All of lore.kernel.org
 help / color / mirror / Atom feed
From: Petr Vorel <pvorel@suse.cz>
To: Wei Gao <wegao@suse.com>
Cc: ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v1] tst_filesystems01.c: Add test for .filesystems
Date: Tue, 16 Sep 2025 13:02:48 +0200	[thread overview]
Message-ID: <20250916110248.GB251329@pevik> (raw)
In-Reply-To: <20250529105156.2564268-1-wegao@suse.com>

Hi Wei, Cyril,

Cyril, single note below can wait after the release.
Also, please have look if this fulfils your idea about testing the feature.

> Fixes: https://github.com/linux-test-project/ltp/issues/1243
> Signed-off-by: Wei Gao <wegao@suse.com>
> ---
>  lib/newlib_tests/tst_filesystems01.c | 118 +++++++++++++++++++++++++++
>  1 file changed, 118 insertions(+)
>  create mode 100644 lib/newlib_tests/tst_filesystems01.c

The test is running fine on github, please add it to
lib/newlib_tests/runtest.sh.

+++ lib/newlib_tests/runtest.sh
@@ -24,6 +24,7 @@ tst_checkpoint_wait_timeout
 tst_checkpoint_wake_timeout
 tst_device
 tst_expiration_timer
+tst_filesystems01
 tst_fuzzy_sync0[1-3]
 tst_needs_cmds0[1-36-8]
 tst_res_hexd

...
> +	snprintf(str_size, sizeof(str_size), "%u", size);
> +	snprintf(path, sizeof(path), "tune2fs -l %s 2>&1", tst_device->dev);
> +	tune2fs = SAFE_POPEN(path, "r");
> +	if (tune2fs == NULL)
> +		tst_brk(TBROK, "Can not popen %s", path);
The point of SAFE_* functions/macros is to avoid having to check in the tests :).
Please remove these 2 lines.

> +	while (fgets(line, PATH_MAX, tune2fs)) {
> +		if (*line && strstr(line, "Inode size:") && strstr(line, str_size)) {
Common patter for fgets() is to check for != NULL, then you don't need "*line &&":
	while (fgets(line, PATH_MAX, tune2fs) != NULL) {
		if (strstr(line, "Inode size:") && strstr(line, str_size)) {

> +			tst_res(TPASS, "check inode size pass");
> +			return 0;
> +		}
> +	}
> +
> +	pclose(tune2fs);
> +	tst_res(TFAIL, "check inode size failed");
> +	return -1;
> +}
> +
> +static int check_mnt_data(char *opt)
> +{
> +	FILE *fp;
> +	char line[PATH_MAX];
> +
> +	fp = SAFE_FOPEN("/proc/mounts", "r");
> +	if (fp == NULL)
> +		tst_brk(TBROK, "Can not open /proc/mounts");
Also here.

> +	while (fgets(line, PATH_MAX, fp)) {
> +		if (*line && strstr(line, tst_device->dev) && strstr(line, opt)) {
Also here.
> +			tst_res(TPASS, "check mnt data pass");
> +			return 0;
> +		}
> +	}
> +	SAFE_FCLOSE(fp);
> +	tst_res(TFAIL, "check mnt data failed");
> +	return -1;
> +}
> +
> +static int check_mkfs_size_opt(unsigned int size)
> +{
> +	char path[PATH_MAX];
> +	char line[PATH_MAX];
> +	FILE *dumpe2fs;
> +	char str_size[NAME_MAX];
> +
> +	snprintf(str_size, sizeof(str_size), "%u", size);
> +	snprintf(path, sizeof(path), "dumpe2fs -h %s 2>&1", tst_device->dev);
> +	dumpe2fs = SAFE_POPEN(path, "r");
> +	if (dumpe2fs == NULL)
> +		tst_brk(TBROK, "Can not popen %s", path);
And here.

> +
> +	while (fgets(line, PATH_MAX, dumpe2fs)) {
> +		if (*line && strstr(line, "Block count:") && strstr(line, str_size)) {
And here.

> +			tst_res(TPASS, "check mkfs size opt pass");
> +			return 0;
> +		}
> +	}
> +
> +	pclose(dumpe2fs);
> +	tst_res(TFAIL, "check mkfs size opt failed");
> +	return -1;
> +}
> +
> +static void do_test(void)
> +{
> +	long fs_type;
> +
> +	fs_type = tst_fs_type(MOUNT_POINT);
> +
> +	if (fs_type == TST_EXT234_MAGIC) {
> +		check_inode_size(128);
check_inode_size() and the other 2 test functions return int but you don't use
the return value. Either use it or change return type to void.

