Re: [PATCH] selftests/ftrace: skip ftrace test if FTRACE was not enabled

[Masami Hiramatsu <mhiramat@kernel.org>]

On Tue, 2 Jul 2019 21:07:30 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> On Wed, 3 Jul 2019 09:11:47 +0900
> Masami Hiramatsu <mhiramat@kernel.org> wrote:
> 
> > Hi Po-Hsu Lin,
> > 
> > On Tue, 2 Jul 2019 13:22:26 -0600
> > shuah <shuah@kernel.org> wrote:
> > 
> > > Hi Po-Hsu Lin,
> > > 
> > > On 7/2/19 12:23 AM, Po-Hsu Lin wrote:  
> > > > The ftrace test will need to have CONFIG_FTRACE enabled to make the
> > > > ftrace directory available.
> > > > 
> > > > Add an additional check to skip this test if the CONFIG_FTRACE was not
> > > > enabled.  
> > 
> > Sorry, NAK for config check.
> 
> Agreed, as  my test boxes do not hold the source code of the kernel
> they are running.
> 
> > 
> > > > 
> > > > This will be helpful to avoid a false-positive test result when testing
> > > > it directly with the following commad against a kernel that does not
> > > > have CONFIG_FTRACE enabled:  
> > 
> > Would you know tools/testing/selftests/ftrace/config (and other config files
> > in each tests) ?
> > 
> > Since each selftest depends specific configurations, those configs are
> > written in config file, and tester must enable it using 
> > "scripts/kconfig/merge_config.sh".
> > 
> > We can not check the kernel config in some cases, e.g. distro kernel,
> > cross-build kernel, remote build kernel etc. Also, the .config file
> > can be a config file for another kernel build.
> > 
> > So please take care of your kernel configuration. If you find any test
> > failed even if you enable configs in config file under that test, please
> > report it, since that is a bug.
> > 
> > 
> > Thank you,
> > 
> > > >      make -C tools/testing/selftests TARGETS=ftrace run_tests
> > > > 
> > > > The test result on an Ubuntu KVM kernel will be changed from:
> > > >      selftests: ftrace: ftracetest
> > > >      ========================================
> > > >      Error: No ftrace directory found
> > > >      not ok 1..1 selftests: ftrace: ftracetest [FAIL]
> > > > To:  
> > > 
> > > Thanks for the patch.
> > > 
> > > Check patch fails with the above To:
> > > 
> > > WARNING: Use a single space after To:
> > > #107:
> > > To:
> > > 
> > > ERROR: Unrecognized email address: ''
> > > #107:
> > > To:
> > > 
> > > total: 1 errors, 1 warnings, 23 lines checked
> > > 
> > > 
> > > Please fix and send v2.
> > >   
> > > >      selftests: ftrace: ftracetest
> > > >      ========================================
> > > >      CONFIG_FTRACE was not enabled, test skipped.
> > > >      not ok 1..1 selftests: ftrace: ftracetest [SKIP]
> > > > 
> > > > Signed-off-by: Po-Hsu Lin <po-hsu.lin@canonical.com>
> > > > ---
> > > >   tools/testing/selftests/ftrace/ftracetest | 11 ++++++++++-
> > > >   1 file changed, 10 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/tools/testing/selftests/ftrace/ftracetest b/tools/testing/selftests/ftrace/ftracetest
> > > > index 6d5e9e8..6c8322e 100755
> > > > --- a/tools/testing/selftests/ftrace/ftracetest
> > > > +++ b/tools/testing/selftests/ftrace/ftracetest
> > > > @@ -7,6 +7,9 @@
> > > >   #  Written by Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
> > > >   #  
> > > 
> > > Hmm. You havem't cc'ed Masami on this. Adding Masami.
> > > 
> > > I would think Masami should be on the Signed-off-by as well,
> > > since he is the author.
> > >   
> > > >   
> > > > +# Kselftest framework requirement - SKIP code is 4.
> > > > +ksft_skip=4
> > > > +
> > > >   usage() { # errno [message]
> > > >   [ ! -z "$2" ] && echo $2
> > > >   echo "Usage: ftracetest [options] [testcase(s)] [testcase-directory(s)]"
> > > > @@ -139,7 +142,13 @@ parse_opts $*
> > > >   
> > > >   # Verify parameters
> > > >   if [ -z "$TRACING_DIR" -o ! -d "$TRACING_DIR" ]; then
> 
> I'm thinking if we didn't find the TRACING_DIR (-z "$TRACING_DIR"
> returns true), then we exit with the skip. I don't believe we should be
> testing ftrace if tracefs isn't even loaded. Or something like:

Yeah, it should be skipped.

> 
> 	err_ret=1
> 	err_skip=4
> 
> 	errexit() {
> 	  echo "Error: $1" 1>&2
> 	  exit $err_ret
> 	}
> 
> 	[..]
> 
> 	if [ -z "$TRACING_DIR" ]; then
> 	  save_err=$err_ret
> 	  err_ret=$err_skip
> 	  mount -t tracefs nodev /sys/kernel/tracing ||
> 	    errexit "kernel does not have tracefs"
> 	  err_ret=$save_err
> 	  TRACING_DIR="/sys/kernel/tracing"
> 	fi
> 	if [ ! -d "$TRACING_DIR" ]; then
> 	  errexit "tracefs is not a directory?"
> 	fi
> 
> Would something like that work?

For older kernel, I think we'd better try to mount debugfs first.

Thank you,

-- 
Masami Hiramatsu <mhiramat@kernel.org>