Michael S. Tsirkin
2017-Feb-03 03:49 UTC
[PATCH] Revert "vring: Force use of DMA API for ARM-based systems with legacy devices"
This reverts commit c7070619f3408d9a0dffbed9149e6f00479cf43b. This has been shown to regress on some ARM systems: by forcing on DMA API usage for ARM systems, we have inadvertently kicked open a hornets' nest in terms of cache-coherency. Namely that unless the virtio device is explicitly described as capable of coherent DMA by firmware, the DMA APIs on ARM and other DT-based platforms will assume it is non-coherent. This turns out to cause a big problem for the likes of QEMU and kvmtool, which generate virtio-mmio devices in their guest DTs but neglect to add the often-overlooked "dma-coherent" property; as a result, we end up with the guest making non-cacheable accesses to the vring, the host doing so cacheably, both talking past each other and things going horribly wrong. We are working on a safer work-around. Fixes: c7070619f340 ("vring: Force use of DMA API for ARM-based systems with legacy devices") Reported-by: Robin Murphy <robin.murphy at arm.com> Cc: <stable at vger.kernel.org> Signed-off-by: Will Deacon <will.deacon at arm.com> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> Acked-by: Marc Zyngier <marc.zyngier at arm.com> --- I'll merge this for 4.10. Let's fix it properly for 4.11. drivers/virtio/virtio_ring.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 7e38ed7..409aeaa 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -159,13 +159,6 @@ static bool vring_use_dma_api(struct virtio_device *vdev) if (xen_domain()) return true; - /* - * On ARM-based machines, the DMA ops will do the right thing, - * so always use them with legacy devices. - */ - if (IS_ENABLED(CONFIG_ARM) || IS_ENABLED(CONFIG_ARM64)) - return !virtio_has_feature(vdev, VIRTIO_F_VERSION_1); - return false; } -- MST
Will Deacon
2017-Feb-03 09:57 UTC
[PATCH] Revert "vring: Force use of DMA API for ARM-based systems with legacy devices"
On Fri, Feb 03, 2017 at 05:49:11AM +0200, Michael S. Tsirkin wrote:> This reverts commit c7070619f3408d9a0dffbed9149e6f00479cf43b. > > This has been shown to regress on some ARM systems: > > by forcing on DMA API usage for ARM systems, we have inadvertently > kicked open a hornets' nest in terms of cache-coherency. Namely that > unless the virtio device is explicitly described as capable of coherent > DMA by firmware, the DMA APIs on ARM and other DT-based platforms will > assume it is non-coherent. This turns out to cause a big problem for the > likes of QEMU and kvmtool, which generate virtio-mmio devices in their > guest DTs but neglect to add the often-overlooked "dma-coherent" > property; as a result, we end up with the guest making non-cacheable > accesses to the vring, the host doing so cacheably, both talking past > each other and things going horribly wrong. > > We are working on a safer work-around. > > Fixes: c7070619f340 ("vring: Force use of DMA API for ARM-based systems with legacy devices") > Reported-by: Robin Murphy <robin.murphy at arm.com> > Cc: <stable at vger.kernel.org> > Signed-off-by: Will Deacon <will.deacon at arm.com> > Signed-off-by: Michael S. Tsirkin <mst at redhat.com> > Acked-by: Marc Zyngier <marc.zyngier at arm.com> > --- > > I'll merge this for 4.10. Let's fix it properly for 4.11.Acked-by: Will Deacon <will.deacon at arm.com> Will
Possibly Parallel Threads
- [PATCH] Revert "vring: Force use of DMA API for ARM-based systems with legacy devices"
- [PATCH] virtio: Try to untangle DMA coherency
- [PATCH] virtio: Try to untangle DMA coherency
- [PATCH v2 2/2] vring: Force use of DMA API for ARM-based systems
- [PATCH v2 2/2] vring: Force use of DMA API for ARM-based systems