summaryrefslogtreecommitdiff
path: root/regress/sys
AgeCommit message (Collapse)Author
2021-10-22knf nitsAnton Lindqvist
2021-10-22add pipe select hangup test case; ok mpi@Anton Lindqvist
2021-10-13Sending live tcpdump output over SSH while running tests may dropAlexander Bluhm
bpf logs in the kernel. Better write pcap files onto remote machine's disk and collect and convert after testing with live packets has finished. Move the TCP path MTU tests to the end. Otherwise TCP packets floating around could affect the packet counters of the subsequent tests.
2021-10-06This regress uses vnd devices to create a softraid volume. Raid levelsAlexander Bluhm
RAID 0, RAID 1, RAID 5, CRYPTO, CONCAT, RAID 1 + CRYPTO are created. The volume disk is then mounted and stressed a little bit.
2021-10-05add missing sudoAnton Lindqvist
2021-10-05add missing sudoAnton Lindqvist
2021-10-05build prog onceAnton Lindqvist
2021-10-03Extend filesystem tests to test mfs and tmpfs. Hook up mfs but leave tmpfsClaudio Jeker
out for now since it is not enabled by default. mfs reports the same errors as ffs (no real surprise), tmpfs has a few different errors. OK bluhm@
2021-10-02quote sudoAnton Lindqvist
2021-10-02add missing sudoAnton Lindqvist
2021-09-30Quote SUDO value during propagation, allowing `doas -n' to be correctlyAnton Lindqvist
honored which I use while running regress. The same principle is already applied to the PYTHON variable here.
2021-09-28Turns out that older SPARC CPUs (like the UltraSPARC II) do report an exactMark Kettenis
fault address, while others (like the UltraSPARC T2) report an inexact address. Deal with this by using the EXPADDR_MASK on both the expected and the reported fault address. ok bluhm@
2021-09-28Use -Wl,-z,wxneeded instead of -z wxneeded such that this builds with baseMark Kettenis
gcc. ok patrick@, millert@, jca@
2021-09-28add missing sudoAnton Lindqvist
2021-09-27Return 0 from main() otherwise the exit code is garbage on sparc64.Alexander Bluhm
Collect status of the child process to detect test failures. OK tb@
2021-09-20Use proper sigsuspend() instead of old pause() and use sigprocmask() toClaudio Jeker
block delivery of signals outside of sigsuspend(). With this the test is more reliable. pause() is implemented as two syscalls and so it is possible to catch a signal on the first syscall and than be stuck on the second waiting for something that already happened. OK millert@ deraadt@ bluhm@
2021-08-30Test for the pathetic case that the 2nd unveil inClaudio Jeker
unveil(NULL, NULL); if (fork() == 0) unveil("/", "rwx") fails with EPERM.
2021-08-30Check that fchdir() works. The problem in unveil was fixed some time ago.Claudio Jeker
2021-08-30Make lines not wrap on 80 char terminals. Code gets more readable.Claudio Jeker
2021-07-14Shrink cluster size for FAT32 to fullfil the new minimum clusterAlexander Bluhm
limit of newfs_msdos for our small 64 MB vnd disk image. suggested by jsg@
2021-07-10Pass expected result first when diff(1)ing for result.Martin Pieuchot
2021-07-06This regress checks that 400 singals are evenly distributed over 4Alexander Bluhm
threads. It works well on idle 4 CPU machines, but with load or fewer cores the distribution of the threads' signal handlers is worse. Reduce the limit of signals that each thread must receive from 75 to 50. This should prevent random test fails. from Christian Ludwig
2021-07-06alarm(3) cannot fail, remove error handling.Alexander Bluhm
suggested by millert@
2021-07-06Fix typo in comment.Alexander Bluhm
2019-09-05Test IPv6 multicast sender, receiver, router. Minimal mc6routeAlexander Bluhm
implmentation does not care about ICMPv6, it simply adds static multicast routes. Multicast routing needs at least two test machines. Otherwise only send and receive on localhost is tested. Currently receiving link-local packets via loopback does not work due to wrong scope id. Everything is mostly copied from IPv4 regress.
