Re: [PATCHv5 05/19] mm/page_alloc: Handle allocation for encrypted memory

From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Michal Hocko <mhocko@suse.com>,
	"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	Ingo Molnar <mingo@redhat.com>,
	x86@kernel.org, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Tom Lendacky <thomas.lendacky@amd.com>,
	Kai Huang <kai.huang@linux.intel.com>,
	Jacob Pan <jacob.jun.pan@linux.intel.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCHv5 05/19] mm/page_alloc: Handle allocation for encrypted memory
Date: Fri, 20 Jul 2018 15:25:16 +0300	[thread overview]
Message-ID: <20180720122516.zm35yk4r4tcy752s@kshutemo-mobl1> (raw)
In-Reply-To: <b0a92a2f-cf14-c976-9fbd-fd9aa4ebcf96@intel.com>

On Thu, Jul 19, 2018 at 07:05:36AM -0700, Dave Hansen wrote:
> On 07/19/2018 01:27 AM, Kirill A. Shutemov wrote:
> >> What other code might need prep_encrypted_page()?
> > 
> > Custom pages allocators if these pages can end up in encrypted VMAs.
> > 
> > It this case compaction creates own pool of pages to be used for
> > allocation during page migration.
> 
> OK, that makes sense.  It also sounds like some great information to add
> near prep_encrypted_page().

Okay.

> Do we have any ability to catch cases like this if we get them wrong, or
> will we just silently corrupt data?

I cannot come up with any reasonable way to detect this immediately.
I'll think about this more.

-- 
 Kirill A. Shutemov