summaryrefslogtreecommitdiff
path: root/regress/sys
AgeCommit message (Collapse)Author
2014-10-30Add some pkill and sleep to make socket splicing performance testsAlexander Bluhm
more reliable.
2014-10-18Simple malloc() to reallocarray().Doug Hogan
ok deraadt@
2014-10-09Add POLLHUP to expected revents for POLLIN on writer EOF.Todd C. Miller
2014-10-08Add tests for EOF behavior (which currently fail) and expected output.Todd C. Miller
The expected output for the POLLIN test could also include POLLHUP, though it does not currently do so.
2014-09-27Fix mmap() calls that check for a result other than MAP_FAILED.Doug Hogan
ok tedu@
2014-08-27Regression tests for setuid-and-friends.Bret Lambert
Thanks to djm@ for good suggestions.
2014-08-18Make the perl modules consistent for the multiple regression tests.Alexander Bluhm
This includes coding style, better error messages and variable naming.
2013-01-03Reengineer the socket splicing regression tests:Alexander Bluhm
- Move the tests from splice to new sosplice directory for consistent naming. - Split the API tests and the TCP splicing tests into separate directories. - Create some tests for the upcoming UDP splicing. - Tests can be run in obj directories now. - The API tests can run both on a local and on a remote machine now. - Fix the forking TCP tests which splice and read or write simultaneously. - Bunch of little fixes for races in the tests. - Deduplicate code, move checks into common functions.
2014-08-18Add a description of what the relayd and sosplice regression testsAlexander Bluhm
do. Also explain the optional environment variables.
2014-08-18Make out of band data tests more reliable. As multiple tcp urgentAlexander Bluhm
bytes cannot be handled correctly, allow another variation.
2014-08-18Make the forking tests more reliable. During copy the read(2)Alexander Bluhm
system call might fail with EAGAIN when the other forked process has already read the data.
2014-08-18Make the multiple UDP packet test more reliable. The receive bufferAlexander Bluhm
limit also counts the mbufs in it, so packets could be dropped. Increase both send and receive socket buffer size.
2014-08-15Convert 3DES regress test from /dev/crypto to using kernel source codeMike Belopuhov
directly with a simplified CBC implementation.
2014-08-15Convert AES-CTR regress test from /dev/crypto to using kernel sourceMike Belopuhov
code directly. This test is converted the same way jsing@ has recently converted an XTS test by pulling in xform.c code.
2014-08-15Convert AES regress test from /dev/crypto to using kernel source codeMike Belopuhov
directly. This test case uses ECB vectors, therefore no chaining is required.
2014-07-20Make sure the correct errno is reported by warn* or err* and notPhilip Guenther
the errno of an intervening cleanup operation like close/unlink/etc. Diff from Doug Hogan (doug (at) acyclic.org)
2014-07-19Make sure struct sockaddr_in gets completely initialized by setting it to zeroMiod Vallat
before setting the few fields we are interested in; Doug Hogan
2014-07-18The pf forward tests were running rdr-to and nat-to simultaneouslyAlexander Bluhm
only. Change address layout and add individual tests for each feature rdr-to and nat-to and rdr-to together with nat-to.
2014-07-16zap trailing newlines; "go for it" deraadtOkan Demirmen
2014-07-13Make the pf_fragment test pass again. pf does not adjust theAlexander Bluhm
checksum of a NATed UDP or TCP packet in the payload of an ICMP packet anymore. Disable all test that rely on this feature. Check that the router's interface mtu has been set to 1300.
2014-07-13Make the pf_forward test pass again. Check that the router'sAlexander Bluhm
interface mtu has been set to 1300. New netcat needs -N for shutdown. The IPv4 and IPv6 addresses for the af-to tests must have corresponding host numbers.
2014-07-12Automatically load the required pf rules into the kernel of theAlexander Bluhm
remote test machine.
2014-07-12Teach these pf regress tests to run with obj directory and to failAlexander Bluhm
softly if the setup is incomplete. Link them to the build.
2014-07-12Use a saner and consistent address layout for my network tests thatAlexander Bluhm
run over 4 machines. Add a check-setup target to verify that interface addresses and routes are properly set up.
