lkml.org 
[lkml]   [2013]   [Nov]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v2 0/2] vmpslice support for zero-copy gifting of pages
On 10/25/2013 05:46 PM, Robert Jennings wrote:
> From: Robert C Jennings <rcj@linux.vnet.ibm.com>
>
> This patch set would add the ability to move anonymous user pages from one
> process to another through vmsplice without copying data. Moving pages
> rather than copying is implemented for a narrow case in this RFC to meet
> the needs of QEMU's usage (below).
>
> Among the restrictions the source address and destination addresses must
> be page aligned, the size argument must be a multiple of page size,
> and by the time the reader calls vmsplice, the page must no longer be
> mapped in the source. If a move is not possible the code transparently
> falls back to copying data.
>
> This comes from work in QEMU[1] to migrate a VM from one QEMU instance
> to another with minimal down-time for the VM. This would allow for an
> update of the QEMU executable under the VM.

Hello,

since this seems somewhat narrow use case for a syscall change, it would
be helpful if you included a larger discussion of considered existing
alternatives, with benchmark results justifying the changed syscall. E.g.:
- Cross Memory Attach comes to mind as one alternative to vmsplice.
Although it does perform a single copy, there are results suggesting
zero-copy doesn't necessarily add that much gain:
http://marc.info/?l=linux-mm&m=130105930902915&w=2
- Would it be possible for QEMU to use shared memory to begin with?
Since you are already restricting this to page-aligned regions.

Ideally the benchmark results would also include the THP support when
complete.

Thanks,
Vlastimil

> New flag usage
> This introduces use of the SPLICE_F_MOVE flag for vmsplice, previously
> unused. Proposed usage is as follows:
>
> Writer gifts pages to pipe, can not access original contents after gift:
> vmsplice(fd, iov, nr_segs, (SPLICE_F_GIFT | SPLICE_F_MOVE);
> Reader asks kernel to move pages from pipe to memory described by iovec:
> vmsplice(fd, iov, nr_segs, SPLICE_F_MOVE);
>
> Moving pages rather than copying is implemented for a narrow case in
> this RFC to meet the needs of QEMU's usage. If a move is not possible
> the code transparently falls back to copying data.
>
> For older kernels the SPLICE_F_MOVE would be ignored and a copy would occur.
>
> [1] QEMU localhost live migration:
> http://lists.gnu.org/archive/html/qemu-devel/2013-10/msg02787.html
>
> Changes from V1:
> - Cleanup zap coalescing in splice_to_pipe for readability
> - Field added to struct partial_page in v1 was unnecessary, using
> private field instead.
> - Read-side code in pipe_to_user pulled out into a new function
> - Improved documentation of read-side flipping code
> - Fixed locking issue in read-size flipping code found by sparse
> - Updated vmsplice comments for vmsplice_to_user(),
> vmsplice_to_pipe, and vmsplice syscall
> _______________________________________________________
>
> vmsplice: unmap gifted pages for recipient
> vmsplice: Add limited zero copy to vmsplice
>
> fs/splice.c | 159 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++----
> 1 file changed, 150 insertions(+), 9 deletions(-)
>



\
 
 \ /
  Last update: 2013-11-04 17:21    [W:0.057 / U:3.088 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site