Received-SPF: pass (google.com: domain of linux-kernel+bounces-110185-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Thu, 21 Mar 2024 16:58:52 +0200
From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To: Ashish Kalra <Ashish.Kalra@amd.com>
Cc: tglx@linutronix.de, mingo@redhat.com, dave.hansen@linux.intel.com, 
	rafael@kernel.org, peterz@infradead.org, adrian.hunter@intel.com, 
	sathyanarayanan.kuppuswamy@linux.intel.com, elena.reshetova@intel.com, jun.nakajima@intel.com, 
	rick.p.edgecombe@intel.com, thomas.lendacky@amd.com, seanjc@google.com, michael.roth@amd.com, 
	kai.huang@intel.com, bhe@redhat.com, kexec@lists.infradead.org, 
	linux-coco@lists.linux.dev, linux-kernel@vger.kernel.org, bdas@redhat.com, 
	vkuznets@redhat.com, dionnaglaze@google.com, anisinha@redhat.com, jroedel@suse.de
Subject: Re: [PATCH v2 2/3] x86/mm: Do not zap page table entries mapping
 unaccepted memory table during kdump.
Message-ID: <opa56dnwlyllrrnlfrhtbavnwmhxxatiuzifvx3bquv5qksidn@v2qxrvvw53jl>
References: <20240227212452.3228893-1-kirill.shutemov@linux.intel.com>
 <cover.1710744412.git.ashish.kalra@amd.com>
 <8e11825bb3e4a7472bf0cb30c72ae013c25ac8c9.1710744412.git.ashish.kalra@amd.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <8e11825bb3e4a7472bf0cb30c72ae013c25ac8c9.1710744412.git.ashish.kalra@amd.com>

On Mon, Mar 18, 2024 at 07:02:45AM +0000, Ashish Kalra wrote:
> From: Ashish Kalra <ashish.kalra@amd.com>
> 
> During crashkernel boot only pre-allocated crash memory is presented as
> E820_TYPE_RAM. This can cause page table entries mapping unaccepted memory
> table to be zapped during phys_pte_init(), phys_pmd_init(), phys_pud_init()
> and phys_p4d_init() as SNP/TDX guest use E820_TYPE_ACPI to store the
> unaccepted memory table and pass it between the kernels on
> kexec/kdump.
> 
> E820_TYPE_ACPI covers not only ACPI data, but also EFI tables and might
> be required by kernel to function properly.
> 
> The problem was discovered during debugging kdump for SNP guest. The
> unaccepted memory table stored with E820_TYPE_ACPI and passed between
> the kernels on kdump was getting zapped as the PMD entry mapping this
> is above the E820_TYPE_RAM range for the reserved crashkernel memory.
> 
> Signed-off-by: Ashish Kalra <ashish.kalra@amd.com>

Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

I guess it would be better if I take this patch into my kexec patchset. I
guess I just got lucky not to step onto the issue.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov