linux-ext4 - Re: [PATCH linux-kselftest/test v4] ext4: add kunit test for decoding extended timestamps

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20191016200304.GA49718@google.com>
Date:   Wed, 16 Oct 2019 13:03:04 -0700
From:   Brendan Higgins <brendanhiggins@...gle.com>
To:     Iurii Zaikin <yzaikin@...gle.com>
Cc:     linux-kselftest@...r.kernel.org, linux-ext4@...r.kernel.org,
        skhan@...uxfoundation.org, tytso@....edu, adilger.kernel@...ger.ca,
        Tim.Bird@...y.com, kunit-dev@...glegroups.com
Subject: Re: [PATCH linux-kselftest/test v4] ext4: add kunit test for
 decoding extended timestamps

On Fri, Oct 11, 2019 at 07:37:57PM -0700, Iurii Zaikin wrote:
> KUnit tests for decoding extended 64 bit timestamps
> that verify the seconds part of [a/c/m]
> timestamps in ext4 inode structs are decoded correctly.
> KUnit tests, which run on boot and output
> the results to the debug log in TAP format (http://testanything.org/).
> are only useful for kernel devs running KUnit test harness. Not for
> inclusion into a production build.
> Test data is derive from the table under
nit:                ^
Should be:     derived from ...

> Documentation/filesystems/ext4/inodes.rst Inode Timestamps.
> 
> Signed-off-by: Iurii Zaikin <yzaikin@...gle.com>

Reviewed-by: Brendan Higgins <brendanhiggins@...gle.com>
Tested-by: Brendan Higgins <brendanhiggins@...gle.com>

> ---
>  fs/ext4/Kconfig      |  14 +++
>  fs/ext4/Makefile     |   1 +
>  fs/ext4/inode-test.c | 239 +++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 254 insertions(+)
>  create mode 100644 fs/ext4/inode-test.c
>
[...]
> diff --git a/fs/ext4/inode-test.c b/fs/ext4/inode-test.c
> new file mode 100644
> index 000000000000..3b3a453ff382
> --- /dev/null
> +++ b/fs/ext4/inode-test.c
> @@ -0,0 +1,239 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * KUnit test of ext4 inode that verify the seconds part of [a/c/m]
> + * timestamps in ext4 inode structs are decoded correctly.
> + */
> +
> +#include <kunit/test.h>
> +#include <linux/kernel.h>
> +#include <linux/time64.h>
> +
> +#include "ext4.h"
> +
> +/* binary: 00000000 00000000 00000000 00000000 */
> +#define LOWER_MSB_0 0L
> +/* binary: 01111111 11111111 11111111 11111111 */
> +#define UPPER_MSB_0 0x7fffffffL
> +/* binary: 10000000 00000000 00000000 00000000 */
> +#define LOWER_MSB_1 (-0x80000000L)
> +/* binary: 11111111 11111111 11111111 11111111 */
> +#define UPPER_MSB_1 (-1L)
> +/* binary: 00111111 11111111 11111111 11111111 */
> +#define MAX_NANOSECONDS ((1L << 30) - 1)
> +
> +#define CASE_NAME_FORMAT "%s: msb:%x lower_bound:%x extra_bits: %x"
> +
> +struct timestamp_expectation {
> +	const char *test_case_name;
> +	struct timespec64 expected;
> +	u32 extra_bits;
> +	bool msb_set;
> +	bool lower_bound;
> +};
> +
> +static time64_t get_32bit_time(const struct timestamp_expectation * const test)
> +{
> +	if (test->msb_set) {
> +		if (test->lower_bound)
> +			return LOWER_MSB_1;
> +
> +		return UPPER_MSB_1;
> +	}
> +
> +	if (test->lower_bound)
> +		return LOWER_MSB_0;
> +	return UPPER_MSB_0;
> +}
> +
> +
> +/*
> + * These tests are derived from the table under
> + * Documentation/filesystems/ext4/inodes.rst Inode Timestamps
> + */
> +static void inode_test_xtimestamp_decoding(struct kunit *test)
> +{
> +	const struct timestamp_expectation test_data[] = {
> +		{
> +			.test_case_name =
> +		"1901-12-13 Lower bound of 32bit < 0 timestamp, no extra bits.",

nit: Maybe drop the period at the end (here and elsewhere)? Otherwise if
the test fails you have a period right next to a colon and it looks a
bit off.

> +			.msb_set = true,
> +			.lower_bound = true,
> +			.extra_bits = 0,
> +			.expected = {.tv_sec = -0x80000000LL, .tv_nsec = 0L},
> +		},

Feel free to ignore my nits if you don't need to send another version.
Also note that Ted has given a Reviewed-by on an earlier revision.

Thanks!