On Mon, 17 Jul 2023 05:24:43 +0000
Ajay Kaher [off-list ref] wrote:

quoted

quoted

As this patch as is will break when running on older kernels, I was
wondering if we should do this instead?  

+1 since the latest kselftest is used also for checking the older
stable kernels, the test case has to check the environment and
change the parameter (or make it unsupported for new feature)
So below looks good to me.  

+1, many ftrace tests are unsupported in my setup and may few require
changes, not sure. Does any auto job takes care of this?

You mean like some kernel CI? Not that I know of.

Shuah, do you run these selftests on older kernels to make sure they don't
fail just because the test is unsupported?

-- Steve