All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Sandeen <sandeen@sandeen.net>
To: fstests <fstests@vger.kernel.org>
Cc: Dmitry Monakhov <dmonakhov@openvz.org>, Jeff Moyer <jmoyer@redhat.com>
Subject: [PATCH V2] loosen input requirements on defrag testing
Date: Wed, 2 Dec 2015 16:09:26 -0600	[thread overview]
Message-ID: <565F6C16.8090405@sandeen.net> (raw)
In-Reply-To: <565F2D7B.90501@sandeen.net>

We don't have perfect control of file allocation for these tests;
in some cases we may fail to adequately fragment a file prior to
defragmentation testing, and today that will fail the test.

Attack this on 2 fronts:

1) Explicitly allow fewer extents on one of the input files in
   generic/018 where the allocator has discretion.
2) _notrun rather than _fail if we don't create enough extents;
   this is a defrag test, not an allocator/fragmentation test,
   so just skip the test if we can't create an acceptable file
   for defrag testing.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---

V2 fix args to _defrag and fix 018.out.  what the ... I swear I tested
this... :/  This time for realz.

diff --git a/common/defrag b/common/defrag
index f923dc0..d2b137e 100644
--- a/common/defrag
+++ b/common/defrag
@@ -53,15 +53,24 @@ _check_extent_count()
 	max=$2
 	ext_cnt=$3
 
-	[ "$min" -gt "$ext_cnt" ] && _fail "Found $ext_cnt extents  min:$max"
-	[ "$max" -ne -1 ] && [ "$ext_cnt" -gt "$max" ] && _fail "Found $ext_cnt max: $max"
+	# Return failure if $3 isn't between $1 and $2; let caller decide
+	# action if it's not, we just return status here.
+	if [ "$min" -gt "$ext_cnt" ]; then
+		echo "Found $ext_cnt extents  min:$max"
+		return 1;
+	fi
+	if [ "$max" -ne -1 ] && [ "$ext_cnt" -gt "$max" ]; then
+		echo "Found $ext_cnt max: $max"
+		return 1;
+	fi
 
 	if [ $max -ne $min ]; then
 	    echo "in_range($min, $max)"
 	else
 	    echo "$ext_cnt"
 	fi
-	return $ext_cnt
+	# success
+	return 0
 }
 
 # Defrag file, check it, and remove it.
@@ -129,6 +138,7 @@ _defrag()
 	echo -n "Before: "
 	ext_before=$(_extent_count $1)
 	_check_extent_count $min_before $max_before $ext_before
+	[ $? -eq 0 ] || _notrun "Could not adequately fragment file"
 
 	[ ! -z $csum ] && CSUM_BEFORE=`md5sum $1`
 	STAT_BEFORE=`stat -c "a: %x m: %y c: %z" $1`
@@ -141,6 +151,7 @@ _defrag()
 	echo -n "After: "
 	ext_after=$(_extent_count $1)
 	_check_extent_count $min_after $max_after $ext_after
+	[ $? -eq 0 ] || _fail "Failed to adequately defragment file"
 
 	[ "$ext_before" -lt "$ext_after" ] && \
 		_fail "Number of extents increased after defragmentation," \
diff --git a/tests/generic/018 b/tests/generic/018
index d97bb88..e8dbab3 100755
--- a/tests/generic/018
+++ b/tests/generic/018
@@ -77,7 +77,9 @@ for i in `seq 9 -1 0`; do
 	$XFS_IO_PROG -fs -c "pwrite -b $bsize $((i * bsize)) $bsize" $fragfile \
 					> /dev/null
 done
-_defrag --before 10 --after 1 $fragfile
+
+# Accept fewer fragments than we might expect; we don't have perfect control.
+_defrag --max_before 10 --min_before 5 --after 1 $fragfile
 
 echo "Write backwards sync leaving holes - defrag should do nothing" | tee -a $seqres.full
 for i in `seq 31 -2 0`; do
diff --git a/tests/generic/018.out b/tests/generic/018.out
index 5f265d1..1746b32 100644
--- a/tests/generic/018.out
+++ b/tests/generic/018.out
@@ -9,7 +9,7 @@ Contiguous file:
 Before: 1
 After: 1
 Write backwards sync, but contiguous - should defrag to 1 extent
-Before: 10
+Before: in_range(5, 10)
 After: 1
 Write backwards sync leaving holes - defrag should do nothing
 Before: 16



  reply	other threads:[~2015-12-02 22:09 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-02 17:42 [PATCH] loosen input requirements on defrag testing Eric Sandeen
2015-12-02 22:09 ` Eric Sandeen [this message]
2015-12-03  8:46   ` [PATCH V2] " Dmitry Monakhov

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=565F6C16.8090405@sandeen.net \
    --to=sandeen@sandeen.net \
    --cc=dmonakhov@openvz.org \
    --cc=fstests@vger.kernel.org \
    --cc=jmoyer@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.