Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Marek Marczykowski-Górecki
On Fri, Sep 23, 2016 at 11:00:42PM +0200, Samuel Thibault wrote: > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 20:56:43 +0200, wrote: > > 1. How to do this? ;) I.e. what synchronization primitives are available > > in mini-os? Just pthread_mutex_lock/unlock? > > pthread_mutex_lock are nops :o)

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Samuel Thibault
Marek Marczykowski-Górecki, on Fri 23 Sep 2016 20:56:43 +0200, wrote: > 1. How to do this? ;) I.e. what synchronization primitives are available > in mini-os? Just pthread_mutex_lock/unlock? pthread_mutex_lock are nops :o) because we don't have pthread_create. But for mini-os itself there are sync

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Marek Marczykowski-Górecki
On Fri, Sep 23, 2016 at 04:51:33PM +0200, Samuel Thibault wrote: > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 16:25:41 +0200, wrote: > > Toolstack during (stub)domain startup setup two things, mostly > > asynchronously: > > 1. pciback/pcifront (through standard xenstore entries) > > 2. instruct

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Marek Marczykowski-Górecki
On Fri, Sep 23, 2016 at 11:35:27AM -0400, Konrad Rzeszutek Wilk wrote: > On Fri, Sep 23, 2016 at 04:25:41PM +0200, Marek Marczykowski-Górecki wrote: > > On Fri, Sep 23, 2016 at 03:27:07PM +0200, Samuel Thibault wrote: > > > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 10:48:14 +0200, wrote: > > >

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Konrad Rzeszutek Wilk
On Fri, Sep 23, 2016 at 04:25:41PM +0200, Marek Marczykowski-Górecki wrote: > On Fri, Sep 23, 2016 at 03:27:07PM +0200, Samuel Thibault wrote: > > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 10:48:14 +0200, wrote: > > > I'm still trying to get PCI passthrough working with stubdomain and > > > wi

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Samuel Thibault
Marek Marczykowski-Górecki, on Fri 23 Sep 2016 16:25:41 +0200, wrote: > On Fri, Sep 23, 2016 at 03:27:07PM +0200, Samuel Thibault wrote: > > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 10:48:14 +0200, wrote: > > > I'm still trying to get PCI passthrough working with stubdomain and > > > without

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Marek Marczykowski-Górecki
On Fri, Sep 23, 2016 at 03:27:07PM +0200, Samuel Thibault wrote: > Marek Marczykowski-Górecki, on Fri 23 Sep 2016 10:48:14 +0200, wrote: > > I'm still trying to get PCI passthrough working with stubdomain and > > without qemu in dom0 (even for just vfb/vkbd backends). How is this > > supposed to wo

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Samuel Thibault
Marek Marczykowski-Górecki, on Fri 23 Sep 2016 10:48:14 +0200, wrote: > I'm still trying to get PCI passthrough working with stubdomain and > without qemu in dom0 (even for just vfb/vkbd backends). How is this > supposed to work? Just as I recall from my memory: > 1. Should xen-pcifront in the ta

Re: [Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Jan Beulich
>>> On 23.09.16 at 10:48, wrote: > 2. What operations are done through stubdomain and what are handled > directly by Xen (itself, or with help of IOMMU)? I'd guess config space > accesses are done through device model. Anything else? MSI-X table handling I think (guest accesses really mainly get

[Xen-devel] PCI passthrough with stubdomain

2016-09-23 Thread Marek Marczykowski-Górecki
Hi, I'm still trying to get PCI passthrough working with stubdomain and without qemu in dom0 (even for just vfb/vkbd backends). How is this supposed to work? 1. Should xen-pcifront in the target domain be used (and consequently, should xen-pciback be set for it)? Currently xen-pciback is set for