Re: [RFC v4 PATCH 00/49] Initial support of multi-process qemu

[Stefan Hajnoczi <stefanha@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 2321 bytes --]

On Thu, Oct 24, 2019 at 05:08:41AM -0400, Jagannathan Raman wrote:
> Started with the presentation in October 2017 made by Marc-Andre (Red Hat)
> and Konrad Wilk (Oracle) [1], and continued by Jag's BoF at KVM Forum 2018,
> the multi-process project is now a prototype and presented in this patchset.
> John & Elena will present the status of this project in KVM Forum 2019.
> 
> This first series enables the emulation of lsi53c895a in a separate process.
> 
> We posted the Proof Of Concept patches [2] before the BoF session in 2018.
> Subsequently, we posted RFC v1 [3], RFC v2 [4] and RFC v3 [5] of this series. 
> 
> We want to present version 4 of this series, which incorporates the feedback
> we received for v3 & adds support for live migrating the remote process.
> 
> Following people contributed to this patchset:
> 
> John G Johnson <john.g.johnson@oracle.com>
> Jagannathan Raman <jag.raman@oracle.com>
> Elena Ufimtseva <elena.ufimtseva@oracle.com>
> Kanth Ghatraju <kanth.ghatraju@oracle.com>
> 
> For full concept writeup about QEMU disaggregation refer to
> docs/devel/qemu-multiprocess.rst. Please refer to 
> docs/qemu-multiprocess.txt for usage information.
> 
> We are planning on making the following improvements in the future:
>  - Performance improvements
>  - Libvirt support
>  - Enforcement of security policies
>  - blockdev support
> 
> We welcome all your ideas, concerns, and questions for this patchset.
> 
> Thank you!

I've wrapped up for v4.  There is more to review in detail but I've
posted enough comments so that I'd like to see the next revision before
investing more time.

The main topics:

1. It's possible to have just one proxy device per bus type (PCI, USB,
   etc).  The proxy device instance can be initialized by communicating
   with the remote process to inquire about its memory regions,
   interrupts, etc.  This removes the need to hardcode that information
   into per-device proxy objects, which is tedious and can get
   out-of-sync with the real device emulation code.

   This is becoming similar to doing VFIO or muser over a socket...

2. Security and code quality.  Missing input validation and resource
   leaks don't inspire confidence :(.

   Please run scripts/checkpatch.pl on the code.

Stefan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]