Re: [PATCH v5 8/8] mm: huge_memory: enable debugfs to split huge pages to any order.

From: Mark Brown <broonie@kernel.org>
To: Ryan Roberts <ryan.roberts@arm.com>
Cc: "Zi Yan" <ziy@nvidia.com>,
	"Aishwarya TCV" <aishwarya.tcv@arm.com>,
	"Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>,
	linux-mm@kvack.org,
	"Matthew Wilcox (Oracle)" <willy@infradead.org>,
	"David Hildenbrand" <david@redhat.com>,
	"Yang Shi" <shy828301@gmail.com>, "Yu Zhao" <yuzhao@google.com>,
	"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
	"Michal Koutný" <mkoutny@suse.com>,
	"Roman Gushchin" <roman.gushchin@linux.dev>,
	"Zach O'Keefe" <zokeefe@google.com>,
	"Hugh Dickins" <hughd@google.com>,
	"Luis Chamberlain" <mcgrof@kernel.org>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, linux-kselftest@vger.kernel.org
Subject: Re: [PATCH v5 8/8] mm: huge_memory: enable debugfs to split huge pages to any order.
Date: Fri, 1 Mar 2024 14:24:05 +0000	[thread overview]
Message-ID: <888bedc2-9677-4958-8dd7-71fbbd66b08d@sirena.org.uk> (raw)
In-Reply-To: <f7a3d07d-290b-46d6-884e-fa288901c3c6@arm.com>

[-- Attachment #1: Type: text/plain, Size: 1308 bytes --]

On Fri, Mar 01, 2024 at 01:09:04PM +0000, Ryan Roberts wrote:
> On 01/03/2024 12:52, Zi Yan wrote:

> > I can do that. But is this a new requirement that self tests have to be finish
> > in CI/CD environment? Can you provide a guideline for it? 

> I'm not sure what's written down, but certainly anyone should be able to run the
> selftests with as little knowledge as possible, and they should only fail if
> they detect a real problem. By convention a test should be skipped if the

It does get mentioned in talks and things but TBH I think it's just
generally a good practice thing.

> > Since I always assume
> > selftests are just ran by human who can set up environment. 

> I believe kernelci have been running mm skeftests on x86 for a long time. We
> have started running them against arm64 on our CI for the last couple of months
> and it had found a number of real issues in the kernel in -next, so this is
> helping find and fix things early. So there is definitely benefit to keeping
> these tests clean and robust.

They're also being routinely run on at least some platforms by LKFT
(though from a quick check it seems not with the magic command line
options to set up secretmem and huge pages) and CKI, possibly also some
of the other CI systems I'm less aware of.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]