<feed xmlns='http://www.w3.org/2005/Atom'>
<title>nng, branch tcp-dialer-local-test</title>
<subtitle>A mirror of https://github.com/nanomsg/nng
</subtitle>
<id>https://cogarr.net/source/cgit.cgi/nng/atom?h=tcp-dialer-local-test</id>
<link rel='self' href='https://cogarr.net/source/cgit.cgi/nng/atom?h=tcp-dialer-local-test'/>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/'/>
<updated>2024-12-31T07:27:03Z</updated>
<entry>
<title>pipes: fix crash caused by use after free race in rejection</title>
<updated>2024-12-31T07:27:03Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T07:25:13Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=5f7a9fe784f3d4074b9c6ad750f05f29966097ef'/>
<id>urn:sha1:5f7a9fe784f3d4074b9c6ad750f05f29966097ef</id>
<content type='text'>
</content>
</entry>
<entry>
<title>sockaddr: ingore the sin6_len field</title>
<updated>2024-12-31T07:07:32Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T07:07:32Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=e183d733b4c3a658c011f1e3c06fbbd03b5e97db'/>
<id>urn:sha1:e183d733b4c3a658c011f1e3c06fbbd03b5e97db</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Use AF instead of PF for socket address families</title>
<updated>2024-12-31T06:29:08Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T06:29:08Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=c30939ab17ba52e5404c2a97949a1f515730a542'/>
<id>urn:sha1:c30939ab17ba52e5404c2a97949a1f515730a542</id>
<content type='text'>
</content>
</entry>
<entry>
<title>tcp dialer tests: add local address tests</title>
<updated>2024-12-31T06:28:32Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T03:59:27Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=6ff9e9beb18ee46e64d7bf929baf57590f23c789'/>
<id>urn:sha1:6ff9e9beb18ee46e64d7bf929baf57590f23c789</id>
<content type='text'>
</content>
</entry>
<entry>
<title>poll poller: support concurrency for pollers using poll()</title>
<updated>2024-12-31T02:45:46Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T02:45:46Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=071de8a7049839b5ed9a7478f29c8921455ae50a'/>
<id>urn:sha1:071de8a7049839b5ed9a7478f29c8921455ae50a</id>
<content type='text'>
</content>
</entry>
<entry>
<title>poller: port events scalability (illumos/Solaris)</title>
<updated>2024-12-31T02:45:35Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T02:38:06Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=8dbcb5def2656798af46da835a90421a99cd6f36'/>
<id>urn:sha1:8dbcb5def2656798af46da835a90421a99cd6f36</id>
<content type='text'>
</content>
</entry>
<entry>
<title>fixes #530 POSIX pollqs should scale horizontally (epoll)</title>
<updated>2024-12-31T02:01:20Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T01:55:47Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=90faa53628ebef4b800a9dccd6cf36f52916ebd2'/>
<id>urn:sha1:90faa53628ebef4b800a9dccd6cf36f52916ebd2</id>
<content type='text'>
This should help Linux platforms scale even further with NNG.

Port events and *maybe* poll are the last to do.  Probably select
will remain left in the cold, because honestly select based systems
are already performance constrained.
</content>
</entry>
<entry>
<title>kqueue: support concurrent pollers</title>
<updated>2024-12-31T01:42:32Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-31T01:42:32Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=0fb33189287f643b98760a663b75e6016537878a'/>
<id>urn:sha1:0fb33189287f643b98760a663b75e6016537878a</id>
<content type='text'>
This allows greatly increased scalability for kqueue based systems
with lots of cores (more likely FreeBSD than Darwin, as most macs
only have a smattering of cores), but even for macs we can engage
a few cores for system calls giving improvements.

The implementation here is pretty simple -- each file descriptor
gets assigned to its own kqueue by taking the numeric value of the
file descriptor modulo the number of kqueues we have opened.

The same approach will be adopted for epoll and Solaris/illumos port events.
</content>
</entry>
<entry>
<title>ipc test: add a case for IPC that never connects</title>
<updated>2024-12-30T23:47:04Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-30T23:39:50Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=7f7a8194a60e9f5883866dd8b8c22d4576fc1abc'/>
<id>urn:sha1:7f7a8194a60e9f5883866dd8b8c22d4576fc1abc</id>
<content type='text'>
This involved test-specific hacks, since connect() for UNIX domain
sockets always completes synchronously one way or the other, even though
it is documented that it might not.  This found a bug, with an uninitialized
poll FD as well!
</content>
</entry>
<entry>
<title>ipc/tcp stream tests: use dup() to avoid confusion and hang in teardown</title>
<updated>2024-12-30T22:27:43Z</updated>
<author>
<name>Garrett D'Amore</name>
<email>garrett@damore.org</email>
</author>
<published>2024-12-30T22:27:43Z</published>
<link rel='alternate' type='text/html' href='https://cogarr.net/source/cgit.cgi/nng/commit/?id=a74d853241fa21ef4b6352fd39c25b4d4c6a4ab3'/>
<id>urn:sha1:a74d853241fa21ef4b6352fd39c25b4d4c6a4ab3</id>
<content type='text'>
This was observed only in the select poller, but the behavior of having
the same file descriptor registered in the poller more than once is undefined.
</content>
</entry>
</feed>
