On Sun, May 25, 2014 at 08:20:29PM -0700, Jim Meyering wrote: > On Sun, May 25, 2014 at 7:35 PM, Jim Meyering <[email protected]> wrote: > > On Sun, May 25, 2014 at 3:32 PM, Phillip Susi <[email protected]> wrote: > >> -----BEGIN PGP SIGNED MESSAGE----- > >> Hash: SHA512 > >> > >> On 04/16/2014 10:54 PM, Brian C. Lane wrote: > >>> Newer versions of mkfs.xfs appear to be using more blocks for the > >>> journal, causing t4100-* tests to fail. Make the image larger. > >>> > >>> * tests/t4100-dvh-partition-limits.sh: Make the xfs image larger * > >>> tests/t4100-msdos-partition-limits.sh: Make the xfs image larger > >> > >> These three look good to me. > > > > Likewise. Thanks. I'd just noticed those two failures. > > Actually, I needed the additional changes (see attached) to make those two > tests pass using xfsprogs-3.1.1-7.el6.x86_64. Feel free to include in yours, > or I'll push separately.
Ha, I was just fixing this for rawhide, I have this patch that I'm adding today (previously on my rhel7 branch) -- Brian C. Lane | Anaconda Team | IRC: bcl #anaconda | Port Orchard, WA (PST8PDT)
>From 5800abc0b2162b4145ab2091acc8e9a3ecdb3c88 Mon Sep 17 00:00:00 2001 From: "Brian C. Lane" <[email protected]> Date: Tue, 27 May 2014 07:38:36 -0700 Subject: [PATCH] tests: Use mkfs.xfs to create files (#1101112) On s390 there is a bug with mkfs.xfs and pre-existing files. Work around it by creating the file directly with mkfs.xfs. This also works on other arches. * tests/t1700-probe-fs.sh: Check for xfs and use direct file creation * tests/t4100-dvh-partition-limits.sh: Use mkfs.xfs -dfile * tests/t4100-msdos-partition-limits.sh: Use mkfs.xfs -dfile --- tests/t1700-probe-fs.sh | 9 +++++++-- tests/t4100-dvh-partition-limits.sh | 3 +-- tests/t4100-msdos-partition-limits.sh | 3 +-- 3 files changed, 9 insertions(+), 6 deletions(-) diff --git a/tests/t1700-probe-fs.sh b/tests/t1700-probe-fs.sh index e090a5a..4cd3b8d 100755 --- a/tests/t1700-probe-fs.sh +++ b/tests/t1700-probe-fs.sh @@ -40,8 +40,13 @@ for type in ext2 ext3 ext4 btrfs xfs nilfs2 ntfs vfat hfsplus; do esac # create an $type file system - dd if=/dev/null of=$dev bs=$ss seek=$n_sectors >/dev/null || fail=1 - mkfs.$type $force $dev || { warn_ $ME: mkfs.$type failed; fail=1; continue; } + if [ "$type" == "xfs" ]; then + # Work around a problem with s390 + mkfs.xfs -ssize=$ss -dfile,name=$dev,size=${n_sectors}s || fail=1 + else + dd if=/dev/null of=$dev bs=$ss seek=$n_sectors >/dev/null || fail=1 + mkfs.$type $force $dev || { warn_ $ME: mkfs.$type failed; fail=1; continue; } + fi # probe the $type file system parted -m -s $dev u s print >out 2>&1 || fail=1 diff --git a/tests/t4100-dvh-partition-limits.sh b/tests/t4100-dvh-partition-limits.sh index fb10025..781f792 100755 --- a/tests/t4100-dvh-partition-limits.sh +++ b/tests/t4100-dvh-partition-limits.sh @@ -37,8 +37,7 @@ mp=`pwd`/mount-point n=4096 # create an XFS file system -dd if=/dev/null of=$fs bs=1MB seek=50 || fail=1 -mkfs.xfs -f -q $fs || fail=1 +mkfs.xfs -dfile,name=$fs,size=100m || fail=1 mkdir "$mp" || fail=1 # Unmount upon interrupt, failure, etc., as well as upon normal completion. diff --git a/tests/t4100-msdos-partition-limits.sh b/tests/t4100-msdos-partition-limits.sh index af7907f..c7a611e 100755 --- a/tests/t4100-msdos-partition-limits.sh +++ b/tests/t4100-msdos-partition-limits.sh @@ -37,8 +37,7 @@ mp=`pwd`/mount-point n=4096 # create an XFS file system -dd if=/dev/null of=$fs bs=1MB seek=50 || fail=1 -mkfs.xfs -f -q $fs || fail=1 +mkfs.xfs -dfile,name=$fs,size=100m || fail=1 mkdir "$mp" || fail=1 # Unmount upon interrupt, failure, etc., as well as upon normal completion. -- 1.9.0
