On Mon, 2020-01-27 at 13:51 -0500, Steven Rostedt wrote:
On Mon, 27 Jan 2020 12:36:21 -0600 Tom Zanussi zanussi@kernel.org wrote:
Hi Steve,
For 4.19, this patch applies if these patches are applied first:
commit 656fe2ba85e81d00e4447bf77b8da2be3c47acb2 Author: Tom Zanussi tom.zanussi@linux.intel.com Date: Tue Dec 18 14:33:24 2018 -0600
tracing: Use hist trigger's var_ref array to destroy var_refs
commit de40f033d4e84e843d6a12266e3869015ea9097c Author: Tom Zanussi tom.zanussi@linux.intel.com Date: Tue Dec 18 14:33:23 2018 -0600
tracing: Remove open-coding of hist trigger var_ref management
After applying all 3 to 4.19, I built and ran the ftrace/trigger selftests and didn't see any problems.
Thanks, I was cherry-picking parts of those changes, but it wasn't fixing the issue itself. Let me just backport both of them in full and see if that works.
It does fix the issue for me and passes the selftests. Remember that 4.19 doesn't have the .trace() hist action - you need to use the event name e.g. .first()
Tom
-- Steve