[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230920160424.GD13733@nvidia.com>
Date: Wed, 20 Sep 2023 13:04:24 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: ankita@...dia.com, alex.williamson@...hat.com, tony.luck@...el.com,
bp@...en8.de, naoya.horiguchi@....com, linmiaohe@...wei.com,
aniketa@...dia.com, cjia@...dia.com, kwankhede@...dia.com,
targupta@...dia.com, vsethi@...dia.com, acurrid@...dia.com,
anuaggarwal@...dia.com, linux-kernel@...r.kernel.org,
linux-mm@...ck.org, linux-edac@...r.kernel.org, kvm@...r.kernel.org
Subject: Re: [PATCH v1 0/4] mm: Implement ECC handling for pfn with no struct
page
On Wed, Sep 20, 2023 at 09:02:22AM -0700, Andrew Morton wrote:
> On Wed, 20 Sep 2023 19:32:06 +0530 <ankita@...dia.com> wrote:
>
> > The kernel MM currently handles ECC errors / poison only on memory page
> > backed by struct page. As part of [1], the nvgrace-gpu-vfio-pci module
> > maps the device memory to user VA (Qemu) using remap_pfn_range without
> > being added to the kernel. These pages are not backed by struct page.
>
> Are you able to identify any other drivers which can (or will) use
> this? Or is it likely that this feature will only ever be for
> nvgrace-gpu-vfio-pci?
I think a future vfio-cxl will have a similar desire at least.
Jason
Powered by blists - more mailing lists