[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20200603215457.146447-1-ebiggers@kernel.org>
Date: Wed, 3 Jun 2020 14:54:57 -0700
From: Eric Biggers <ebiggers@...nel.org>
To: Theodore Ts'o <tytso@....edu>
Cc: linux-ext4@...r.kernel.org, linux-fscrypt@...r.kernel.org
Subject: [xfstests-bld PATCH] test-appliance: exclude ext4/023 and ext4/028 from encrypt config
From: Eric Biggers <ebiggers@...gle.com>
In Linux 5.8, the test_dummy_encryption mount option will use v2
encryption policies rather than v1 as it previously did. This increases
the size of the encryption xattr slightly, causing two ext4 tests to
start failing due to xattr spillover. Exclude these tests.
See kernel commit ed318a6cc0b6 ("fscrypt: support
test_dummy_encryption=v2") for more details.
Signed-off-by: Eric Biggers <ebiggers@...gle.com>
---
.../files/root/fs/ext4/cfg/encrypt.exclude | 2 ++
.../files/root/fs/ext4/cfg/encrypt_1k.exclude | 13 +++++++------
2 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt.exclude b/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt.exclude
index 304201e..47c26e7 100644
--- a/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt.exclude
+++ b/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt.exclude
@@ -5,7 +5,9 @@ ext4/004 # dump/restore doesn't handle quotas
# xattr size. This causes problems with encryption
# which requires its own xattrs which take space.
ext4/022
+ext4/023
ext4/026
+ext4/028
# file systems with encryption enabled can't be mounted with ext3
ext4/044
diff --git a/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt_1k.exclude b/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt_1k.exclude
index e31c371..cd60151 100644
--- a/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt_1k.exclude
+++ b/kvm-xfstests/test-appliance/files/root/fs/ext4/cfg/encrypt_1k.exclude
@@ -1,8 +1,15 @@
# These tests are also excluded in encrypt.exclude.
# See there for the reasons.
+#
+# Due to the 1k block size, ext4/023 and ext4/028 also fail for a second reason:
+# they use _scratch_populate_cached() which tries to create a 1023-byte symlink,
+# which fails with encrypt_1k because encrypted symlinks are limited to
+# blocksize-3 bytes, not blocksize-1 as is the case for no encryption.
ext4/004
ext4/022
+ext4/023
ext4/026
+ext4/028
generic/082
generic/219
generic/230
@@ -19,9 +26,3 @@ generic/204
ext4/034
generic/273
generic/454
-
-# These tests use _scratch_populate_cached() which tries to create a 1023-byte
-# symlink, which fails with encrypt_1k because encrypted symlinks are limited to
-# blocksize-3 bytes, not blocksize-1 as is the case for no encryption.
-ext4/023
-ext4/028
--
2.26.2
Powered by blists - more mailing lists