src - OpenBSD base system

Age	Commit message (Collapse)	Author
2017-01-10	Cleanup the mount point and vnd devices before and after the test.	Alexander Bluhm
	Use objdir and curdir correctly. Link ffs regress to the build.
2016-12-13	Print better error messages in case an expected packet is missing.	Alexander Bluhm

2016-12-12	Set promiscuous mode during srp1() to receive the arp reply directed	Alexander Bluhm
	to the ethernet multicast address. This fixes the test on my hardware setup, it was only passing in qemu before. At least scapy 2.3.3p1 is needed to support the promisc=1 parameter here.
2016-11-30	Fix typos	Vincent Gross

2016-11-28	Enable the pmtu and traceroute subtests with af-to. pf has been	Alexander Bluhm
	fixed now.
2016-11-28	The setuid regression test builds and runs a binary that is setuid	Alexander Bluhm
	or setgid nobody. Since /usr/obj is 0770, user nobody cannot access other files there anymore. Install all programs into a temporary directory and run them there. Check that /tmp is mounted without nosuid.
2016-11-21	Print better error message if test fails.	Alexander Bluhm

2016-11-21	Test that IPv6 framents with invalid length or more-fragment bit	Alexander Bluhm
	drop the whole fragment cache.
2016-11-20	SRPL_FIRST() and SRPL_FOLLOW().	Martin Pieuchot

2016-11-17	Put all make targets in a loop and print a little more what each	Alexander Bluhm
	IPv6 fragment header test is doing.
2016-11-17	Send fragment with Hop-by-Hop extension header after the fragment	Alexander Bluhm
	header. I must not be processed.
2016-11-16	Test multiple combinations of divert connections. The socket cleanup	Alexander Bluhm
	of the first connection must remove the pf state so that the second connection can succeed. Disable two raw IP tests that do not work with the current implementation in the kernel.
2016-11-15	Mark pf states created by this regress test with a label. Remove	Alexander Bluhm
	all such states before basic testing to make sure that these subtests are independent.
2016-11-15	Sync rdomain_is_used()	Martin Pieuchot

2016-11-15	rtable_init() now creates the default routing table.	Martin Pieuchot

2016-11-15	Fix rdomain_is_used() now that lo(4) is created automagically.	Martin Pieuchot

2016-11-15	Check for correct flags now that ::1 is set automagically.	Martin Pieuchot

2016-11-14	Regression test for creating a rdomain default loopback interface.	Martin Pieuchot
	Even if a lo(4) interface will now be created for every rdomain, we must ensure that condiguring such interface via netstart(8) still work, for example: $ cat /etc/hostname.lo1 rdomain 1 127.0.0.1 inet6 ::1/128 alias 88.76.0.3/32
2016-11-09	Remove the path MTU workaround in the af-to tcp test. This subtest	Alexander Bluhm
	passes without it. For ping with af-to the expected mtu is special. Adapt although pf is still broken here and this subtest is disabled.
2016-11-09	Split the pf forwarding test into more subtests to make debugging	Alexander Bluhm
	specific failures easier.
2016-11-09	Use netstat -n in target check-setup to avoid DNS timeout.	Alexander Bluhm
	Requested by mpi@
2016-11-09	Use variable REMOTE_SSH to check the setup of the remote machine.	Alexander Bluhm
	This was a copy & paste bug from another test. Found by mpi@.
2016-11-05	Add regress tests for multicasts and dynamic vxlans	Vincent Gross

2016-10-26	In addition to forwarding and path MTU discovery with pf, check	Alexander Bluhm
	that ICMP Time Exceeded packets are generated. Traceroute output from packets forwarded through pf and a router must contain 3 hops and every hop must respond. Do not test with af-to and reply-to rules as this does not work right now.
2016-10-26	Cleanup pflow test before fixing it. Be more verbose to make	Alexander Bluhm
	debugging easier. Handle EINPROGRESS connect(2) error correctly. Perl pack Q fails on 32 bit machines, add workaround. Restore interfaces in END block, works also if script dies. OK florian@
2016-10-24	Add regress tests for carp(4).	YASUOKA Masahiko

