Message-ID: <CABVgOS=ySmWfWqxra4N7z9dqWsk86GY=LTDZ2juPXAQGp462EA@mail.gmail.com>
Date:   Wed, 11 Jan 2023 15:18:08 +0800
From:   David Gow <davidgow@...gle.com>
To:     Kees Cook <keescook@...omium.org>
Cc:     Guenter Roeck <linux@...ck-us.net>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Nathan Chancellor <nathan@...nel.org>,
        linux-hardening@...r.kernel.org,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Josh Poimboeuf <jpoimboe@...nel.org>,
        Geert Uytterhoeven <geert+renesas@...der.be>,
        Miguel Ojeda <ojeda@...nel.org>,
        Isabella Basso <isabbasso@...eup.net>,
        Vlastimil Babka <vbabka@...e.cz>,
        Dan Williams <dan.j.williams@...el.com>,
        Rasmus Villemoes <linux@...musvillemoes.dk>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kunit: memcpy: Split slow memcpy tests into MEMCPY_SLOW_KUNIT_TEST

On Sat, 7 Jan 2023 at 12:02, Kees Cook <keescook@...omium.org> wrote:
>
> Since the long memcpy tests may stall a system for tens of seconds
> in virtualized architecture environments, split those tests off under
> CONFIG_MEMCPY_SLOW_KUNIT_TEST so they can be separately disabled.
>
> Reported-by: Guenter Roeck <linux@...ck-us.net>
> Link: https://lore.kernel.org/lkml/20221226195206.GA2626419@roeck-us.net
> Cc: Andrew Morton <akpm@...ux-foundation.org>
> Cc: Nathan Chancellor <nathan@...nel.org>
> Cc: linux-hardening@...r.kernel.org
> Signed-off-by: Kees Cook <keescook@...omium.org>
> ---
> Guenter, does this give you the needed flexibility to turn on the memcpy
> kunit tests again in your slower environments?
> ---

Sorry for the delay: I've just got back from holidays.

This looks good to me (modulo the tristate/bool issue mentioned).

Having a specific way to tag tests as "slow" is something we've
considered having in KUnit for a while, but (at least initially),
there weren't any tests slow enough for it to be worthwhile. It looks
like that's changing (alongside this, there are a few DRM tests which
are quite slow, as well as some of the KCSAN tests. Even the
time64_to_tm_test_date_range test, while very quick under UML on a
modern system, is noticeably slow under qemu and times out on
something like an old 486...). Daniel's comments elsewhere in this
thread have some good ideas.

That being said, it'll probably take long enough to work out and
implement a good way of giving tests "properties" and requesting only
fast tests run that just putting tests which are slow enough to cause
problems behind a kconfig entry seems a pretty solid intermediate
solution.

So this is:
Reviewed-by: David Gow <davidgow@...gle.com>

(assuming the tristate goes to bool), and I'll look into getting a way
of marking tests as "slow" and enabling/disabling them at runtime.

Cheers,
-- David

>  lib/Kconfig.debug  |  9 +++++++++
>  lib/memcpy_kunit.c | 17 +++++++++++++----
>  2 files changed, 22 insertions(+), 4 deletions(-)
>
> diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
> index c2c78d0e761c..b5e94807f41c 100644
> --- a/lib/Kconfig.debug
> +++ b/lib/Kconfig.debug
> @@ -2621,6 +2621,15 @@ config MEMCPY_KUNIT_TEST
>
>           If unsure, say N.
>
> +config MEMCPY_SLOW_KUNIT_TEST
> +       tristate "Include exhaustive memcpy tests" if !KUNIT_ALL_TESTS

As mentioned, if this is just going to be a toggle, it can be a "bool".


> +       depends on MEMCPY_KUNIT_TEST
> +       default KUNIT_ALL_TESTS
> +       help
> +         Some memcpy tests are quite exhaustive in checking for overlaps
> +         and bit ranges. These can be very slow, so they are split out
> +         as a separate config.
> +
>  config IS_SIGNED_TYPE_KUNIT_TEST
>         tristate "Test is_signed_type() macro" if !KUNIT_ALL_TESTS
>         depends on KUNIT
> diff --git a/lib/memcpy_kunit.c b/lib/memcpy_kunit.c
> index 89128551448d..cc1f36335a9b 100644
> --- a/lib/memcpy_kunit.c
> +++ b/lib/memcpy_kunit.c
> @@ -307,8 +307,12 @@ static void set_random_nonzero(struct kunit *test, u8 *byte)
>         }
>  }
>
> -static void init_large(struct kunit *test)
> +static int init_large(struct kunit *test)
>  {
> +       if (!IS_ENABLED(CONFIG_MEMCPY_SLOW_KUNIT_TEST)) {
> +               kunit_skip(test, "Slow test skipped. Enable with CONFIG_MEMCPY_SLOW_KUNIT_TEST=y");
> +               return -EBUSY;
> +       }
>
>         /* Get many bit patterns. */
>         get_random_bytes(large_src, ARRAY_SIZE(large_src));
> @@ -319,6 +323,8 @@ static void init_large(struct kunit *test)
>
>         /* Explicitly zero the entire destination. */
>         memset(large_dst, 0, ARRAY_SIZE(large_dst));
> +
> +       return 0;
>  }
>
>  /*
> @@ -327,7 +333,9 @@ static void init_large(struct kunit *test)
>   */
>  static void copy_large_test(struct kunit *test, bool use_memmove)
>  {
> -       init_large(test);
> +
> +       if (init_large(test))
> +               return;
>
>         /* Copy a growing number of non-overlapping bytes ... */
>         for (int bytes = 1; bytes <= ARRAY_SIZE(large_src); bytes++) {
> @@ -472,7 +480,8 @@ static void memmove_overlap_test(struct kunit *test)
>         static const int bytes_start = 1;
>         static const int bytes_end = ARRAY_SIZE(large_src) + 1;
>
> -       init_large(test);
> +       if (init_large(test))
> +               return;
>
>         /* Copy a growing number of overlapping bytes ... */
>         for (int bytes = bytes_start; bytes < bytes_end;
> @@ -549,8 +558,8 @@ static void strtomem_test(struct kunit *test)
>  static struct kunit_case memcpy_test_cases[] = {
>         KUNIT_CASE(memset_test),
>         KUNIT_CASE(memcpy_test),
> -       KUNIT_CASE(memcpy_large_test),
>         KUNIT_CASE(memmove_test),
> +       KUNIT_CASE(memcpy_large_test),
>         KUNIT_CASE(memmove_large_test),
>         KUNIT_CASE(memmove_overlap_test),
>         KUNIT_CASE(strtomem_test),
> --
> 2.34.1
>

Download attachment "smime.p7s" of type "application/pkcs7-signature" (4003 bytes)

Powered by blists - more mailing lists