<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/fs/dlm/lowcomms.c, branch v5.16</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>fs: dlm: implement delayed ack handling</title>
<updated>2021-08-19T16:33:03+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-08-18T20:27:13+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=b97f85259fca5accc2cd5f7c4f42fa0dd8efda48'/>
<id>b97f85259fca5accc2cd5f7c4f42fa0dd8efda48</id>
<content type='text'>
This patch changes that we don't ack each message. Lowcomms will take
care about to send an ack back after a bulk of messages was processed.
Currently it's only when the whole receive buffer was processed, there
might better positions to send an ack back but only the lowcomms
implementation know when there are more data to receive. This patch has
also disadvantages that we might retransmit more on errors, however this
is a very rare case.

Tested with make_panic on gfs2 with three nodes by running:

trace-cmd record -p function -l 'dlm_send_ack' sleep 100

and

trace-cmd report | wc -l

Before patch:
- 20548
- 21376
- 21398

After patch:
- 18338
- 20679
- 19949

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch changes that we don't ack each message. Lowcomms will take
care about to send an ack back after a bulk of messages was processed.
Currently it's only when the whole receive buffer was processed, there
might better positions to send an ack back but only the lowcomms
implementation know when there are more data to receive. This patch has
also disadvantages that we might retransmit more on errors, however this
is a very rare case.

Tested with make_panic on gfs2 with three nodes by running:

trace-cmd record -p function -l 'dlm_send_ack' sleep 100

and

trace-cmd report | wc -l

Before patch:
- 20548
- 21376
- 21398

After patch:
- 18338
- 20679
- 19949

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: move receive loop into receive handler</title>
<updated>2021-07-19T16:56:36+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=62699b3f0a62435fceb8debf295e90a5ea259e04'/>
<id>62699b3f0a62435fceb8debf295e90a5ea259e04</id>
<content type='text'>
This patch moves the kernel_recvmsg() loop call into the
receive_from_sock() function instead of doing the loop outside the
function and abort the loop over it's return value.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch moves the kernel_recvmsg() loop call into the
receive_from_sock() function instead of doing the loop outside the
function and abort the loop over it's return value.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: fix multiple empty writequeue alloc</title>
<updated>2021-07-19T16:56:35+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=c51b0221798b695ac1beac6597d3a0acf039384b'/>
<id>c51b0221798b695ac1beac6597d3a0acf039384b</id>
<content type='text'>
This patch will add a mutex that a connection can allocate a writequeue
entry buffer only at a sleepable context at one time. If multiple caller
waits at the writequeue spinlock and the spinlock gets release it could
be that multiple new writequeue page buffers were allocated instead of
allocate one writequeue page buffer and other waiters will use remaining
buffer of it. It will only be the case for sleepable context which is
the common case. In non-sleepable contexts like retransmission we just
don't care about such behaviour.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch will add a mutex that a connection can allocate a writequeue
entry buffer only at a sleepable context at one time. If multiple caller
waits at the writequeue spinlock and the spinlock gets release it could
be that multiple new writequeue page buffers were allocated instead of
allocate one writequeue page buffer and other waiters will use remaining
buffer of it. It will only be the case for sleepable context which is
the common case. In non-sleepable contexts like retransmission we just
don't care about such behaviour.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: generic connect func</title>
<updated>2021-07-19T16:56:34+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:43+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=8728a455d20ddadecd767337475fc1371e031d79'/>
<id>8728a455d20ddadecd767337475fc1371e031d79</id>
<content type='text'>
This patch adds a generic connect function for TCP and SCTP. If the
connect functionality differs from each other additional callbacks in
dlm_proto_ops were added. The sockopts callback handling will guarantee
that sockets created by connect() will use the same options as sockets
created by accept().

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch adds a generic connect function for TCP and SCTP. If the
connect functionality differs from each other additional callbacks in
dlm_proto_ops were added. The sockopts callback handling will guarantee
that sockets created by connect() will use the same options as sockets
created by accept().

