[PATCH 0/6] tracing: open/delete fixes
From: Oleg Nesterov
Date: Tue Jul 23 2013 - 17:04:23 EST
Hello.
Untested again, only for review. Sorry, I know this is annoying.
Will try to test tomorrow.
The only essential change is that "clear d_subdirs->i_private"
comes as a last patch, this needs the additional comment in the
changelogs of 1-4 but avoids the temporary complications.
I am sending this series right now because:
- I promised to do this today ;)
- Whatever I do, correctness-wise this needs your review
anyway.
- Even if correct, the very intent needs an ack from Masami
who seems to disagree with this approach.
This series shifts the validation from f_op->open() (which
currently doesn't work) to ->read/write.
While I personally think this makes more sense (simpler
and does not block rmdir/event_remove), I understand that
this is subjective, so I won't argue too much if you prefer
to add the additional ref-counts instead.
>From 6/6:
Note: this doesn't not fix other problems, event_remove() can
destroy the active ftrace_event_call, we need more changes but
those changes are completely orthogonal.
Oleg.
kernel/trace/trace.h | 4 +-
kernel/trace/trace_events.c | 176 ++++++++++++++++++------------------
kernel/trace/trace_events_filter.c | 21 ++---
3 files changed, 99 insertions(+), 102 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/