2014-07-11Put a license file into the test directories with multiple argumentAlexander Bluhm
files. We do not want to put a license header into each subtest. suggested by reyk@
2014-07-11Make the test output easier to read by splitting the test casesAlexander Bluhm
with an empty line and a header line.
2014-07-11Verify the checksum of the echo reply fragment based on the echoAlexander Bluhm
request.
2014-07-11Add a test that sends a large ICMP6 packet in many small fragments.Alexander Bluhm
The echo reply has to be fragmented.
2013-01-03Reengineer the socket splicing regression tests:Alexander Bluhm
- Move the tests from splice to new sosplice directory for consistent naming. - Split the API tests and the TCP splicing tests into separate directories. - Create some tests for the upcoming UDP splicing. - Tests can be run in obj directories now. - The API tests can run both on a local and on a remote machine now. - Fix the forking TCP tests which splice and read or write simultaneously. - Bunch of little fixes for races in the tests. - Deduplicate code, move checks into common functions.
2014-07-03Enter the REGRESS_FULL target subdirs for clean, cleandir and obj targets.Miod Vallat
2014-06-24POSIX doesn't specify MAP_FILE and it's a noop anyway, so remove it toMatthew Dempsky
make test portable to other OSes.
2014-06-23Add regress test for SIGSEGV and SIGBUS siginfo semantics.Matthew Dempsky
According to POSIX, SIGSEGV should specify SEGV_ACCERR if the memory pages are mapped, but the protections don't match the user's access attempts, while SEGV_MAPERR should only be specified for pages that are unmapped. Some platforms currently handle this correctly, but not all. Additionally, SIGBUS/BUS_ADRERR should be generated instead of SIGSEGV for access to file mapped pages that exceed the end of the file. (Thanks to kettenis@ for suggesting this test.) Currently failing, so not linked into the regress tree yet.
2014-06-19Add a basic regress test for __MAP_NOREPLACE.Matthew Dempsky
2014-06-13Add regress test for MAP_INHERIT_ZERO.Matthew Dempsky
2014-05-29use calloc, from Benjamin BaierTed Unangst
2014-05-20Use errc/warnc to simplify code.Philip Guenther
Also, in 'ftp', always put the error message last, after the hostname/ipaddr. ok jsing@ krw@ millert@
2014-04-27Regress for access()Philip Guenther
2014-04-24The switch statement given/when was introduced in Perl 5.10 and gotAlexander Bluhm
experimental in Perl 5.18. So we see warnings in the tests. given is experimental at funcs.pl line 62. when is experimental at funcs.pl line 63. Just suppress this type of warnings for now. If Perl will disable smart match, the simple regression code still works. If they will choose to remove given/when, we can fix it later. Discussed with and OK andre@ afresh1@
2014-04-18It's been a quarter century: we can assume volatile is present with that name.Philip Guenther
2014-03-29pflowproto 9 is no more.Florian Obser
While there fix regression for inaddr_any flowsrc.
2014-03-18Retire hp300, mvme68k and mvme88k ports. These ports have no users, keepingMiod Vallat
this hardware alive is becoming increasingly difficult, and I should heed the message sent by the three disks which have died on me over the last few days. Noone sane will mourn these ports anyway. So long, and thanks for the fish.
2014-02-28... also check that writing 0 bytes works...Marc Espie
2014-01-22Missing #includePhilip Guenther
2014-01-18Remove -Wbounded: it is now the compiler default.Martynas Venckus
2014-01-10Use arc4random instead of random in the flock regress tests.Joel Sing
ok deraadt@
2014-01-10Make this work on hppa.Joel Sing
2014-01-10Do not include MD assembly code in a sys regress test. Untested on alpha,Joel Sing
however it has a much better chance of compiling than it did previously.
2014-01-08Fix the tests for the pf divert state and socket reuse.Alexander Bluhm
2014-01-08Set socket buffer size to fixed value of 100000 octets. ThisAlexander Bluhm
prevents that the socket splicing performance test interferes with the dynamic TCP buffer size adjusting.
2013-12-21Match the type to the callsPhilip Guenther
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-22 21:03:53 +0000