Re: [PATCH v3 2/2] vhost/vsock: keep IOTLB across feature updates
Stefano Garzarella <[email protected]>
| Newsgroups | org.kernel.vger.kvm,dev.linux.lists.virtualization,org.kernel.vger.linux-kernel,org.kernel.vger.netdev |
|---|---|
| Message-ID | <an2RzeZ2zI8mWFRM@sgarzare-redhat> |
On Mon, Aug 10, 2026 at 09:40:18PM +0800, Jia Jia wrote: >VHOST_SET_FEATURES is also used to update logging while a device is >running. When ACCESS_PLATFORM stays enabled, allocating a new empty >IOTLB on every call drops valid translations and forces avoidable >misses. > >Initialize the device IOTLB only when one does not already exist. > Do we need a Fixes tag here? >Signed-off-by: Jia Jia <[email protected]> >--- > drivers/vhost/vsock.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > >diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c >index 7372c22691de..1beae7f5b9a9 100644 >--- a/drivers/vhost/vsock.c >+++ b/drivers/vhost/vsock.c >@@ -889,7 +889,8 @@ static int vhost_vsock_set_features(struct vhost_vsock *vsock, u64 features) > goto err; > } > >- if ((features & (1ULL << VIRTIO_F_ACCESS_PLATFORM))) { >+ if ((features & (1ULL << VIRTIO_F_ACCESS_PLATFORM)) && >+ !vsock->dev.iotlb) { Do we need to do the same in vhost-net too? Thanks, Stefano > if (vhost_init_device_iotlb(&vsock->dev)) > goto err; > } >-- >2.34.1 >