
Subject: Re: [Regression v5.19-rc1] crash kexec fails to boot the 2nd kernel (Re: [PATCH v12 38/46] x86/sev: Add SEV-SNP feature detection/setup)
Date: Wed, 29 Jun 2022 10:35:35 -0500	[thread overview]
Message-ID: <20220629153535.ivrp2x6j74cnpsv7@amd.com> (raw)
In-Reply-To: <20220629135254.cicvyp5clm4s4n35@amd.com>

On Wed, Jun 29, 2022 at 08:52:54AM -0500, Michael Roth wrote:
> 
> So the runtime kernel relies on boot/compressed for handling EFI, and
> simply uses boot_params to pass that on to runtime kernel.
> 
> But it has its own handling for cc_blob passed via setup_data entry,
> because that entry point is meant for boot loaders / VMMs that use stuff
> like CONFIG_PVH to jump directly into decompressed kernel: those are the
> cases where we'd expect cc blob via setup_data entry. But this is a
> critical thing to support right now, and if we can't find a proper fix
> we can simply disable this patch and not check for setup_data entry in
> runtime kernel.

Sorry for the confusing typos:

I meant to say that this (setup_data-based cc blob for decompressed kernel)
*isn't* a critical thing to support right now, and disabling that boot
path is certainly an option if that ends up being necessary.

But I do think the boot/compressed issue with how it doesn't map setup_data
entries into identity map should be fixed either way, as it seems like a
latent issue that snp_init() just happens to be the first to trigger.