From: Cyril Hrubis <chrubis@suse.cz>
To: Wei Gao <wegao@suse.com>
Cc: s.mesoraca16@gmail.com, ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v2] open15: allow restricted O_CREAT of FIFOs and regular files
Date: Mon, 26 Feb 2024 14:37:19 +0100 [thread overview]
Message-ID: <ZdyUD_rhoMTKMHFD@yuki> (raw)
In-Reply-To: <20231227130555.29035-1-wegao@suse.com>
Hi!
> +/*\
> + * [Description]
> + *
> + * Verify disallows open of FIFOs or regular files not owned by the user in world
> + * writable sticky directories
> + *
> + * commit 30aba6656f61ed44cba445a3c0d38b296fa9e8f5
> + * Author: Salvatore Mesoraca <s.mesoraca16@gmail.com>
> + * Date: Thu Aug 23 17:00:35 2018 -0700
> + * namei: allow restricted O_CREAT of FIFOs and regular files
> + */
> +
> +#include <pwd.h>
> +#include <stdlib.h>
> +#include "tst_test.h"
> +#include "tst_safe_file_at.h"
> +
> +#define FILENAME "setuid04_testfile"
> +#define DIR "ltp_tmp_check1"
> +#define TEST_FILE "test_file_1"
> +#define TEST_FIFO "test_fifo_1"
> +#define LTP_USR_UID1 1000
> +#define LTP_USR_UID2 1001
> +#define CONCAT(dir, filename) dir "/" filename
> +#define PROTECTED_REGULAR "/proc/sys/fs/protected_regular"
> +#define PROTECTED_FIFOS "/proc/sys/fs/protected_fifos"
> +
> +static int dir_fd;
> +
> +static void run(void)
> +{
> + int pid;
> +
> + SAFE_FILE_PRINTF(PROTECTED_REGULAR, "%d", 0);
> + SAFE_FILE_PRINTF(PROTECTED_FIFOS, "%d", 0);
> +
> + pid = SAFE_FORK();
No need to store the pid if we are not using it. We can do instead just:
if (!SAFE_FORK()) {
> + if (pid == 0) {
> + SAFE_SETUID(LTP_USR_UID1);
> +
> + int fd = TST_EXP_FD(openat(dir_fd, TEST_FILE, O_CREAT | O_RDWR, 0777));
> +
> + SAFE_CLOSE(fd);
> + fd = SAFE_MKFIFO(CONCAT(DIR, TEST_FIFO), 0777);
> + SAFE_CLOSE(fd);
This part has to be done in the test setup() otherwise the test will
fail with EEXIST with -i 2.
> + exit(0);
> + }
> +
> + tst_reap_children();
> +
> + pid = SAFE_FORK();
> +
> + if (pid == 0) {
> + SAFE_SETUID(LTP_USR_UID2);
> +
> + int fd = TST_EXP_FD(openat(dir_fd, TEST_FILE, O_CREAT | O_RDWR, 0777));
> +
> + tst_res(TPASS, "check protect_regural == 0 pass");
The TST_EXP_FD() should print TPASS message, there is no point in adding
another.
> + SAFE_CLOSE(fd);
This should be closed only if the fd is valid.
> + fd = SAFE_OPEN(CONCAT(DIR, TEST_FIFO), O_RDWR | O_CREAT);
> + tst_res(TPASS, "check protect_fifos == 0 pass");
> + SAFE_CLOSE(fd);
Again this should be TST_EXP_FD().
> + exit(0);
> + }
> +
> + tst_reap_children();
> +
> + SAFE_FILE_PRINTF(PROTECTED_REGULAR, "%d", 1);
> + SAFE_FILE_PRINTF(PROTECTED_FIFOS, "%d", 1);
> +
> + pid = SAFE_FORK();
> +
> + if (pid == 0) {
> + SAFE_SETUID(LTP_USR_UID2);
> +
> + TST_EXP_FAIL(openat(dir_fd, TEST_FILE, O_RDWR | O_CREAT, 0777), EACCES);
> +
> + TST_EXP_FAIL(open(CONCAT(DIR, TEST_FIFO), O_RDWR | O_CREAT, 0777), EACCES);
> +
> + exit(0);
> + }
> +
> + tst_reap_children();
> +
> + SAFE_FILE_PRINTF(PROTECTED_REGULAR, "%d", 2);
> + SAFE_FILE_PRINTF(PROTECTED_FIFOS, "%d", 2);
> + SAFE_CHMOD(DIR, 0020 | S_ISVTX);
I suppose that this will break the test with -i 2 as well, you need to
chmod the directory back at the end of the test.
> + pid = SAFE_FORK();
> +
> + if (pid == 0) {
> + SAFE_SETUID(LTP_USR_UID2);
> +
> + TST_EXP_FAIL(openat(dir_fd, TEST_FILE, O_RDWR | O_CREAT, 0777), EACCES);
> +
> + TST_EXP_FAIL(open(CONCAT(DIR, TEST_FIFO), O_RDWR | O_CREAT, 0777), EACCES);
> +
> + exit(0);
> + }
> +
> + tst_reap_children();
> +}
> +
> +static void setup(void)
> +{
> + umask(0);
> + SAFE_MKDIR(DIR, 0777 | S_ISVTX);
> +
> + dir_fd = SAFE_OPEN(DIR, O_DIRECTORY);
> +}
> +
> +static void cleanup(void)
> +{
> + SAFE_CLOSE(dir_fd);
> +}
> +
> +static struct tst_test test = {
> + .setup = setup,
> + .cleanup = cleanup,
> + .needs_root = 1,
> + .test_all = run,
> + .needs_tmpdir = 1,
> + .forks_child = 1,
> + .save_restore = (const struct tst_path_val[]) {
> + {PROTECTED_REGULAR, NULL, TST_SR_TCONF},
> + {PROTECTED_FIFOS, NULL, TST_SR_TCONF},
> + {}
> + },
> + .needs_checkpoints = 1,
This should have been removed.
> +};
> --
> 2.35.3
>
>
> --
> Mailing list info: https://lists.linux.it/listinfo/ltp
--
Cyril Hrubis
chrubis@suse.cz
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2024-02-26 13:38 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-09 11:20 [LTP] [PATCH v1] open15: allow restricted O_CREAT of FIFOs and regular files Wei Gao via ltp
2023-12-08 15:34 ` Cyril Hrubis
2023-12-27 13:05 ` [LTP] [PATCH v2] " Wei Gao via ltp
2024-02-26 13:37 ` Cyril Hrubis [this message]
2024-06-03 12:55 ` [LTP] [PATCH v3] " Wei Gao via ltp
2025-02-21 10:01 ` Andrea Cervesato via ltp
2025-03-19 14:23 ` [LTP] [PATCH v4] open16: " Wei Gao via ltp
2025-07-11 12:04 ` Cyril Hrubis
2025-07-23 15:46 ` [LTP] [PATCH v5] " Wei Gao via ltp
2026-02-18 13:21 ` Andrea Cervesato 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=ZdyUD_rhoMTKMHFD@yuki \
--to=chrubis@suse.cz \
--cc=ltp@lists.linux.it \
--cc=s.mesoraca16@gmail.com \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox