[PATCH v2 2/2] ioctl console test v2

[Dmitry V. Levin]

On Mon, Mar 09, 2020 at 06:41:53PM +0530, Bran S wrote:
> > > -// #include "tests.h"
> >
> > All tests should include this header file unless they have a strong reason
> > not to include it.
> 
> I removed it because it wasn't being used. Should I add it ?

tests.h also includes config.h, so yes, it should be included.

> > This change probably fixes those two warnings reported by the compiler,
> > but the test still fails.  I'm not sure that's the purpose of submitting
> > a test that's expected to fail.
> 
> PASS: ioctl_block.gen.test
> PASS: ioctl_console.gen.test
> PASS: ioctl_dm.gen.test
> 
> FAIL: strace-DDD
> ================
> 
> 1351 1127 26203
> strace-DDD.test: failed test: ../../strace -D -enone -esignal=none
> ../tracer_ppid_pgid_sid: unexpected output
> FAIL strace-DDD.test (exit status: 1)
> 
> The test report from my system. Only one test i.e. strace-DDD.test
> fails. This test has been known to fail.

Yes, it's a bug in the test that assumes something about the operating
system which is not always the case.  This test has to be fixed.

> ioctl_console.gen.test passes.

It passes because it tests very little.  Try changing ul_status from 0
to a different integer value and the test will fail.

> > By the way, have you seen a "work in progress" branch esyr/kd?
> 
> I see it. Do I need to make changes in the test w.r.t. esyr/kd ?

No, you don't have to rebase your change on top of esyr/kd branch for a
very simple reason: its tests/ioctl_kd.c already tests both KDGETLED and
KDSETLED in a greater extent than your patch.


-- 
ldv