<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/drivers/vhost/net.c, branch v3.8-rc2</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>vhost-net: enable zerocopy tx by default</title>
<updated>2012-12-06T15:09:18+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-12-06T12:56:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=f9611c43ab0ddaf547b395c90fb842f55959334c'/>
<id>f9611c43ab0ddaf547b395c90fb842f55959334c</id>
<content type='text'>
Zero copy TX has been around for a while now.
We seem to be down to eliminating theoretical bugs
and performance tuning at this point:
it's probably time to enable it by default so that
most users get the benefit.

Keep the flag around meanwhile so users can experiment
with disabling this if they experience regressions.
I expect that we will remove it in the future.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Zero copy TX has been around for a while now.
We seem to be down to eliminating theoretical bugs
and performance tuning at this point:
it's probably time to enable it by default so that
most users get the benefit.

Keep the flag around meanwhile so users can experiment
with disabling this if they experience regressions.
I expect that we will remove it in the future.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: skip head management if no outstanding</title>
<updated>2012-12-06T15:09:18+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-12-06T15:00:18+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=cedb9bdce099206290a2bdd02ce47a7b253b6a84'/>
<id>cedb9bdce099206290a2bdd02ce47a7b253b6a84</id>
<content type='text'>
For short packets zerocopy mode adds overhead
of managing heads which isn't necessary: we
could simly update used ring directly
same as with zerocopy disabled.

Things seem to run a bit faster if we detect
and bypass head management when zcopy isn't used.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For short packets zerocopy mode adds overhead
of managing heads which isn't necessary: we
could simly update used ring directly
same as with zerocopy disabled.

Things seem to run a bit faster if we detect
and bypass head management when zcopy isn't used.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: flush outstanding DMAs on memory change</title>
<updated>2012-12-06T15:09:18+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-12-03T22:17:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=1280c27f8e29acf4af2da914e80ec27c3dbd5c01'/>
<id>1280c27f8e29acf4af2da914e80ec27c3dbd5c01</id>
<content type='text'>
When memory map changes, we need to flush outstanding
DMAs as they might in theory reference old memory addresses.
To do this simply stop initiating new DMAs
and wait for ubufs ref count to drop to 0.
Afterwards reset the count back to 1.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When memory map changes, we need to flush outstanding
DMAs as they might in theory reference old memory addresses.
To do this simply stop initiating new DMAs
and wait for ubufs ref count to drop to 0.
Afterwards reset the count back to 1.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost: avoid backend flush on vring ops</title>
<updated>2012-12-06T15:09:18+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-12-06T12:03:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=935cdee7ee159569b0aaa10bd9244660f6672b08'/>
<id>935cdee7ee159569b0aaa10bd9244660f6672b08</id>
<content type='text'>
vring changes already do a flush internally where appropriate, so we do
not need a second flush.

It's currently not very expensive but a follow-up patch makes flush more
heavy-weight, so remove the extra flush here to avoid regressing
performance if call or kick fds are changed on data path.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
vring changes already do a flush internally where appropriate, so we do
not need a second flush.

It's currently not very expensive but a follow-up patch makes flush more
heavy-weight, so remove the extra flush here to avoid regressing
performance if call or kick fds are changed on data path.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: initialize zcopy packet counters</title>
<updated>2012-12-03T18:42:17+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-12-03T07:31:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=64e9a9b8a0eef9cbd9f86f0e79fc4eeea24abe76'/>
<id>64e9a9b8a0eef9cbd9f86f0e79fc4eeea24abe76</id>
<content type='text'>
These packet counters are used to drive the zercopy
selection heuristic so nothing too bad happens if they are off a bit -
and they are also reset once in a while.
But it's cleaner to clear them when backend is set so that
we start in a known state.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
These packet counters are used to drive the zercopy
selection heuristic so nothing too bad happens if they are off a bit -
and they are also reset once in a while.
But it's cleaner to clear them when backend is set so that
we start in a known state.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net</title>
<updated>2012-11-10T23:32:51+00:00</updated>
<author>
<name>David S. Miller</name>
<email>davem@davemloft.net</email>
</author>
<published>2012-11-10T23:32:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=d4185bbf62a5d8d777ee445db1581beb17882a07'/>
<id>d4185bbf62a5d8d777ee445db1581beb17882a07</id>
<content type='text'>
Conflicts:
	drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c

Minor conflict between the BCM_CNIC define removal in net-next
and a bug fix added to net.  Based upon a conflict resolution
patch posted by Stephen Rothwell.

Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Conflicts:
	drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c

Minor conflict between the BCM_CNIC define removal in net-next
and a bug fix added to net.  Based upon a conflict resolution
patch posted by Stephen Rothwell.

Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: reduce vq polling on tx zerocopy</title>
<updated>2012-11-03T01:29:58+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-11-01T09:16:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=24eb21a14896e5c1c7cbca8e9a40405a51745f1d'/>
<id>24eb21a14896e5c1c7cbca8e9a40405a51745f1d</id>
<content type='text'>
It seems that to avoid deadlocks it is enough to poll vq before
 we are going to use the last buffer.  This is faster than
c70aa540c7a9f67add11ad3161096fb95233aa2e.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
It seems that to avoid deadlocks it is enough to poll vq before
 we are going to use the last buffer.  This is faster than
c70aa540c7a9f67add11ad3161096fb95233aa2e.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: select tx zero copy dynamically</title>
<updated>2012-11-03T01:29:58+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-11-01T09:16:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=eaae8132ef6032ffe612e789e8ff145dcf5bc2bb'/>
<id>eaae8132ef6032ffe612e789e8ff145dcf5bc2bb</id>
<content type='text'>
Even when vhost-net is in zero-copy transmit mode,
net core might still decide to copy the skb later
which is somewhat slower than a copy in user
context: data copy overhead is added to the cost of
page pin/unpin. The result is that enabling tx zero copy
option leads to higher CPU utilization for guest to guest
and guest to host traffic.

To fix this, suppress zero copy tx after a given number of
packets triggered late data copy. Re-enable periodically
to detect workload changes.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Even when vhost-net is in zero-copy transmit mode,
net core might still decide to copy the skb later
which is somewhat slower than a copy in user
context: data copy overhead is added to the cost of
page pin/unpin. The result is that enabling tx zero copy
option leads to higher CPU utilization for guest to guest
and guest to host traffic.

To fix this, suppress zero copy tx after a given number of
packets triggered late data copy. Re-enable periodically
to detect workload changes.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost: move -net specific code out</title>
<updated>2012-11-03T01:29:58+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-11-01T09:16:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=b211616d712551874db3ce0fb44196f6faad2c34'/>
<id>b211616d712551874db3ce0fb44196f6faad2c34</id>
<content type='text'>
Zerocopy handling code is vhost-net specific.
Move it from vhost.c/vhost.h out to net.c

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Zerocopy handling code is vhost-net specific.
Move it from vhost.c/vhost.h out to net.c

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-net: cleanup macros for DMA status tracking</title>
<updated>2012-11-03T01:29:57+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2012-11-01T09:16:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=70e4cb9aafb3caa9e5eb25614a5bc40b0ffa9ffd'/>
<id>70e4cb9aafb3caa9e5eb25614a5bc40b0ffa9ffd</id>
<content type='text'>
Better document macros for DMA tracking. Add an
explicit one for DMA in progress instead of
relying on user supplying len != 1.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Better document macros for DMA tracking. Add an
explicit one for DMA in progress instead of
relying on user supplying len != 1.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</pre>
</div>
</content>
</entry>
</feed>
