From: Chuck Lever <chuck.lever@xxxxxxxxxx>
I've dropped the alloc_bulk_pages_array() patch, as it seemed
slightly less reliable than allocating a page at a time on my
small-memory test system.
I do not currently have a set up that can test slice/concat
configurations, but the simple/scsi volume type is stable (except
for generic/450).
---
Changes since v2:
- Support more complex volume types
- Dropped the alloc_bulk_pages_array() patch
Changes since RFC:
- series re-ordered to place fixes first
- address review comments as best I can
Chuck Lever (3):
nfs/blocklayout: Fix premature PR key unregistration
nfs/blocklayout: Report only when /no/ device is found
nfs/blocklayout: SCSI layout trace points for reservation key
reg/unreg
fs/nfs/blocklayout/blocklayout.c | 25 ++++---
fs/nfs/blocklayout/blocklayout.h | 9 ++-
fs/nfs/blocklayout/dev.c | 116 +++++++++++++++++++++++--------
fs/nfs/nfs4trace.c | 7 ++
fs/nfs/nfs4trace.h | 88 +++++++++++++++++++++++
5 files changed, 205 insertions(+), 40 deletions(-)
--
2.45.1