2016-10-20	During sniffing filter strictly on icmp6 "packet too big" to avoid	Alexander Bluhm
	that "neighbor discovery" confuses the test.
2016-10-20	By removing all the sleeps I created a race in the pf tests. The	Alexander Bluhm
	packet was sent and its reply received before the sniffer was up and running. So sleep a second after starting the sniffer thread but before sending the packet. With pf_forward 0m51.54s, pf_fragment 1m33.59s, pf_state 0m03.64s the test are still faster than with forking.
2016-10-20	Replace fork() and sleep() with a Python thread for sniffing packets.	Alexander Bluhm
	This reduces test execution time from 2m21.95s to 1m09.80s.
2016-10-20	Replace fork() and sleep() with a Python thread for sniffing packets.	Alexander Bluhm
	This reduces test execution time from 1m20.34s to 0m37.32s.
2016-10-20	Make the test faster. Move all the packet matching code into the	Alexander Bluhm
	pcap filter. That means the first packet sniffed is the correct one. In the success case, we can stop without waiting for a timeout. OK sashan@
2016-10-19	Add a test suite for pf state handling. It needs a remote machine	Alexander Bluhm
	to exchange packets, the remote kernel is tested. For now it only contains a challenge ack test written by sashan@. OK sashan@
2016-10-19	Link pf_state test to build.	Alexander Bluhm

2016-10-19	Add a test suite for pf state handling. It needs a remote machine	Alexander Bluhm
	to exchange packets, the remote kernel is tested. For now it only contains a challenge ack test written by sashan@. OK sashan@
2016-10-19	The clean:, cleandir:, and obj: targets of src/Makefile descend into	Theo Buehler
	regress/ and run python to set the IMPORT_PYTHON variable in several test Makefiles. This results in W^X violations during 'make build'. Avoid doing that. ok bluhm
2016-10-19	Do not change routes during test run, but expect that everything	Alexander Bluhm
	is setup correctly before. Use make check-setup to verify that.
2016-10-10	Move check for sysctl settings from shell script to makefile to	Alexander Bluhm
	handle skipping consistently like other tests.
2016-10-07	Add simple regress test for vxlan(4) and etherip(4).	YASUOKA Masahiko
	ok bluhm
2016-10-05	Link the setuid regress test to the build in a way that the obj	Alexander Bluhm
	directory gets created. But do not run the test during a global make regress, as the test puts a setuid binary into the object directory. Also remove the setuid bit after the test has been run manually. OK otto@
2016-10-03	The setuid tests needs an obj directory that allows to run programs	Alexander Bluhm
	with set user id. Skip the test if the current filesystem has set the nosuid mount option. OK otto@
2016-09-30	crypto(4) is no more	Otto Moerbeek

2016-09-28	If stdout is not line buffered, messages got lost when assert() was	Alexander Bluhm
	called. Print errors to stderr instead.
2016-09-24	On sparc64 we need "retl" not "ret" to do an immediate return.	Mark Kettenis
	ok deraadt@, tedu@
2016-09-24	recpect W^X, sparc64 still has a problem: non-exec mmap: execute	Otto Moerbeek
	#1 is done, the others not. ok kettenis@ millert@
2016-09-23	Making the regress target depends on building PROG.	Alexander Bluhm

2016-09-23	Zap mquery test for good	Otto Moerbeek

2016-09-23	Zap. Has been disabled for a long time and makes unholy assumptions about	Otto Moerbeek
	the address pace layout. ok bluhhm@
2016-09-21	Use /dev/console as a template for the pseudo tty, stdin may be not	Alexander Bluhm
	a terminal. Check wether tcgetattr() can read the values. Do not only verify that kqueue does not report unwanted events but also that expected events exist.
2016-09-21	Enter rtable	Martin Pieuchot

2016-09-21	Test ART implementation by default.	Martin Pieuchot
	RADIX code can be tested by doing "make RADIX=1"