Paul Moore [off-list ref] writes:

On Tue, Aug 24, 2021 at 1:11 PM Christophe Leroy
[off-list ref] wrote:

quoted

Le 24/08/2021 à 16:47, Paul Moore a écrit :

quoted

On Tue, Aug 24, 2021 at 9:36 AM Christophe Leroy
[off-list ref] wrote:

quoted

Commit e65e1fc2d24b ("[PATCH] syscall class hookup for all normal
targets") added generic support for AUDIT but that didn't include
support for bi-arch like powerpc.

Commit 4b58841149dc ("audit: Add generic compat syscall support")
added generic support for bi-arch.

Convert powerpc to that bi-arch generic audit support.

Cc: Paul Moore <paul@paul-moore.com>
Cc: Eric Paris <eparis@redhat.com>
Signed-off-by: Christophe Leroy <redacted>
---
Resending v2 with Audit people in Cc

v2:
- Missing 'git add' for arch/powerpc/include/asm/unistd32.h
- Finalised commit description
---
  arch/powerpc/Kconfig                |  5 +-
  arch/powerpc/include/asm/unistd32.h |  7 +++
  arch/powerpc/kernel/Makefile        |  3 --
  arch/powerpc/kernel/audit.c         | 84 -----------------------------
  arch/powerpc/kernel/compat_audit.c  | 44 ---------------
  5 files changed, 8 insertions(+), 135 deletions(-)
  create mode 100644 arch/powerpc/include/asm/unistd32.h
  delete mode 100644 arch/powerpc/kernel/audit.c
  delete mode 100644 arch/powerpc/kernel/compat_audit.c

Can you explain, in detail please, the testing you have done to verify
this patch?

I built ppc64_defconfig and checked that the generated code is functionnaly equivalent.

ppc32_classify_syscall() is exactly the same as audit_classify_compat_syscall() except that the
later takes the syscall as second argument (ie in r4) whereas the former takes it as first argument
(ie in r3).

audit_classify_arch() and powerpc audit_classify_syscall() are slightly different between the
powerpc version and the generic version because the powerpc version checks whether it is
AUDIT_ARCH_PPC or not (ie value 20), while the generic one checks whether it has bit
__AUDIT_ARCH_64BIT set or not (__AUDIT_ARCH_64BIT is the sign bit of a word), but taking into
account that the abi is either AUDIT_ARCH_PPC, AUDIT_ARCH_PPC64 or AUDIT_ARCH_PPC64LE, the result is
the same.

If you are asking I guess you saw something wrong ?

I was asking because I didn't see any mention of testing, and when you
are enabling something significant like this it is nice to see that it
has been verified to work :)

While binary dumps and comparisons are nice, it is always good to see
verification from a test suite.  I don't have access to the necessary
hardware to test this, but could you verify that the audit-testsuite
passes on your test system with your patches applied?

 * https://github.com/linux-audit/audit-testsuite

I tested on ppc64le. Both before and after the patch I get the result
below.

So I guess the patch is OK, but maybe we have some existing issue.

I had a bit of a look at the test code, but my perl is limited. I think
it was running the command below, and it returned "<no matches>", but
not really sure what that means.

  $ sudo ausearch -i -m SYSCALL -p 216440 -ui 0 -gi 0 -ul 0 -su unconfined _u:unconfined_r:unconfined_t:s0-s0:c0.c1023 -ts recent
  <no matches>

cheers



Running as   user    root
        with context unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
        on   system  Fedora

backlog_wait_time_actual_reset/test .. ok
exec_execve/test ..................... ok
exec_name/test ....................... ok
file_create/test ..................... ok
file_delete/test ..................... ok
file_rename/test ..................... ok
filter_exclude/test .................. 1/21
# Test 20 got: "256" (filter_exclude/test at line 167)
#    Expected: "0"
#  filter_exclude/test line 167 is: ok( $result, 0 );
# Test 21 got: "0" (filter_exclude/test at line 179)
#    Expected: "1"
#  filter_exclude/test line 179 is: ok( $found_msg, 1 );
filter_exclude/test .................. Failed 2/21 subtests
filter_saddr_fam/test ................ ok
filter_sessionid/test ................ ok
login_tty/test ....................... ok
lost_reset/test ...................... ok
netfilter_pkt/test ................... ok
syscalls_file/test ................... ok
syscall_module/test .................. ok
time_change/test ..................... ok
user_msg/test ........................ ok
fanotify/test ........................ ok
bpf/test ............................. ok

Test Summary Report
-------------------
filter_exclude/test                (Wstat: 0 Tests: 21 Failed: 2)
  Failed tests:  20-21
Files=18, Tests=202, 45 wallclock secs ( 0.18 usr  0.03 sys + 20.15 cusr  0.92 csys = 21.28 CPU)
Result: FAIL
Failed 1/18 test programs. 2/202 subtests failed.