Re: [PATCH v1 06/11] mm: support GUP-triggered unsharing via FAULT_FLAG_UNSHARE (!hugetlb)

From: Jason Gunthorpe
Date: Tue Dec 21 2021 - 18:59:24 EST

Next message: kernel test robot: "[axboe-block:nvme-passthru-wip.2 15/23] exit.c:(.text+0xac0): multiple definition of `io_uring_cmd_import_fixed'; kernel/fork.o:fork.c:(.text+0x600): first defined here"
Previous message: David E. Box: "[PATCH 3/4] net/mlx5e: Use auxiliary_device driver data helpers"
In reply to: Jason Gunthorpe: "Re: [PATCH v1 06/11] mm: support GUP-triggered unsharing via FAULT_FLAG_UNSHARE (!hugetlb)"
Next in thread: David Hildenbrand: "Re: [PATCH v1 06/11] mm: support GUP-triggered unsharing via FAULT_FLAG_UNSHARE (!hugetlb)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Tue, Dec 21, 2021 at 06:40:30PM +0100, David Hildenbrand wrote:

> What adds complexity to correctly maintain the "exclusive" state are at
> least:
> * KSM (might be harder, have to think about it)

I know little about it, but isn't KSM like fork where it is trying to
WP pages with the intention of copying them? Shouldn't KSM completely
reject WP'ing a page that is under any kind of writable GUP?

Jason

Next message: kernel test robot: "[axboe-block:nvme-passthru-wip.2 15/23] exit.c:(.text+0xac0): multiple definition of `io_uring_cmd_import_fixed'; kernel/fork.o:fork.c:(.text+0x600): first defined here"
Previous message: David E. Box: "[PATCH 3/4] net/mlx5e: Use auxiliary_device driver data helpers"
In reply to: Jason Gunthorpe: "Re: [PATCH v1 06/11] mm: support GUP-triggered unsharing via FAULT_FLAG_UNSHARE (!hugetlb)"
Next in thread: David Hildenbrand: "Re: [PATCH v1 06/11] mm: support GUP-triggered unsharing via FAULT_FLAG_UNSHARE (!hugetlb)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]