On Sun, Oct 30, 2011 at 10:25:55AM +0800, Eryu Guan wrote:
> In commit c18bf42de9d5d5fa05025754df1ff63f2147bd12 a warning message
> will be printed to stderr if filesystem doesn't support fallocate(2).
> The warning message will fail these tests though fsx returns SUCCESS.
>
> FSTYP -- ext3
> PLATFORM -- Linux/i686 debian-dev 3.1.0+
> MKFS_OPTIONS -- /dev/sda3
> MOUNT_OPTIONS -- -o acl,user_xattr /dev/sda3 /mnt/sda3
>
> 075 12s ... - output mismatch (see 075.out.bad)
> --- 075.out 2011-10-28 21:35:41.872573530 +0800
> +++ 075.out.bad 2011-10-30 10:31:02.607058029 +0800
> @@ -4,6 +4,9 @@
> -----------------------------------------------
> fsx.0 : -d -N numops -S 0
> -----------------------------------------------
> +fsx: main: filesystem does not support fallocate, disabling
> +: Operation not supported
>
> So redirect stderr of fsx to /dev/null too.
>
> Signed-off-by: Eryu Guan <guaneryu@xxxxxxxxx>
> ---
> 075 | 2 +-
> 112 | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/075 b/075
> index 0f0a6ed..87d273f 100755
> --- a/075
> +++ b/075
> @@ -77,7 +77,7 @@ _do_test()
>
> # This cd and use of -P gets full debug on $here (not TEST_DEV)
> cd $out
> - if ! $here/ltp/fsx $_param -P $here $seq.$_n >/dev/null
> + if ! $here/ltp/fsx $_param -P $here $seq.$_n &>/dev/null
> then
> echo " fsx ($_param) failed, $? - compare $seq.$_n.{good,bad,fsxlog}"
> mv $out/$seq.$_n $here/$seq.$_n.full
> diff --git a/112 b/112
> index 557350a..763b17a 100755
> --- a/112
> +++ b/112
> @@ -77,7 +77,7 @@ _do_test()
>
> # This cd and use of -P gets full debug on $here (not TEST_DEV)
> cd $out
> - if ! $here/ltp/fsx $_param -P $here $seq.$_n >/dev/null
> + if ! $here/ltp/fsx $_param -P $here $seq.$_n &>/dev/null
> then
> echo " fsx ($_param) returned $? - see $seq.$_n.full"
> mv $seq.$_n.fsxlog $here/$seq.$_n.full
If this hasn't been committed, the correct fix is to pass the "-q"
option to fsx to tell it not to emit these messages. We still really
want any other output to stderr to cause a test failure....
Cheers,
Dave.
--
Dave Chinner
david@xxxxxxxxxxxxx
|