> +		check_mkfs_size_opt(10240);
> +	}
> +
> +	if (fs_type == TST_XFS_MAGIC)
> +		check_mnt_data("usrquota");
While this works, it expect that nobody changes .filesystems below.

@Cyril: wouldn't be useful if the test itself had it's struct tst_fs member
from filesystems[i] available? Static variable, either pointer to the member or
instance of i from run_tcases_per_fs()?

Because we always expect filesystems (.type) is unique in .filesystems. But in
the future we might have test which uses several items in .filesystems with a
different data but an unique filesystem (not needed so far).

Kind regards,
Petr

> +}
> +
> +static struct tst_test test = {
> +	.test_all = do_test,
> +	.needs_root = 1,
> +	.mntpoint = MOUNT_POINT,
> +	.mount_device = 1,
> +	.needs_cmds = (const char *[]) {
> +		"tune2fs",
> +		"dumpe2fs",
> +		NULL
> +	},
> +	.filesystems = (struct tst_fs []) {
> +		{
> +			.type = "ext3",
> +			.mkfs_opts = (const char *const []){"-I", "128", NULL},
> +			.mkfs_size_opt = "10240",
> +		},
> +		{
> +			.type = "xfs",
> +			.mnt_data = "usrquota",
> +		},
> +		{}
> +	},
> +
> +};

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

  reply	other threads:[~2025-09-16 11:03 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-29 10:51 [LTP] [PATCH v1] tst_filesystems01.c: Add test for .filesystems Wei Gao via ltp
2025-09-16 11:02 ` Petr Vorel [this message]
2025-09-24  8:42 ` [LTP] [PATCH v2] " Wei Gao via ltp
2025-10-30 20:17   ` Petr Vorel
2025-10-31  6:33     ` Wei Gao via ltp
2025-10-31  9:13       ` Petr Vorel
2025-11-01  1:20   ` [LTP] [PATCH v3 0/2] " Wei Gao via ltp
2025-11-01  1:20     ` [LTP] [PATCH v3 1/2] " Wei Gao via ltp
2025-11-11 12:06       ` Petr Vorel
2025-11-12  6:51         ` Wei Gao via ltp
2025-11-12  9:22           ` Petr Vorel
2025-11-12 10:32             ` Cyril Hrubis
2025-11-12 11:34               ` Wei Gao via ltp
2025-11-12 11:44                 ` Cyril Hrubis
2025-11-12 12:14                   ` Wei Gao via ltp
2025-11-12 12:21               ` Petr Vorel
2025-11-13  1:28                 ` Wei Gao via ltp
2025-11-01  1:20     ` [LTP] [PATCH v3 2/2] tst_format_device.sh: Add e2fsprogs package for CI test Wei Gao via ltp
2025-11-11 12:22       ` Petr Vorel
2025-11-13  1:52     ` [LTP] [PATCH v4 0/3] tst_filesystems01.c: Add test for .filesystems Wei Gao via ltp
2025-11-13  1:52       ` [LTP] [PATCH v4 1/3] " Wei Gao via ltp
2026-01-22  3:37         ` Li Wang via ltp
2026-01-22  8:02           ` Petr Vorel
2026-01-22  8:54             ` Wei Gao via ltp
2025-11-13  1:52       ` [LTP] [PATCH v4 2/3] ci: Add mount operation for busybox Wei Gao via ltp
2025-11-13  1:52       ` [LTP] [PATCH v4 3/3] ci: Add e2fsprogs package for CI test Wei Gao via ltp
2025-12-18 13:49         ` Andrea Cervesato via ltp
2025-12-18 13:51           ` Andrea Cervesato via ltp
2026-01-23  2:36       ` [LTP] [PATCH v5 0/3] tst_filesystems01.c: Add test for .filesystems Wei Gao via ltp
2026-01-23  2:36         ` [LTP] [PATCH v5 1/3] " Wei Gao via ltp
2026-01-28  0:49           ` Li Wang via ltp
2026-01-23  2:36         ` [LTP] [PATCH v5 2/3] ci: Add mount operation for busybox Wei Gao via ltp
2026-01-23  2:36         ` [LTP] [PATCH v5 3/3] ci: Add e2fsprogs package for CI test Wei Gao via ltp

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250916110248.GB251329@pevik \
    --to=pvorel@suse.cz \
    --cc=ltp@lists.linux.it \
    --cc=wegao@suse.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.