Hi,
> Having the support for vmapping dmabuf's is required to share
> dmabufs to drivers that want CPU access. This is the case of
> a vivid to virtio-gpu pipeline, where the virtio-gpu driver
> exports dmabufs to the video4linux vivid driver.
>
> The first patch adds virtio_gpu_object_kunmap()
On Thu, Sep 20, 2018 at 10:30:06AM -0700, Nadav Amit wrote:
> This patch-set adds the following enhancements to the VMware balloon
> driver:
>
> 1. Balloon compaction support.
> 2. Report the number of inflated/deflated ballooned pages through vmstat.
> 3. Memory shrinker to avoid balloon over-inf
On Tue, 25 Sep 2018 14:13:07 +0200
Halil Pasic wrote:
> The trigger for the series is this bug report:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1788432
>
> Changelog:
> v1 -> v2:
> * improve on commit messages, add cc:stable
> RFC -> v1:
> * do mutual exclusion on a per device bas
From: Tonghao Zhang
This patch improves the guest receive performance.
On the handle_tx side, we poll the sock receive queue at the
same time. handle_rx do that in the same way.
We set the poll-us=100us and use the netperf to test throughput
and mean latency. When running the tests, the vhost-ne
From: Tonghao Zhang
Use the VHOST_NET_VQ_XXX as a subclass for mutex_lock_nested.
Signed-off-by: Tonghao Zhang
Acked-by: Jason Wang
---
drivers/vhost/net.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 1bff6bc..5fe57a
From: Tonghao Zhang
Factor out generic busy polling logic and will be
used for in tx path in the next patch. And with the patch,
qemu can set differently the busyloop_timeout for rx queue.
To avoid duplicate codes, introduce the helper functions:
* sock_has_rx_data(changed from sk_has_rx_data)
*
From: Tonghao Zhang
This patch changes the way that lock all vqs
at the same, to lock them one by one. It will
be used for next patch to avoid the deadlock.
Signed-off-by: Tonghao Zhang
Acked-by: Jason Wang
Signed-off-by: Jason Wang
---
drivers/vhost/vhost.c | 24 +++-
1
From: Tonghao Zhang
This patches improve the guest receive performance.
On the handle_tx side, we poll the sock receive queue
at the same time. handle_rx do that in the same way.
For more performance report, see patch 4
Tonghao Zhang (4):
net: vhost: lock the vqs one by one
net: vhost: repl