From: Petr Vorel <pvorel@suse.cz>
To: Yang Xu <xuyang2018.jy@fujitsu.com>
Cc: ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v4 6/7] syscalls/swapon03: Simply this case
Date: Fri, 23 Feb 2024 12:48:17 +0100 [thread overview]
Message-ID: <20240223114817.GH1423688@pevik> (raw)
In-Reply-To: <20240220074218.13487-6-xuyang2018.jy@fujitsu.com>
Hi Yang Xu,
> By moving swapfile create stage from verify_swaopon and
> test EPERM error more accurate. Also use glibc wrapper by
> using swapon/swapoff instead of call syscall number directly
> because glibc/musl/binoic also support them since long time ago.
+1 thanks for checking.
FYI uClibc-ng only support with UCLIBC_LINUX_SPECIFIC config enabled, but it's
by default enabled. And I guess nobody runs uClibc-ng on Linux without this
enabled => really safe to depend on libc wrapper.
...
> + TST_EXP_PASS_SILENT(swapon(filename, 0));
+1
> }
> exit(0);
> } else
> @@ -145,13 +61,40 @@ static int setup_swap(void)
> if (WEXITSTATUS(status))
> tst_brk(TFAIL, "Failed to setup swaps");
nit: s/swaps/swap files/
> - /* Create all needed extra swapfiles for testing */
> - for (j = 0; j < testfiles; j++)
> - make_swapfile(swap_testfiles[j].filename, 10, 0);
> + tst_res(TINFO, "Successfully created %d swapfiles", swapfiles);
nit: s/swapfiles/swap files/
> + make_swapfile(TEST_FILE, 10, 0);
> return 0;
> }
> +/*
> + * Check if the file is at /proc/swaps and remove it giving swapoff
> + */
> +static int check_and_swapoff(const char *filename)
> +{
> + char cmd_buffer[256];
> + int rc = -1;
> +
> + if (snprintf(cmd_buffer, sizeof(cmd_buffer),
> + "grep -q '%s.*file' /proc/swaps", filename) < 0) {
> + tst_res(TWARN, "sprintf() failed to create the command string");
nit: we don't have SAFE_SNPRINTF() and don't even check snprintf() / sprintf()
return value. Shouldn't we add SAFE_SNPRINTF() which TBROK?
This can be handled later, thus I would here either use plain snprintf() or
tst_brk(TBROK).
if you add return -1 here, the following block does not have to be in else
(=> fewer indentation => text can be longer fewer string splits).
> + } else {
> + rc = 0;
> + if (system(cmd_buffer) == 0) {
> + /* now we need to swapoff the file */
> + if (swapoff(filename) != 0) {
Why not single if?
if (system(cmd_buffer) == 0) && swapoff(filename) != 0) {
> + tst_res(TWARN, "Failed to turn off swap "
> + "file. system reboot after "
> + "execution of LTP test suite "
> + "is recommended");
Then this string would not need to be split several times (bad for searching
with 'git grep'). Maybe shorten just to
"Failed to swapoff %", filename"
=> more important than suggest to reboot (which is obvious) is to point out
problematic swap file, which was kept on.
> + rc = -1;
> + }
> + }
> + }
> +
> + return rc;
> +}
> +
> /*
> * Turn off all swapfiles previously turned on
> */
> @@ -169,49 +112,17 @@ static int clean_swap(void)
Return code of clean_swap() is not used. How about to make it void?
> }
> }
> - for (j = 0; j < testfiles; j++) {
> - if (check_and_swapoff(swap_testfiles[j].filename) != 0) {
> - tst_res(TWARN, "Failed to turn off swap file %s.",
> - swap_testfiles[j].filename);
> - return -1;
> - }
> + if (check_and_swapoff("testfile") != 0) {
> + tst_res(TWARN, "Failed to turn off swap file testfile");
We have the warning in the function, why also here?
> + return -1;
> }
> return 0;
> }
...
> +static void verify_swapon(void)
> {
> + TST_EXP_FAIL(swapon(TEST_FILE, 0), EPERM, "swapon(%s, 0)", TEST_FILE);
+1
Kind regards,
Petr
> }
> static void setup(void)
> @@ -220,6 +131,11 @@ static void setup(void)
> tst_brk(TCONF, "swap not supported by kernel");
> is_swap_supported(TEST_FILE);
> +
> + if (setup_swap() < 0) {
> + clean_swap();
> + tst_brk(TBROK, "Setup failed, quitting the test");
> + }
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2024-02-23 11:48 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-20 7:42 [LTP] [PATCH v4 1/7] libltpswap: Add tst_max_swapfiles api Yang Xu via ltp
2024-02-20 7:42 ` [LTP] [PATCH v4 2/7] libltpswap: alter tst_count_swaps api Yang Xu via ltp
2024-02-23 9:37 ` Petr Vorel
2024-02-20 7:42 ` [LTP] [PATCH v4 3/7] syscalls/swapon03: use tst_max_swapfiles() and GET_USED_SWAPFILES() api Yang Xu via ltp
2024-02-21 9:37 ` Petr Vorel
2024-02-22 1:27 ` Yang Xu (Fujitsu) via ltp
2024-02-23 9:48 ` Petr Vorel
2024-02-20 7:42 ` [LTP] [PATCH v4 4/7] swaponoff.h: Remove useless header Yang Xu via ltp
2024-02-23 10:15 ` Petr Vorel
2024-02-20 7:42 ` [LTP] [PATCH v4 5/7] swapon/Makefile: Remove useless section for MAX_SWAPFILES Yang Xu via ltp
2024-02-23 10:48 ` Petr Vorel
2024-02-20 7:42 ` [LTP] [PATCH v4 6/7] syscalls/swapon03: Simply this case Yang Xu via ltp
2024-02-23 11:48 ` Petr Vorel [this message]
2024-02-20 7:42 ` [LTP] [PATCH v4 7/7] Add fallback for RHEL9 Yang Xu via ltp
2024-02-23 11:54 ` Petr Vorel
2024-02-23 9:04 ` [LTP] [PATCH v4 1/7] libltpswap: Add tst_max_swapfiles api Petr Vorel
2024-02-23 11:56 ` Petr Vorel
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=20240223114817.GH1423688@pevik \
--to=pvorel@suse.cz \
--cc=ltp@lists.linux.it \
--cc=xuyang2018.jy@fujitsu.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