[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Zwle6aDpkf_0ZCaX@smile.fi.intel.com>
Date: Fri, 11 Oct 2024 20:22:49 +0300
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Christophe Leroy <christophe.leroy@...roup.eu>
Cc: Geert Uytterhoeven <geert@...ux-m68k.org>,
David Gow <davidgow@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Shuah Khan <skhan@...uxfoundation.org>,
Brendan Higgins <brendanhiggins@...gle.com>,
Rae Moar <rmoar@...gle.com>, Kees Cook <kees@...nel.org>,
linux-kselftest@...r.kernel.org, kunit-dev@...glegroups.com,
linux-kernel@...r.kernel.org,
Stephen Rothwell <sfr@...b.auug.org.au>,
Yury Norov <yury.norov@...il.com>,
Rasmus Villemoes <linux@...musvillemoes.dk>,
"Jason A . Donenfeld" <Jason@...c4.com>,
"Naveen N . Rao" <naveen.n.rao@...ux.ibm.com>,
Anil S Keshavamurthy <anil.s.keshavamurthy@...el.com>,
"David S . Miller" <davem@...emloft.net>,
Masami Hiramatsu <mhiramat@...nel.org>,
Mark Brown <broonie@...nel.org>, linux-hardening@...r.kernel.org,
linux-trace-kernel@...r.kernel.org,
Palmer Dabbelt <palmer@...osinc.com>,
Charlie Jenkins <charlie@...osinc.com>,
Simon Horman <horms@...nel.org>, Jakub Kicinski <kuba@...nel.org>,
Arnd Bergmann <arnd@...db.de>, Daniel Latypov <dlatypov@...gle.com>,
Guenter Roeck <linux@...ck-us.net>,
David Howells <dhowells@...hat.com>,
Mickaël Salaün <mic@...ikod.net>,
Marco Elver <elver@...gle.com>, Mark Rutland <mark.rutland@....com>,
"Gustavo A. R. Silva" <gustavoars@...nel.org>,
Vlastimil Babka <vbabka@...e.cz>,
Nathan Chancellor <nathan@...nel.org>,
Fangrui Song <maskray@...gle.com>,
"Steven Rostedt (Google)" <rostedt@...dmis.org>
Subject: Re: [PATCH 3/6] lib: Move KUnit tests into tests/ subdirectory
On Fri, Oct 11, 2024 at 06:29:49PM +0200, Christophe Leroy wrote:
> Le 11/10/2024 à 17:21, Andy Shevchenko a écrit :
> > [Vous ne recevez pas souvent de courriers de andy.shevchenko@...il.com. Découvrez pourquoi ceci est important à https://aka.ms/LearnAboutSenderIdentification ]
> >
> > On Fri, Oct 11, 2024 at 5:20 PM Geert Uytterhoeven <geert@...ux-m68k.org> wrote:
> > > On Fri, Oct 11, 2024 at 3:57 PM Andy Shevchenko <andy@...nel.org> wrote:
> > > > On Fri, Oct 11, 2024 at 03:38:00PM +0200, Geert Uytterhoeven wrote:
> > > > > On Fri, Oct 11, 2024 at 12:44 PM Andy Shevchenko <andy@...nel.org> wrote:
> > > > > > On Fri, Oct 11, 2024 at 03:25:07PM +0800, David Gow wrote:
...
> > > > > > > rename lib/{ => tests}/bitfield_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/checksum_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/cmdline_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/cpumask_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/fortify_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/hashtable_test.c (100%)
> > > > > > > rename lib/{ => tests}/is_signed_type_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/kunit_iov_iter.c (100%)
> > > > > > > rename lib/{ => tests}/list-test.c (100%)
> > > > > > > rename lib/{ => tests}/memcpy_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/overflow_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/siphash_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/slub_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/stackinit_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/string_helpers_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/string_kunit.c (100%)
> > > > > > > rename lib/{ => tests}/test_bits.c (100%)
> > > > > > > rename lib/{ => tests}/test_fprobe.c (100%)
> > > > > > > rename lib/{ => tests}/test_hash.c (100%)
> > > > > > > rename lib/{ => tests}/test_kprobes.c (100%)
> > > > > > > rename lib/{ => tests}/test_linear_ranges.c (100%)
> > > > > > > rename lib/{ => tests}/test_list_sort.c (100%)
> > > > > > > rename lib/{ => tests}/test_sort.c (100%)
> > > > > > > rename lib/{ => tests}/usercopy_kunit.c (100%)
> > > > > >
> > > > > > While I support the idea, I think this adds an additional churn in creating a
> > > > > > duplicate 'test' in the filenames. Why they all can't be cut while removing?
> > > > > > (at least this question is not answered in the commit message)
> > > > >
> > > > > To avoid duplicate *.ko file names?
> > > >
> > > > With what? Sorry, but I don't see how it's a problem. These are test cases.
> > > > Do they use kernel command line parameters? If so, shouldn't KUnit take care
> > > > about it in a more proper way?
> > >
> > > If .e.g. lib/list_sort.o could be modular, its module would be called
> > > "list_sort.ko", conflicting with the "list_sort.ko" test module.
>
> But as it is now in lib/tests/, the module will also be installed in tests/
> subdir, so it shouldn't clash anymore ?
>
> You'd have:
>
> /lib/modules/lib/list_sort.ko and
> /lib/modules/lib/tests/list_sort.ko
>
> Or did I miss something ?
I believe Geert is talking about `modprobe list-sort` in the userspace.
Which one will be loaded?
> > Can't this be solved by automatically adding a prefix in Makefile for
> > kunit tests, for example?
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists