xfs/019: fix golden output for files created in setgid dir
authorDarrick J. Wong <djwong@kernel.org>
Tue, 19 Apr 2022 17:32:00 +0000 (10:32 -0700)
committerZorro Lang <zlang@kernel.org>
Tue, 26 Apr 2022 18:54:00 +0000 (02:54 +0800)
A recent change to xfs/019 exposed a long-standing bug in mkfs where
it would always set the gid of a new child created in a setgid directory
to match the gid parent directory instead of what's in the protofile.

Ignoring the user's directions is not the correct behavior, so update
this test to reflect that.  Also don't erase the $seqres.full file,
because that makes forensic analysis pointlessly difficult.

Cc: Catherine Hoang <catherine.hoang@oracle.com>
Fixes: 7834a740 ("xfs/019: extend protofile test")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Catherine Hoang <catherine.hoang@oracle.com>
Reviewed-by: Zorro Lang <zlang@redhat.com>
Signed-off-by: Zorro Lang <zlang@kernel.org>
tests/xfs/019
tests/xfs/019.out

index 535b7af1a769e9fedb2d17c1867a05434f1ef537..790a6821a32b8f93e637ecb33bb143b3fdabb500 100755 (executable)
@@ -10,6 +10,7 @@
 _begin_fstest mkfs auto quick
 
 seqfull="$seqres.full"
+rm -f $seqfull
 # Import common functions.
 . ./common/filter
 
@@ -97,7 +98,6 @@ _verify_fs()
        echo "*** create FS version $1"
        VERSION="-n version=$1"
 
-       rm -f $seqfull
        _scratch_unmount >/dev/null 2>&1
 
        _full "mkfs"
@@ -131,6 +131,5 @@ _verify_fs()
 _verify_fs 2
 
 echo "*** done"
-rm $seqfull
 status=0
 exit
index 8584f593e96c615218ff7cfad7a400920258b6c6..9db157f9b413b9b6896b1dfbf53ac0f784eacf03 100644 (file)
@@ -61,7 +61,7 @@ Device: <DEVICE> Inode: <INODE> Links: 2
 
  File: "./directory_setgid/file_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx_5"
  Size: 5 Filetype: Regular File
- Mode: (0755/-rwxr-xr-x) Uid: (3) Gid: (2)
+ Mode: (0755/-rwxr-xr-x) Uid: (3) Gid: (1)
 Device: <DEVICE> Inode: <INODE> Links: 1 
 
  File: "./pipe"