Skip to contentSkip to conversation

ext4: test for DAX + journaling corruption

PATCH New Ross Zwisler 1 participant 2 messages Jan 4, 2018 libnvdimm: non-volatile memory device

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 ext4: test for DAX + journaling corruption

Add a regression test for the following kernel commit:

ext4: prevent data corruption with journaling + DAX

The test passes if either we successfully compare the data between the mmap with journaling turned on and the one with journaling turned off, or if we fail the chattr command to turn on or off journaling. The latter is how we prevent this issue in the kernel.

---

Changes since v1:

- Reordered .gitignore entry. (Eryu)

- Added comments about how "chattr +j" turns off DAX and about why we need the 'nodelalloc' mount option. (Eryu)

- Added a _require_command for chattr. (Eryu)

- Added $here for src/t_ext4_dax_journal_corruption command. (Eryu)

The previous version of this series is here:

https://lists.01.org/pipermail/linux-nvdimm/2017-September/012463.html

The related kernel patches were merged in v4.15-rc1.

---
 .gitignore                          |   1 +
 src/Makefile                        |   3 +-
 src/t_ext4_dax_journal_corruption.c | 102 ++++++++++++++++++++++++++++++++++++
 tests/ext4/030                      |  74 ++++++++++++++++++++++++++
 tests/ext4/030.out                  |   2 +
 tests/ext4/group                    |   1 +
 6 files changed, 182 insertions(+), 1 deletion(-)
 create mode 100644 src/t_ext4_dax_journal_corruption.c
 create mode 100755 tests/ext4/030
 create mode 100644 tests/ext4/030.out
.gitignore +1 -0
index f27c30af..840e4fe4 100644
@@ -115,6 +115,7 @@
 /src/t_dir_offset2
 /src/t_dir_type
 /src/t_encrypted_d_revalidate
+/src/t_ext4_dax_journal_corruption
 /src/t_futimens
 /src/t_getcwd
 /src/t_holes
src/Makefile +2 -1
index b1012172..86c5440c 100644
@@ -13,7 +13,8 @@ TARGETS = dirstress fill fill2 getpagesize holes lstat64 \
 	multi_open_unlink dmiperf unwritten_sync genhashnames t_holes \
 	t_mmap_writev t_truncate_cmtime dirhash_collide t_rename_overwrite \
 	holetest t_truncate_self t_mmap_dio af_unix t_mmap_stale_pmd \
-	t_mmap_cow_race t_mmap_fallocate fsync-err t_mmap_write_ro
+	t_mmap_cow_race t_mmap_fallocate fsync-err t_mmap_write_ro \
+	t_ext4_dax_journal_corruption
 
 LINUX_TARGETS = xfsctl bstat t_mtab getdevicesize preallo_rw_pattern_reader \
 	preallo_rw_pattern_writer ftrunc trunc fs_perms testx looptest \
src/t_ext4_dax_journal_corruption.c new file +102 -0

102 lines not on this page · raw message

tests/ext4/030 new file +74 -0

74 lines not on this page · raw message

tests/ext4/030.out new file +2 -0
new file mode 100644
index 00000000..06a1c8fe
@@ -0,0 +1,2 @@
+QA output created by 030
+Silence is golden
tests/ext4/group +1 -0
index 257bb646..ef768dff 100644
@@ -32,6 +32,7 @@
 027 auto quick fsmap
 028 auto quick fsmap
 029 auto quick fsmap
+030 auto quick
 271 auto rw quick
 301 aio auto ioctl rw stress defrag
 302 aio auto ioctl rw stress defrag
Signature · 6 lines
2.14.3

_______________________________________________
Linux-nvdimm mailing list
[email protected]
https://lists.01.org/mailman/listinfo/linux-nvdimm
#1 lore raw

PATCH 2/2 ext4: test for inline data + DAX corruption

Add a regression test for the following kernel commit:

ext4: prevent data corruption with inline data + DAX

The test passes either if we don't encounter corruption, or if mounting with DAX + inline data fails. The latter is the way that we prevent this issue in the kernel.

---

Changes since v1:

- Changed ordering of .gitignore entries. (Eryu)

- Added '_require_scratch_ext4_feature "inline_data"' test. (Eryu)

---
 .gitignore                         |  1 +
 src/Makefile                       |  2 +-
 src/t_ext4_dax_inline_corruption.c | 70 +++++++++++++++++++++++++++++++
 tests/ext4/031                     | 86 ++++++++++++++++++++++++++++++++++++++
 tests/ext4/031.out                 |  2 +
 tests/ext4/group                   |  1 +
 6 files changed, 161 insertions(+), 1 deletion(-)
 create mode 100644 src/t_ext4_dax_inline_corruption.c
 create mode 100755 tests/ext4/031
 create mode 100644 tests/ext4/031.out
.gitignore +1 -0
index 840e4fe4..09dd01cb 100644
@@ -116,6 +116,7 @@
 /src/t_dir_type
 /src/t_encrypted_d_revalidate
 /src/t_ext4_dax_journal_corruption
+/src/t_ext4_dax_inline_corruption
 /src/t_futimens
 /src/t_getcwd
 /src/t_holes
src/Makefile +1 -1
index 86c5440c..b96b8cf2 100644
@@ -14,7 +14,7 @@ TARGETS = dirstress fill fill2 getpagesize holes lstat64 \
 	t_mmap_writev t_truncate_cmtime dirhash_collide t_rename_overwrite \
 	holetest t_truncate_self t_mmap_dio af_unix t_mmap_stale_pmd \
 	t_mmap_cow_race t_mmap_fallocate fsync-err t_mmap_write_ro \
-	t_ext4_dax_journal_corruption
+	t_ext4_dax_journal_corruption t_ext4_dax_inline_corruption
 
 LINUX_TARGETS = xfsctl bstat t_mtab getdevicesize preallo_rw_pattern_reader \
 	preallo_rw_pattern_writer ftrunc trunc fs_perms testx looptest \
src/t_ext4_dax_inline_corruption.c new file +70 -0

70 lines not on this page · raw message

tests/ext4/031 new file +86 -0

86 lines not on this page · raw message

tests/ext4/031.out new file +2 -0
new file mode 100644
index 00000000..b3d0bb03
@@ -0,0 +1,2 @@
+QA output created by 031
+Silence is golden
tests/ext4/group +1 -0
index ef768dff..396f963d 100644
@@ -33,6 +33,7 @@
 028 auto quick fsmap
 029 auto quick fsmap
 030 auto quick
+031 auto quick
 271 auto rw quick
 301 aio auto ioctl rw stress defrag
 302 aio auto ioctl rw stress defrag
Signature · 6 lines
2.14.3

_______________________________________________
Linux-nvdimm mailing list
[email protected]
https://lists.01.org/mailman/listinfo/linux-nvdimm