fuzz: distinguish between reading/writing properties in logs

[mrc0mmand]

Also, reorder the for/loop statements, because I have no idea why I put
them like I did in the first place, and check if the remote side is
alive even after just reading a property.

 Interface: org.freedesktop.dfuzzerInterface
  PASS [P] read_only (read)
  PASS [P] write_only (write)
  FAIL [P] crash_on_write (write) - unexpected response while writing to a property
[RE-CONNECTED TO PID: 2479535]
  PASS [P] read_write (read)
  PASS [P] read_write (write)
  PASS [M] df_hello
...

Addresses: https://github.com/matusmarhefka/dfuzzer/pull/86#pullrequestreview-971808731

[mrc0mmand]

Also a note - there's something wrong with the ppc Copr builders, so the ppc Packit jobs currently end up stuck in the queue indefinitely (so there's no point on waiting for them to finish).

[mrc0mmand]

@evverx also, I wonder if it would make sense to tag a release now (without #83), like v2.0, so I could do a new build in Fedora Rawhide, allowing us to drop all the legacy stuff (Makefile, old man page, etc.)?

[mrc0mmand]

@evverx also, I wonder if it would make sense to tag a release now (without #83), like v2.0, so I could do a new build in Fedora Rawhide, allowing us to drop all the legacy stuff (Makefile, old man page, etc.)?

Think about it a bit more, I guess we should resolve (at least) #65 and #67 beforehand.

[evverx]

I guess we should resolve (at least) https://github.com/matusmarhefka/dfuzzer/pull/65 and https://github.com/matusmarhefka/dfuzzer/issues/67 beforehand.

https://github.com/matusmarhefka/dfuzzer/pull/65 was based on the assumption that it's possible to provide a one-size-fits-all config file but at this point I'm not sure if it makes sense. For example, Reload and Rexecute should be blocked when systemd is tested in environments where it's built with ASan/UBsan and reloaded/reexecuted somewhere. In general though I don't think they should be suppressed. Another example would be Freeze, which should be blocked only when the unified hierarchy is used. Ref*, Unref* and Thaw should be removed though.

https://github.com/matusmarhefka/dfuzzer/issues/67 is different in the sense that it should probably be addressed somehow but the only user relying on the output of dfuzzer -l I'm aware of hasn't chimed in yet. I left a link to that issue in os-autoinst/os-autoinst-distri-opensuse#14669 (comment).

All in all, I think those issues can be addressed later.

If it tagging a release could help to finally switch to meson I think it would make sense to do it. Until Makefile is removed meson.build and dfuzzer.h should be kept in sync manually though

meson.build:        version : '1.4',
src/dfuzzer.h:#define DF_VERSION "dfuzzer 1.4\n" \

[evverx]

LGTM. Thanks!