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: auto load sctp module</title>
<updated>2021-07-19T16:56:26+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=90d21fc0479dc0b5d338d664ddb55e5017b44f3e'/>
<id>90d21fc0479dc0b5d338d664ddb55e5017b44f3e</id>
<content type='text'>
This patch adds a "for now" better handling of missing SCTP support in
the kernel and try to load the sctp module if SCTP is set.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch adds a "for now" better handling of missing SCTP support in
the kernel and try to load the sctp module if SCTP is set.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: introduce generic listen</title>
<updated>2021-07-19T16:53:43+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=2dc6b1158c28c3a5e86d162628810312f98d5e97'/>
<id>2dc6b1158c28c3a5e86d162628810312f98d5e97</id>
<content type='text'>
This patch combines each transport layer listen functionality into one
listen function. Per transport layer differences are provided by
additional callbacks in dlm_proto_ops.

This patch drops silently sock_set_keepalive() for listen tcp sockets
only. This socket option is not set at connecting sockets, I also don't
see the sense of set keepalive for sockets which are created by accept()
only.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch combines each transport layer listen functionality into one
listen function. Per transport layer differences are provided by
additional callbacks in dlm_proto_ops.

This patch drops silently sock_set_keepalive() for listen tcp sockets
only. This socket option is not set at connecting sockets, I also don't
see the sense of set keepalive for sockets which are created by accept()
only.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: move to static proto ops</title>
<updated>2021-07-19T16:53:43+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=a66c008cd16a122e340d48f01ddf8cd768368717'/>
<id>a66c008cd16a122e340d48f01ddf8cd768368717</id>
<content type='text'>
This patch moves the per transport socket callbacks to a static const
array. We can support only one transport socket for the init namespace
which will be determinted by reading the dlm config at lowcomms_start().

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch moves the per transport socket callbacks to a static const
array. We can support only one transport socket for the init namespace
which will be determinted by reading the dlm config at lowcomms_start().

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: introduce con_next_wq helper</title>
<updated>2021-07-19T16:53:43+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=66d5955a098d504b5a34c35d56a967c536264673'/>
<id>66d5955a098d504b5a34c35d56a967c536264673</id>
<content type='text'>
This patch introduce a function to determine if something is ready to
being send in the writequeue. It's not just that the writequeue is not
empty additional the first entry need to have a valid length field.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch introduce a function to determine if something is ready to
being send in the writequeue. It's not just that the writequeue is not
empty additional the first entry need to have a valid length field.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: clear CF_APP_LIMITED on close</title>
<updated>2021-07-19T16:53:43+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=052849beeab02306d1420d03da959e7db883b375'/>
<id>052849beeab02306d1420d03da959e7db883b375</id>
<content type='text'>
If send_to_sock() sets CF_APP_LIMITED limited bit and it has not been
cleared by a waiting lowcomms_write_space() yet and a close_connection()
apprears we should clear the CF_APP_LIMITED bit again because the
connection starts from a new state again at reconnect.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If send_to_sock() sets CF_APP_LIMITED limited bit and it has not been
cleared by a waiting lowcomms_write_space() yet and a close_connection()
apprears we should clear the CF_APP_LIMITED bit again because the
connection starts from a new state again at reconnect.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>fs: dlm: use sk-&gt;sk_socket instead of con-&gt;sock</title>
<updated>2021-07-19T16:53:43+00:00</updated>
<author>
<name>Alexander Aring</name>
<email>aahringo@redhat.com</email>
</author>
<published>2021-07-16T20:22:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=feb704bd17786c8ff52a49d7759b8ee4f3a5aaac'/>
<id>feb704bd17786c8ff52a49d7759b8ee4f3a5aaac</id>
<content type='text'>
Instead of dereference "con-&gt;sock" we can get the socket structure over
"sk-&gt;sk_socket" as well. This patch will switch to this behaviour.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Instead of dereference "con-&gt;sock" we can get the socket structure over
"sk-&gt;sk_socket" as well. This patch will switch to this behaviour.

Signed-off-by: Alexander Aring &lt;aahringo@redhat.com&gt;
Signed-off-by: David Teigland &lt;teigland@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
