<feed xmlns='http://www.w3.org/2005/Atom'>
<title>src/sys/netinet, branch release/2.2.8</title>
<subtitle>FreeBSD source tree</subtitle>
<id>https://cgit-dev.freebsd.org/src/atom?h=release%2F2.2.8</id>
<link rel='self' href='https://cgit-dev.freebsd.org/src/atom?h=release%2F2.2.8'/>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/'/>
<updated>1998-11-25T22:19:21Z</updated>
<entry>
<title>MFC: Fix the bind security fix introduced in rev 1.38 to work with multicast.</title>
<updated>1998-11-25T22:19:21Z</updated>
<author>
<name>Nate Williams</name>
<email>nate@FreeBSD.org</email>
</author>
<published>1998-11-25T22:19:21Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=4e55c42d7271641a753750ea89d1b9081ec7c28a'/>
<id>urn:sha1:4e55c42d7271641a753750ea89d1b9081ec7c28a</id>
<content type='text'>
[ This bug allows Multicast in the JDK to work ]
</content>
</entry>
<entry>
<title>MFC: {rev 1.82} copy M_MCAST bit into fragmented packets.</title>
<updated>1998-11-20T03:09:02Z</updated>
<author>
<name>Joseph Koshy</name>
<email>jkoshy@FreeBSD.org</email>
</author>
<published>1998-11-20T03:09:02Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=2b2995aa3d89ce29a5c9165e05797f6b0c1f484a'/>
<id>urn:sha1:2b2995aa3d89ce29a5c9165e05797f6b0c1f484a</id>
<content type='text'>
Approved by:	wollman
</content>
</entry>
<entry>
<title>Brought in fix from rev 1.105: pullup entire header for fragments.</title>
<updated>1998-11-11T21:20:08Z</updated>
<author>
<name>David Greenman</name>
<email>dg@FreeBSD.org</email>
</author>
<published>1998-11-11T21:20:08Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=bb92371a61333e456ea2ff0b2df92331bcf18732'/>
<id>urn:sha1:bb92371a61333e456ea2ff0b2df92331bcf18732</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Make dummynet use m_pkthdr.len, not ip_len, for computing</title>
<updated>1998-10-16T15:00:16Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-10-16T15:00:16Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=c16c35d47d35ed89ac4349d8ea8e23ba0ac44f65'/>
<id>urn:sha1:c16c35d47d35ed89ac4349d8ea8e23ba0ac44f65</id>
<content type='text'>
bandwidths. This is necessary for correct operation at the bridging
level. Problem reported by Richard B. Ernst
</content>
</entry>
<entry>
<title>Fix a potential panic when deleting ipfw rules and packets are</title>
<updated>1998-10-14T16:29:58Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-10-14T16:29:58Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=e08c1d198ec86aefdb607e65f791ca89837a0b4d'/>
<id>urn:sha1:e08c1d198ec86aefdb607e65f791ca89837a0b4d</id>
<content type='text'>
trapped into a pipe, and you allow multiple passees through the
firewall code with dummynet. It has never hit anyone on -stable because
net.inet.ip.fw.one_pass=1 by default.

Reported by  Philippe Regnauld (who else!) while beating the
3.0 version of dummynet.
</content>
</entry>
<entry>
<title>Restore default behaviour for skipto rules, i.e. jump to the first rule</title>
<updated>1998-10-06T09:55:01Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-10-06T09:55:01Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=01e7047b54ade417303b0b08541fe20d31c3f596'/>
<id>urn:sha1:01e7047b54ade417303b0b08541fe20d31c3f596</id>
<content type='text'>
greater or equal to the jump target.
</content>
</entry>
<entry>
<title>Fix the ipfw/natd problem for which people has been shouting at me for the</title>
<updated>1998-09-21T08:55:22Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-09-21T08:55:22Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=efee84f064bc8a812612341d46db11b83cf64b7b'/>
<id>urn:sha1:efee84f064bc8a812612341d46db11b83cf64b7b</id>
<content type='text'>
last few days. It was a stupid uninitialized variable, and the bug was
generated when i cleaned up the code removing redundant(!) initializations
before the commit.

Thanks to Studded@dal.net and Allan Saddi who provided traces to
help track down the problem.
</content>
</entry>
<entry>
<title>remove a diagnostic message and fix statistics when using ipfw on</title>
<updated>1998-09-18T20:53:08Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-09-18T20:53:08Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=fda888c25a008ef0466672582a17a414b59236e6'/>
<id>urn:sha1:fda888c25a008ef0466672582a17a414b59236e6</id>
<content type='text'>
bridged packets
</content>
</entry>
<entry>
<title>bring DUMMYNET and BRIDGE support into -stable</title>
<updated>1998-09-17T18:02:33Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-09-17T18:02:33Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=18f405451d1c66af0cc711c501fa970e527eb458'/>
<id>urn:sha1:18f405451d1c66af0cc711c501fa970e527eb458</id>
<content type='text'>
decouple BPF and PROMISC handling on some if drivers
make ipstat available through sysctl (already in -current)

NOTE: you have to recompile ipfw!
</content>
</entry>
<entry>
<title>introduce a sysctl variable to disable slow-start on "local"</title>
<updated>1998-09-17T04:35:58Z</updated>
<author>
<name>Luigi Rizzo</name>
<email>luigi@FreeBSD.org</email>
</author>
<published>1998-09-17T04:35:58Z</published>
<link rel='alternate' type='text/html' href='https://cgit-dev.freebsd.org/src/commit/?id=cbed57e63091c6194ac041aa6c98a1d3388c37d4'/>
<id>urn:sha1:cbed57e63091c6194ac041aa6c98a1d3388c37d4</id>
<content type='text'>
tcp connections. slow-start is used by default (as it should be)
This fixes a misbehaviour that was introduced when T/TCP was
integrated in the main sources.
</content>
</entry>
</feed>