2021-07-05Write ktrace output files onto NFS mounted file systems.Alexander Bluhm
2021-06-20use __riscv instead of __riscv64Theo de Raadt
2021-06-12Exercise resetting of expired and unexpired timers.Visa Hankala
2021-06-11Adjust futex regress test since futex(2) now returns -1 on error and setsMark Kettenis
errno like any normal syscall. ok mpi@, bluhm@
2021-06-01Introduce the new test which enforces unp_gc() run and closemvs
unreferenced descriptors. ok bluhm@
2021-05-27Introduce the new test which enforces unp_internalize() to follow errormvs
path and FRELE(9) descriptors. ok @bluhm
2021-05-22Test more corner cases of IPv4 fragment reassembly.Alexander Bluhm
2021-05-22On powerpc64 calling a function needs 64 bytes of stack overheadAlexander Bluhm
per recursion. On amd64 it is only 32. Allocate more stack per thread and the stack tests pass everywhere
2021-05-21Fix the calculation of the maximum stack size in the command lineAlexander Bluhm
option -s.
2021-05-21Set the stack size attribute of the pthreads large enough to allocateAlexander Bluhm
the specified amount of stack memory. On 32 bit architectures regress fork-exit failed as the default stack size for pthreads is smaller. With the limit set to the expected size we can test even larger thread stacks.
2021-05-17add riscv supportTheo de Raadt
2021-05-17Link fenv.S test program statically to avoid calling the dynamicAlexander Bluhm
linker.
2021-05-17Skip sosplice loop test if Perl packages are not installed.Alexander Bluhm
2021-05-16add riscv asm for these two security-feature checking regress testsTheo de Raadt
with drahn
2021-05-08Fix a race in the kqueue-read test. If the writer thread managed to fillanton
up the pipe buffer, it will hang in write(2) waiting for the pipe to become writeable again. While doing so, it cannot observe that the kqueue thread as already finished implying that ctx_thread_alive() would return false, meaning no further writes are wanted. Therefore open the pipe in non blocking mode and just ignore EAGAIN errors while writing.
2021-05-07Dynamic socket buffers make tests unreliable, set fixed buffer size.Alexander Bluhm
2021-05-04Create memory pressure in test. Allocate heap and stack in childAlexander Bluhm
processes and threads. Then exit has more work to do.
2021-04-29Extend the fork and exit test with threads. Simultaneously killAlexander Bluhm
30 processes with 30 threads each.
2021-04-28To hunt kernel bugs during exit, terminate processes simultaneously.Alexander Bluhm
Fork 300 children that sleep. Kill them together as process group. Sleeping can optionally be done with individual memory layout by executing sleep(1).
2021-04-21Check that path MTU discovery across routing domains inserts aAlexander Bluhm
dynamic IPv4 route. Add IPv6 test that runs tcpbench over pair interfaces in different routing domains. For that setup, PMTU currently does not work. IPv6 TCP packets get fragmented.
2021-04-14Unify pivot.h; brings powerpc/powerpc64 support to the usr.bin/lastcommMark Kettenis
regress test. ok bluhm@
2021-04-13Unbreak rtable regress test. Define _KERNEL in some crucial spots so thatClaudio Jeker
the include guard in rtsock.h is bypassed and critical defines like rtentry show up. Also remove the old RADIX bits, rtable code is art only for a while.
2021-02-09These regress tests expect coredumps to be written so run them withClaudio Jeker
ulimit -c unlimited. Also simplify the logic a bit as requested by bluhm@. OK bluhm@ deraadt@
2021-02-09These regress test expect coredumps as an effect of the test so runClaudio Jeker
the tests with ulimit -c unlimited to make sure coredumps are written. OK bluhm@ deraadt@
2021-02-05Setup two pair(4) and one lo(4) interface in different routingAlexander Bluhm
domains and with different interface MTU. Test that packet flow and TCP path MTU discovery works.
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-15 15:16:14 +0000