On Thu, Apr 14, 2022 at 2:10 PM Richard Haines <richard_c_haines@xxxxxxxxxxxxxx> wrote: > On Wed, 2022-04-13 at 13:47 +0200, Ondrej Mosnacek wrote: > > Currently the testsuite only verifies that SCTP peeloff works on the > > server side. However, it can just as well be used on the client side > > as > > well, which isn't being tested (and actually is buggy at the time of > > writing). > > > > To correct this, add a simple SCTP peeloff client and a couple tests > > that verify that the client-side peeloff works well with SELinux. > > > > The new tests are enabled on kernels 5.18+, where the fixes were > > introduced. While the fixes are now queued also for some stable > > streams, > > it's much simpler to just check the version against 5.18 and skip the > > test for all older kernels. > > > > Signed-off-by: Ondrej Mosnacek <omosnace@xxxxxxxxxx> > > --- > > > > v4: added a kernel version check > > v3: fixed uninitialized "sinlen" variable in sctp_peeloff_client.c > > v2: rebased and resolved a conflict > > > > tests/sctp/.gitignore | 1 + > > tests/sctp/Makefile | 3 +- > > tests/sctp/sctp_peeloff_client.c | 254 > > +++++++++++++++++++++++++++++++ > > tests/sctp/test | 170 ++++++++++++++++++++- > > 4 files changed, 419 insertions(+), 9 deletions(-) > > create mode 100644 tests/sctp/sctp_peeloff_client.c > > > > Tested on kernel 5.18-rc2 > > Tested-by: Richard Haines <richard_c_haines@xxxxxxxxxxxxxx> Thanks! The patch is now applied. -- Ondrej Mosnacek Software Engineer, Linux Security - SELinux kernel Red Hat, Inc.
