Re: [PATCH] fstests: add fsstress + compaction test

[Luis Chamberlain <mcgrof@kernel.org>]

On Sat, Apr 20, 2024 at 10:02:41PM +0800, Zorro Lang wrote:
> On Wed, Apr 17, 2024 at 05:13:56PM -0700, Luis Chamberlain wrote:
> > +_require_compaction()
> 
> I'm not sure if we should name it as "_require_vm_compaction", does linux
> have other "compaction" or only memory compaction?

I'll color bike shed with "memory compaction" although I am not aware of
other compaction types. However making it clear helps.

> > +++ b/tests/generic/744
> > @@ -0,0 +1,56 @@
> > +#! /bin/bash
> > +# SPDX-License-Identifier: GPL-2.0
> > +# Copyright (c) 2024 Luis Chamberlain.  All Rights Reserved.
> > +#
> > +# FS QA Test 744
> > +#
> > +# fsstress + compaction test
> 
> fsstress + memory compaction ?

Sure.

> Looks like this case is copied from g/476, just add memory_compaction
> test. That makes sense to me from the test side.

It's a generic fsstress + compaction, right.

> you just found. Looks like you're reporting a bug, and provide a test
> case to fstests@ by the way.

This case is hard to reproduce, and so instead of waiting for compaction
to trigger we force it now.

> Anyway, I think there's not objection on
> this test itself, right? And is this test for someone known bug or not?

This reproduces a known kernel bug for which we have a fix now merged,
at the time this test was written it was not even merged on v6.9-rc4.
The fix was merged on v6.9-rc6.

I can now add this to the test:

_fixed_by_git_commit kernel d99e3140a4d3 \                                       
        "mm: turn folio_test_hugetlb into a PageType"

However I also ran into some *other* issues even after that patch was
applied. Now that v6.10-rc1 is out I will retest to clarify the
situation and see if new issues are still lingering with this test.

  Luis