Re: [PATCH v10 6/6] selftests/mm: add hwpoison-panic destructive test

[Andrew Morton]

On Tue, 28 Jul 2026 02:29:39 -0700 Breno Leitao <leitao@xxxxxxxxxx> wrote:

> On Tue, Jul 28, 2026 at 04:57:13PM +0800, Miaohe Lin wrote:
> > On 2026/7/28 12:24, Andrew Morton wrote:
> > > On Thu, 9 Jul 2026 11:34:05 +0800 Miaohe Lin <linmiaohe@xxxxxxxxxx> wrote:
> > > 
> > >>> The script is intentionally NOT wired into run_vmtests.sh: every
> > >>> successful run panics the kernel, which is incompatible with the
> > >>> sequential "run each category in the same VM" model that
> > >>> run_vmtests.sh assumes.  It is also not registered as a TEST_PROGS /
> > >>> ksft_* wrapper so a default kselftest run does not opt itself into
> > >>> a panic.  The script is meant to be executed manually inside a
> > >>> disposable VM (e.g. virtme-ng), one variant per VM boot, and
> > >>> requires RUN_DESTRUCTIVE=1 in the environment as a safety net.
> > >>>
> > >>> Signed-off-by: Breno Leitao <leitao@xxxxxxxxxx>
> > >>
> > >> With Mike's comment addressed:
> > >>
> > >> Acked-by: Miaohe Lin <linmiaohe@xxxxxxxxxx>
> > > 
> > > As I understand it, Breno would prefer not to rewrite the selftest in C
> > > at this time.  IOW, we stick with the bash version for now and add
> > > "convert this to C" to the todo list, OK?
> > 
> > Sure. Should I make this change or should we wait until Breno has time to do it?
> 
> Apologies for the delay. This is on my TODO list, and I'll prioritize it.
> 
> Just to confirm: the feature is already in mm-next without the selftest, so I
> only need to submit the C version of the test, correct?

The bash script is still there, no known issues.

We have the thing and tests are useful.  I'd say retain it until the
rewrite-in-C happens?