Linux Test Project
 help / color / mirror / Atom feed
From: Jan Stancek <jstancek@redhat.com>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH] swapon: commit swapfile caches to disk
Date: Tue, 12 Jun 2018 08:56:12 -0400 (EDT)	[thread overview]
Message-ID: <1468267167.25962192.1528808172867.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <1528806239-14297-1-git-send-email-liwang@redhat.com>


----- Original Message -----
> Testcase include swapon(8) fails on mainline kernel-v4.17+ as:
>   swapon01    1  TFAIL  :  swapon01.c:47: Failed to turn on swapfile:
>   TEST_ERRNO=EINVAL(22): Invalid argument
>   swapon02    4  TFAIL  :  swapon02.c:91: swapon(2) failed to produce
>   expected error: 16, errno: EBUSY and got 22.
>   swapon03    1  TFAIL  :  swapon03.c:243: Failed to setup swaps
>   swapoff01   1  TBROK  :  swapoff01.c:64: Failed to turn on the swap file,
>   skipping test iteration
> 
> And dmesg shows that:
>   [  128.698981] swapon: file is not committed
> 
> The error located in IOMAP_F_DIRTY(linux/fs/iomap.c) checking, which means
> this swapfile have uncommitted metadata in caches. After adding sync() to
> the end of all mkswap opertaion, these errors were gone.
> 
> linux/fs/iomap.c:
> 	static loff_t iomap_swapfile_activate_actor(
> 		struct inode *inode,
> 		loff_t pos,
> 		loff_t count,
> 		void *data,
> 		struct iomap *iomap)
> 	{
> 		struct iomap_swapfile_info *isi = data;
> 		int error;
> 		...
> 
> 		/* No uncommitted metadata or shared blocks. */
> 		if (iomap->flags & IOMAP_F_DIRTY) {
> 			pr_err("swapon: file is not committed\n");
> 			return -EINVAL;
> 		}
> 		...
> 	}
> 									}
> Signed-off-by: Li Wang <liwang@redhat.com>
> ---
> 
> Notes:
>     The new changes about swapfile activation function were merged in
>     kernel-4.17
>     recently, so we didn't hit this failures before.

Comment/code in mkswap, that has been present for years:
	/*
	 * A subsequent swapon() will fail if the signature
	 * is not actually on disk. (This is a kernel bug.)
	 * The fsync() in close_fd() will take care of writing.
	 */

I'm assuming your mkswap is recent enough. Any idea why fsync()
in mkswap is no longer enough?

Regards,
Jan

> 
>  testcases/kernel/syscalls/swapoff/swapoff01.c | 2 ++
>  testcases/kernel/syscalls/swapoff/swapoff02.c | 2 ++
>  testcases/kernel/syscalls/swapon/libswapon.c  | 2 ++
>  3 files changed, 6 insertions(+)
> 
> diff --git a/testcases/kernel/syscalls/swapoff/swapoff01.c
> b/testcases/kernel/syscalls/swapoff/swapoff01.c
> index a63e661..b587b0d 100644
> --- a/testcases/kernel/syscalls/swapoff/swapoff01.c
> +++ b/testcases/kernel/syscalls/swapoff/swapoff01.c
> @@ -105,6 +105,8 @@ static void setup(void)
>  
>  	if (system("mkswap swapfile01 > tmpfile 2>&1") != 0)
>  		tst_brkm(TBROK, cleanup, "Failed to make swapfile");
> +
> +	sync();
>  }
>  
>  static void cleanup(void)
> diff --git a/testcases/kernel/syscalls/swapoff/swapoff02.c
> b/testcases/kernel/syscalls/swapoff/swapoff02.c
> index b5c6312..d8c5889 100644
> --- a/testcases/kernel/syscalls/swapoff/swapoff02.c
> +++ b/testcases/kernel/syscalls/swapoff/swapoff02.c
> @@ -154,6 +154,8 @@ static void setup(void)
>  
>  	if (tst_fill_file("./swapfile01", 0x00, 1024, 1))
>  		tst_brkm(TBROK, cleanup, "Failed to create swapfile");
> +
> +	sync();
>  }
>  
>  static void cleanup(void)
> diff --git a/testcases/kernel/syscalls/swapon/libswapon.c
> b/testcases/kernel/syscalls/swapon/libswapon.c
> index cf6a988..49fa5ed 100644
> --- a/testcases/kernel/syscalls/swapon/libswapon.c
> +++ b/testcases/kernel/syscalls/swapon/libswapon.c
> @@ -46,4 +46,6 @@ void make_swapfile(void (cleanup)(void), const char
> *swapfile)
>  	argv[2] = NULL;
>  
>  	tst_run_cmd(cleanup, argv, "/dev/null", "/dev/null", 0);
> +
> +	sync();
>  }
> --
> 1.9.3
> 
> 
> --
> Mailing list info: https://lists.linux.it/listinfo/ltp
> 

  parent reply	other threads:[~2018-06-12 12:56 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-12 12:23 [LTP] [PATCH] swapon: commit swapfile caches to disk Li Wang
2018-06-12 12:48 ` Cyril Hrubis
2018-06-12 13:05   ` Li Wang
2018-06-12 13:17     ` Li Wang
2018-06-12 12:56 ` Jan Stancek [this message]
2018-06-12 13:12   ` Li Wang
2018-06-14  6:37     ` Li Wang

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=1468267167.25962192.1528808172867.JavaMail.zimbra@redhat.com \
    --to=jstancek@redhat.com \
    --cc=ltp@lists.linux.it \
    /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