[4/8] xfs: fix tests to handle removal of no-alloc create nonfeature

Commit Message

Darrick J. Wong Dec. 13, 2017, 6:03 a.m. UTC

From: Darrick J. Wong <[e-mail]>

We're removing from XFS the ability to perform no-allocation file
creation.  This was added years ago because someone at SGI demanded that
we still be able to create (empty?) files with zero free blocks
remaining so long as there were free inodes and space in existing
directory blocks.  This came at an unacceptable risk of ENOSPC'ing
midway through a transaction and shutting down the fs, so we're removing
it for the create case.

However, some tests fail as a result, so fix them to be more flexible
about not failing when a dir/file creation fails due to ENOSPC.

Signed-off-by: Darrick J. Wong <[e-mail]>
---
 tests/xfs/013 |    6 ++++--
 tests/xfs/014 |    3 +++
 tests/xfs/104 |    2 +-
 tests/xfs/109 |    2 +-
 4 files changed, 9 insertions(+), 4 deletions(-)



--
To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Dave Chinner Dec. 13, 2017, 10:12 p.m. UTC | #1

On Tue, Dec 12, 2017 at 10:03:42PM -0800, Darrick J. Wong wrote:
> From: Darrick J. Wong <[e-mail]>
> 
> We're removing from XFS the ability to perform no-allocation file
> creation.  This was added years ago because someone at SGI demanded that
> we still be able to create (empty?) files with zero free blocks

Just to correct the record, it wasn't SGI that demanded this - it
was implemented by SGI a long, long time ago (~1998) in response to
customer demands.

> remaining so long as there were free inodes and space in existing
> directory blocks.  This came at an unacceptable risk of ENOSPC'ing
> midway through a transaction and shutting down the fs, so we're removing
> it for the create case.

Well, 20 years later we consider it an unacceptible risk. :P

> However, some tests fail as a result, so fix them to be more flexible
> about not failing when a dir/file creation fails due to ENOSPC.
> 
> Signed-off-by: Darrick J. Wong <[e-mail]>

Looks fine.

Reviewed-by: Dave Chinner <[e-mail]>

Patch

diff --git a/tests/xfs/013 b/tests/xfs/013
index 80298ca..394b9bc 100755
--- a/tests/xfs/013
+++ b/tests/xfs/013
@@ -145,8 +145,10 @@  $FSSTRESS_PROG -d $SCRATCH_MNT/fsstress -n 9999999 -p 2 -S t \
 for i in $(seq 1 $LOOPS)
 do
 	# hard link the content of the current directory to the next
-	cp -Rl $SCRATCH_MNT/dir$i $SCRATCH_MNT/dir$((i+1)) 2>&1 | \
-		filter_enospc
+	while ! test -d $SCRATCH_MNT/dir$((i+1)); do
+		cp -Rl $SCRATCH_MNT/dir$i $SCRATCH_MNT/dir$((i+1)) 2>&1 | \
+			filter_enospc
+	done
 
 	# do a random replacement of files in the new directory
 	_rand_replace $SCRATCH_MNT/dir$((i+1)) $COUNT
diff --git a/tests/xfs/014 b/tests/xfs/014
index 875ab40..08cd001 100755
--- a/tests/xfs/014
+++ b/tests/xfs/014
@@ -112,6 +112,9 @@  _test_enospc()
 	# consume 1/2 of the current preallocation across the set of 4 writers
 	write_size=$((TOTAL_PREALLOC / 2 / 4))
 	for i in $(seq 0 3); do
+		touch $dir/file.$i
+	done
+	for i in $(seq 0 3); do
 		$XFS_IO_PROG -f -c "pwrite 0 $write_size" $dir/file.$i \
 			>> $seqres.full &
 	done
diff --git a/tests/xfs/104 b/tests/xfs/104
index 785027e..c3b5977 100755
--- a/tests/xfs/104
+++ b/tests/xfs/104
@@ -65,7 +65,7 @@  _stress_scratch()
 	# -w ensures that the only ops are ones which cause write I/O
 	FSSTRESS_ARGS=`_scale_fsstress_args -d $SCRATCH_MNT -w -p $procs \
 	    -n $nops $FSSTRESS_AVOID`
-	$FSSTRESS_PROG $FSSTRESS_ARGS >> $seqres.full &
+	$FSSTRESS_PROG $FSSTRESS_ARGS >> $seqres.full 2>&1 &
 }
 
 # real QA test starts here
diff --git a/tests/xfs/109 b/tests/xfs/109
index e0fdec3..2625f15 100755
--- a/tests/xfs/109
+++ b/tests/xfs/109
@@ -79,7 +79,7 @@  allocate()
 			while [ $j -lt 100 ]; do
 				$XFS_IO_PROG -f -c 'pwrite -b 64k 0 16m' $file \
 					>/dev/null 2>&1
-				rm $file
+				test -e $file && rm $file
 				let j=$j+1
 			done
 		} &