Lorenzo Fontana

@weierophinney The `object` keyword does not lead to an immediate breakage (like Int, Float and String) because is not marked as reserved in the current implementations of the PHP language...

So, what's the point? From: http://php.net/manual/en/reserved.other-reserved-words.php > These words are also prohibited from being used in namespaces

Yes it is, but unfortunately is not available on 7.3 and on some 7.4 - one could handle this btw

Update, just tested this on a machine with an older kernel and works as expected: ``` sudo ./userspace/sysdig/sysdig evt.type=unlinkat -p" %evt.num %evt.outputtime %evt.cpu %proc.name (%thread.tid) %evt.dir %evt.type %evt.info" 5672 14:29:07.650800968...

Debugged this a bit, it looks like that [ppm_copy_from_user](https://github.com/draios/sysdig/blob/55b1cc826b540292e5e9bd930b2e1c2c7a750af5/driver/ppm_events.c#L92) returns `16` while doing `__copy_from_user_inatomic` by calling `raw_copy_from_user` However `raw_copy_from_user` didn't change between the two reported versions, it actually didn't change...

Update: It happens everytime `val_to_ring` is called with `is_user` true.

@gnosek is your suggestion that what @kristopolous described could be integrated with USDT to make the "system debugger" better at understanding what is the actual path to follow to debug...

When it's about executing and not writing to those binaries like the rule that @AkihiroSuda posted we also have the same kind of issue when dealing with symlinks.

This also happens with overlayfs2, while with fuse-overlayfs it seems good.

Fixed 2 out of 5 broken tests by unpinning the maps before the next execution (was a bug in `BPFTcFilterTests`). Still three to go. ``` Running main() from /var/lib/jenkins/workspace/pf_multibranch_pipeline_lf_cmake/contrib/googletest/googletest/src/gtest_main.cc [==========]...