Re: [PATCH RFC 0/5] mm: process_vm_mmap() -- syscall for duplication a process mapping

From: Jann Horn <jannh@google.com>
To: Kirill Tkhai <ktkhai@virtuozzo.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Dan Williams <dan.j.williams@intel.com>,
	 Michal Hocko <mhocko@suse.com>,
	keith.busch@intel.com,
	 "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
	pasha.tatashin@oracle.com,
	 Alexander Duyck <alexander.h.duyck@linux.intel.com>,
	ira.weiny@intel.com,  Andrey Konovalov <andreyknvl@google.com>,
	arunks@codeaurora.org,  Vlastimil Babka <vbabka@suse.cz>,
	Christoph Lameter <cl@linux.com>, Rik van Riel <riel@surriel.com>,
	 Kees Cook <keescook@chromium.org>,
	hannes@cmpxchg.org, npiggin@gmail.com,
	 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Shakeel Butt <shakeelb@google.com>,  Roman Gushchin <guro@fb.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Hugh Dickins <hughd@google.com>,
	 Jerome Glisse <jglisse@redhat.com>,
	Mel Gorman <mgorman@techsingularity.net>,
	 daniel.m.jordan@oracle.com,
	kernel list <linux-kernel@vger.kernel.org>,
	 Linux-MM <linux-mm@kvack.org>
Subject: Re: [PATCH RFC 0/5] mm: process_vm_mmap() -- syscall for duplication a process mapping
Date: Thu, 16 May 2019 15:14:08 +0200	[thread overview]
Message-ID: <CAG48ez0itiEE1x=SXeMbjKvMGkrj7wxjM6c+ZB00LpXAAhqmiw@mail.gmail.com> (raw)
In-Reply-To: <456c7367-0656-933b-986d-febdcc5ab98e@virtuozzo.com>

On Thu, May 16, 2019 at 3:03 PM Kirill Tkhai <ktkhai@virtuozzo.com> wrote:
> On 15.05.2019 21:46, Jann Horn wrote:
> > On Wed, May 15, 2019 at 5:11 PM Kirill Tkhai <ktkhai@virtuozzo.com> wrote:
> >> This patchset adds a new syscall, which makes possible
> >> to clone a mapping from a process to another process.
> >> The syscall supplements the functionality provided
> >> by process_vm_writev() and process_vm_readv() syscalls,
> >> and it may be useful in many situation.
> >>
> >> For example, it allows to make a zero copy of data,
> >> when process_vm_writev() was previously used:
> > [...]
> >> This syscall may be used for page servers like in example
> >> above, for migration (I assume, even virtual machines may
> >> want something like this), for zero-copy desiring users
> >> of process_vm_writev() and process_vm_readv(), for debug
> >> purposes, etc. It requires the same permittions like
> >> existing proc_vm_xxx() syscalls have.
> >
> > Have you considered using userfaultfd instead? userfaultfd has
> > interfaces (UFFDIO_COPY and UFFDIO_ZERO) for directly shoving pages
> > into the VMAs of other processes. This works without the churn of
> > creating and merging VMAs all the time. userfaultfd is the interface
> > that was written to support virtual machine migration (and it supports
> > live migration, too).
>
> I know about userfaultfd, but it does solve the discussed problem.
> It allocates new pages to make UFFDIO_COPY (see mcopy_atomic_pte()),
> and it accumulates all the disadvantages, the example from [0/5]
> message has.

Sorry, right, I misremembered that.