summaryrefslogtreecommitdiff
path: root/regress/sys
AgeCommit message (Collapse)Author
2019-06-24Do not use a warning flag that base gcc does not understand,Visa Hankala
and fix the warnings. From Moritz Buhl
2019-06-24Fix test after kernel interface change.Visa Hankala
From Moritz Buhl
2019-06-17improve clock_gettime regress:anton
* fix compilation error due to missing cast * do not run sysctl during make clean and the other common regress cleanup targets * make it work using a separate object directory
2019-06-14new regress: clock_gettime(2); test if CLOCK_MONOTONIC is really monotoniccheloha
suggested by kettenis@, tweaked by anton@. ok mpi@ jca@
2019-06-04To avoid SYN retransmit, kill local tcp pf state that will be reused.Alexander Bluhm
This makes the state reuse tests faster and more reliable.
2019-05-30A test for W|X ^ syscall prevention, which is difficult to incorporateTheo de Raadt
properly into regress, because the wxneeded binary must be executed from a wxallowed filesystem..
2019-05-29Add some more relative path testsBob Beck
2019-05-19pass a context struct to each test function in order to reduce future churnanton
2019-05-17Add realpath to regressBob Beck
2019-05-15Add a regression test for path discovery outside of unveil() whichBob Beck
we should eventually fix
2019-05-14test different buffer sizesanton
2019-05-13Document our currently incosnsistent behaviour and our currentBob Beck
sins which have been judged by Dr Posix as sinful...
2019-05-13Link regress sigpthread to build.Alexander Bluhm
2019-05-13Test the interaction of signals with multiple posix threads. ItAlexander Bluhm
covers blocking with signal mask, killing process or thread, invoking handler or waiting for signal.
2019-05-12Remove redundant settings.Alexander Bluhm
2019-05-11let's check /bin/c++Bob Beck
2019-05-10Enable the test on octeon. Ignore the fact that the test cannot workVisa Hankala
on CN3xxx where the TLB does not implement the execute inhibit feature. OK guenther@
2019-05-10Let the test build on mips64.Visa Hankala
OK guenther@
2019-05-10Python for scapy must be executed a bit differently in this test.Alexander Bluhm
2019-05-09Test IPv6 raw sockets with checksum calculation in kernel.Alexander Bluhm
2019-05-10Simplify the detection for installed python and scapy. RefactorAlexander Bluhm
these tests to make them similar. Use the setup and cleanup variables.
2019-05-09Drop reference to sparc and expand 'VAC' so the comment makes sensePhilip Guenther
to semi-gurus
2019-05-09Test IPv6 raw sockets with checksum calculation in kernel.Alexander Bluhm
2019-05-09Link rip6cksum regress to build.Alexander Bluhm
2019-05-09Test IPv6 raw sockets with checksum calculation in kernel.Alexander Bluhm
2019-05-08Add another case for robert@ :)Bob Beck
2019-05-08Mutex functions have no __ prefix anymore. Adjust kernel compatAlexander Bluhm
macros to make art.c compile again in regress.
2019-05-06enable deadlock related testsanton
2019-04-26a bit more evilBob Beck
2019-04-26/usr/include/machine is a good one.Bob Beck
2019-04-23Report slot types recursively for recursive PTE entriesPhilip Guenther
2019-04-23Improve usage messagePhilip Guenther
2019-04-23Simplify presentation of attributes to eliminate duplicationPhilip Guenther
2019-04-23Add a header to explain fieldsPhilip Guenther
2019-04-23Add -u for getting pagetables of pid 1Philip Guenther
User PAGE_SIZE macro
2019-04-23Utility useful for debugging amd64 page tablesPhilip Guenther
2019-04-19Start a realpath(2) regress test, currently not enabled.Bob Beck
This will extend as I add more cases to it. It will come into use once an in-kernel version of realpath(2) becomes a real thing
2019-04-08Check that pf and IPv6 stack handle two ping6 fragments with differentAlexander Bluhm
extension headers correctly and produce a echo reply.
2019-03-31Make ART data structure definitions visible to userland, in order to fixMartin Pieuchot
netstat(1) inspection of routing tables. From Naoki Fukaumi, ok yasuoka@
2019-03-24Make stat(2) and access(2) need UNVEIL_READ instead of UNVEIL_INSPECTBob Beck
UNVEIL_INSPECT is a hack we added to get chrome/glib working. It silently adds permission for stat(2), access(2), and readlink(2) to be used on all path components of any unveil'ed path. robert@ has sucessfully now fixed chrome/glib to not require exessive TOC vs TOU stat(2) and access(2) calls on the paths it uses, so that this no longer needed there. readlink(2) is the sole call that is now permitted by UNVEIL_INSPECT, and this is only needed so that realpath(3) can work. Going forward we will likely make a realpath(2), after which we can completely deprecate UNVEIL_INSPECT. ok deraadt@
2019-03-21Do not run keyboard and mouse device detection during make clean,Alexander Bluhm
cleandir, obj. Eliminates a X11 keyboard problem seen by claudio@ during make build. OK anton@ deraadt@
2019-03-18Disable wscons regress test for now. This test is touching /dev entriesClaudio Jeker
and the result is a rather annoying keyboard repeat issue. Finding this was a bit of a saga. OK by many
2019-03-04add kqueue pty close slave testanton
2019-03-04add support for multiple kqueue pty testsanton
2019-03-04one regress test per line to reduce future churnanton
2019-02-12Skip tests for wscons devices that do not exist on the local machine.Alexander Bluhm
from Moritz Buhl
2019-02-10"non-existant" is one of those words that don't exist, so use "non-existent"Peter Hessler
instead From Pamela Mosiejczuk, many thanks! OK phessler@ deraadt@
2019-02-01make it easier to ignore incompatible device and test case combinationsanton
2019-01-31Instead of skipping the text mmap test, do a non fixed mmap in thisAlexander Bluhm
case. This gives us another test case. Make error messages more verbose. Fix white spaces.
2019-01-25I am retiring my old email address; replace it with my OpenBSD one.Todd C. Miller
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-26 13:31:45 +0000