Benjamin Herrenschmidt
2015-Nov-10 22:27 UTC
[PATCH v4 0/6] virtio core DMA API conversion
On Tue, 2015-11-10 at 10:54 -0800, Andy Lutomirski wrote:>? > Does that work on powerpc on existing kernels? > > Anyway, here's another crazy idea: make the quirk assume that the > IOMMU is bypasses if and only if the weak barriers bit is set on > systems that are missing the new DT binding."New DT bindings" doesn't mean much ... how do we change DT bindings on existing machines with a FW in flash ? What about partition <-> partition virtio such as what we could do on PAPR systems. That would have the weak barrier bit. Ben.
On Tue, Nov 10, 2015 at 2:27 PM, Benjamin Herrenschmidt <benh at kernel.crashing.org> wrote:> On Tue, 2015-11-10 at 10:54 -0800, Andy Lutomirski wrote: >> >> Does that work on powerpc on existing kernels? >> >> Anyway, here's another crazy idea: make the quirk assume that the >> IOMMU is bypasses if and only if the weak barriers bit is set on >> systems that are missing the new DT binding. > > "New DT bindings" doesn't mean much ... how do we change DT bindings on > existing machines with a FW in flash ? > > What about partition <-> partition virtio such as what we could do on > PAPR systems. That would have the weak barrier bit. >Is it partition <-> partition, bypassing IOMMU? I think I'd settle for just something that doesn't regress non-experimental setups that actually work today and that allow new setups (x86 with fixed QEMU and maybe something more complicated on powerpc and/or sparc) to work in all cases. We could certainly just make powerpc and sparc continue bypassing the IOMMU until someone comes up with a way to fix it. I'll send out some patches that do that, and maybe that'll help this make progress. --Andy
Benjamin Herrenschmidt
2015-Nov-11 00:44 UTC
[PATCH v4 0/6] virtio core DMA API conversion
On Tue, 2015-11-10 at 15:44 -0800, Andy Lutomirski wrote:> > > What about partition <-> partition virtio such as what we could do on > > PAPR systems. That would have the weak barrier bit. > > > > Is it partition <-> partition, bypassing IOMMU?No.> I think I'd settle for just something that doesn't regress > non-experimental setups that actually work today and that allow new > setups (x86 with fixed QEMU and maybe something more complicated on > powerpc and/or sparc) to work in all cases. > > We could certainly just make powerpc and sparc continue bypassing the > IOMMU until someone comes up with a way to fix it.? I'll send out some > patches that do that, and maybe that'll help this make progress.But we haven't found a solution that works. All we have come up with is a quirk that will force bypass on virtio always and will not allow us to operate non-bypassing devices on either of those architectures in the future. I'm not too happy about this. Ben.