[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1481121643.11413.14.camel@redhat.com>
Date: Wed, 07 Dec 2016 15:40:43 +0100
From: Gerd Hoffmann <kraxel@...hat.com>
To: Jike Song <jike.song@...el.com>
Cc: Kirti Wankhede <kwankhede@...dia.com>, alex.williamson@...hat.com,
cjia@...dia.com, pbonzini@...hat.com, qemu-devel@...gnu.org,
kvm@...r.kernel.org, kevin.tian@...el.com,
bjsdjshi@...ux.vnet.ibm.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v9 00/12] Add Mediated device support
Hi,
> Will update a topic/gvt-next-kvmgt-2016-12-06 today, sorry for the
> inconvenience :)
Thanks, that brings us one step forward.
Linux guest can see the device (in lspci). Trying to load i915.ko leads
to kernel oopses on both guest and host though. So I guess the guest
driver can't handle the device.
I've tried RHEL-7.3, kernel 3.10.0-514.el7.x86_64. The drivers/gpu/drm/
subsystem is roughly at upstream kernel v4.6 level. Is that too old?
IIRC some vgpu guest patches have been merged in v4.2 timeframe, so I
kind-of expected this to work ...
I'll go try the kvmgt branch in the guest next.
What is the upstream merge status btw? As far I know the bulk of the
gvt code is scheduled for 4.10. What about the gvt/mdev plumbing? Will
that land in 4.10 too or is that planned for 4.11?
thanks,
Gerd
Powered by blists - more mailing lists