[PATCH v3 0/3] vhost-net regression fixes

Michael S. Tsirkin posted 3 patches 2 weeks, 2 days ago
Only 1 patches received!
drivers/vhost/net.c | 44 ++++++++++++++++++++------------------------
1 file changed, 20 insertions(+), 24 deletions(-)
[PATCH v3 0/3] vhost-net regression fixes
Posted by Michael S. Tsirkin 2 weeks, 2 days ago
Two regressions were reported in vhost-net.
This is based on a patchset by Jason, but with
patch 2 from his series split up and made simpler.

Lightly tested.
Jason, Jon could you pls test this as well, and report?


Jason Wang (2):
  vhost-net: unbreak busy polling

Michael S. Tsirkin (1):
  Revert "vhost/net: Defer TX queue re-enable until after sendmsg"
  vhost-net: flush batched before enabling notifications

 drivers/vhost/net.c | 44 ++++++++++++++++++++------------------------
 1 file changed, 20 insertions(+), 24 deletions(-)

-- 
MST
Re: [PATCH v3 0/3] vhost-net regression fixes
Posted by Jon Kohler 2 weeks, 2 days ago

> On Sep 15, 2025, at 12:03 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> 
> !-------------------------------------------------------------------|
>  CAUTION: External Email
> 
> |-------------------------------------------------------------------!
> 
> 
> Two regressions were reported in vhost-net.
> This is based on a patchset by Jason, but with
> patch 2 from his series split up and made simpler.
> 
> Lightly tested.
> Jason, Jon could you pls test this as well, and report?

For the iotlb bit (patch 2/3):
Tried this out on a 6.16 host / guest that locked up with iotlb miss loop,
applied this patch and all was well.

Tested-by: Jon Kohler <jon@nutanix.com> 

> 
> 
> Jason Wang (2):
>  vhost-net: unbreak busy polling
> 
> Michael S. Tsirkin (1):
>  Revert "vhost/net: Defer TX queue re-enable until after sendmsg"
>  vhost-net: flush batched before enabling notifications
> 
> drivers/vhost/net.c | 44 ++++++++++++++++++++------------------------
> 1 file changed, 20 insertions(+), 24 deletions(-)
> 
> -- 
> MST
> 

Re: [PATCH v3 0/3] vhost-net regression fixes
Posted by Lei Yang 2 weeks, 2 days ago
Tested this series of patches with virtio-net regression tests,
everything works fine. Please feel free to let me know if these
patches need me to do more tests.

Tested-by: Lei Yang <leiyang@redhat.com>

On Tue, Sep 16, 2025 at 12:39 AM Jon Kohler <jon@nutanix.com> wrote:
>
>
>
> > On Sep 15, 2025, at 12:03 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > !-------------------------------------------------------------------|
> >  CAUTION: External Email
> >
> > |-------------------------------------------------------------------!
> >
> >
> > Two regressions were reported in vhost-net.
> > This is based on a patchset by Jason, but with
> > patch 2 from his series split up and made simpler.
> >
> > Lightly tested.
> > Jason, Jon could you pls test this as well, and report?
>
> For the iotlb bit (patch 2/3):
> Tried this out on a 6.16 host / guest that locked up with iotlb miss loop,
> applied this patch and all was well.
>
> Tested-by: Jon Kohler <jon@nutanix.com>
>
> >
> >
> > Jason Wang (2):
> >  vhost-net: unbreak busy polling
> >
> > Michael S. Tsirkin (1):
> >  Revert "vhost/net: Defer TX queue re-enable until after sendmsg"
> >  vhost-net: flush batched before enabling notifications
> >
> > drivers/vhost/net.c | 44 ++++++++++++++++++++------------------------
> > 1 file changed, 20 insertions(+), 24 deletions(-)
> >
> > --
> > MST
> >
>